0 Copy ns/nsprpub/config/LinuxELF2.0.mk to ns/nsprpub/config/LinuxELF2.1.mk and then add the follow hacked patch (there should be real LINUX2_2 flags in the source later I guess): *** Linux.mk Thu Apr 9 16:18:45 1998 --- Linux.mk.orig Thu Apr 9 16:18:20 1998 *************** ifeq ($(OS_RELEASE),2.0) *** 77,86 **** PLATFORM_FLAGS += -DLINUX2_0 endif - ifeq ($(OS_RELEASE),2.1) - PLATFORM_FLAGS += -DLINUX2_0 - endif - # # Linux ppc and 2.0 have shared libraries. # --- 77,82 ---- Updated by Wan-Teh Chang (wtc@netscape.com) on Wednesday, April 22, 1998 12:56:37 PM PDT 1 Large client side image maps take a long time to be processed. (Compared to say Netscape 2.x, where it was much faster). In addition, the user doesn Updated by Shawn Steele (shawn@terra.org) on Tuesday, April 14, 1998 1:07:56 PM PDT 1 testing please ignroe aasdfasdflasd;fl asd asdfl;asjd fasdf asa;ldfj sadf sda f asdf asd f
asdf 1111
" 1 A suggestion for later model DEC alpha (OSF1) Dec Unix platforms; Dont use GCC/G++/GNU - it leads to code problems since that compiler is not very picky about ANSI C++ standards on callings and arguments. I have made a few ports to this 64-bit platform, here are a few useful things I found /* +========================================================================+ */ /* | Playground, Copyright 1996 Washington University All Rights Reserved. | */ /* |------------------------------------------------------------------------| */ /* | FILE: CompilerDefs.hh */ /* |------------------------------------------------------------------------| */ /* | Who | Date | Implementation Modification */ /* |------------------------------------------------------------------------| */ /* | jyp | Feb 01, 97 | Initial Code Generation */ /* +========================================================================+ */ #ifndef CompilerDefs_h_ #define CompilerDefs_h_ /* */ /* list of weird definitions and other tricks needed to work around */ /* various compilers idiosyncrasies */ /* */ /* The following pre-defined symbols are provided by the compiler */ /* _MSC_VER - Visual C++ version ID 4.2E */ /* __BCPLUSPLUS__ - Borland C++ V5.0 */ /* __SUNPRO_CC - Sun Sparcworks Pro V3.1 */ /* __DECCXX - Digital CXX V5.6 */ /* __DECCXX_VER = 600300 1 This bug is being submitted using the optimized native version of Mozilla 5.0! A list of all altered files... ./mozilla/cmd/xfe/src/MozillaApp.cpp ./mozilla/cmd/xfe/src/HistoryFrame.cpp ./mozilla/cmd/xfe/src/Frame.cpp ./mozilla/cmd/xfe/src/NavCenterFrame.cpp ./mozilla/cmd/xfe/src/EditorFrame.cpp ./mozilla/cmd/xfe/src/HistoryView.cpp * ./mozilla/cmd/xfe/src/Menu.cpp ./mozilla/cmd/xfe/src/EditorToolbar.cpp ./mozilla/cmd/xfe/src/NavCenterView.cpp ./mozilla/cmd/xfe/src/PopupMenu.cpp ./mozilla/cmd/xfe/src/EditorView.cpp * ./mozilla/cmd/xfe/src/URLBar.cpp ./mozilla/cmd/xfe/src/HTMLDialogFrame.cpp ./mozilla/cmd/xfe/cxxlink-driver.sh ./mozilla/nsprpub/pr/include/md/_osf1.cfg ./mozilla/nsprpub/config/OSF1.mk ./mozilla/nsprpub/config/UNIX.mk ./mozilla/config/OSF1.mk I should have submitted a change for each file, all explicit type casts, except for the OSF1.mk and cxxlink-driver.sh In the configs, I removed -taso everywhere (not needed, and it makes things slower) diff mozilla/cmd/xfe/cxxlink-driver.sh $X/mozilla/cmd/xfe/cxxlink-driver.sh 79,82c79 < # HEAD=$LD_DIR/ld -G 8 -g2 -call_shared -nocount $LD_DIR/crt0.o < # $CXX_DIR/_main.o -count -taso < HEAD=$LD_DIR/ld -G 8 -g2 -call_shared -nocount $LD_DIR/crt0.o < $CXX_DIR/_main.o -count --- > HEAD=$LD_DIR/ld -G 8 -g2 -call_shared -nocount $LD_DIR/crt0.o $CXX_DIR/_main.o -count -taso diff mozi 1 Like, does it work? Oh, and what happens if I say
 here? And whats the handling of apostrophes these days? Updated by Terry Weissman (terry@netscape.com) on Monday, May 18, 1998 9:05:49 AM PDT 
1  test -11"
1 "testing ------- Additional Comments From arunr 05/19/98 14:27 ------- testing again please ignore"
1  testing if email notification is being received while creating new bugs Updated by Arun Ramachandran (T1) (arunr@test.com) on Wednesday, June 10, 1998 3:55:31 PM PDT 
1  Bugzilla test Updated by Arun Ramachandran Test -2 (arunr-test2@netscape.com) on Wednesday, June 10, 1998 2:10:28 PM PDT 
1  testing again gagina gasfsdfasdf a sdfasdf Updated by Arun Ramachandran (arunr@netscape.com) on Thursday, June 11, 1998 12:00:07 PM PDT 
1  I installed a windows NT network here, with one server and 4 clients. After installing netscape I noticed that netscape stores the specific user information under the LOCAL_MACHINE key in the registry. Since only the CURRENT_USER key gets saved per user this leads to some problems. For example: user x saves the profile under U:profiles u: being the mounted user-directory from the server. Now he logs out and another user starts netscape. Since he doesnt have u:profiles on his u: drive netscpae asks him to create a profile. He does it..and everything works. But this way you get a pretty large number of profiles on one machine and the worst part is, when the user goes to another machine he has to do it again. A workaround is to tell the users to create a profile with a special name default and store it under u: etscape. So netscape looks for it on the network drive and finds it. But that way it is impossible for a user to have more than one profile for himself."
0  Currently Mozilla still uses ~/.netscape/ directory for its config files, although ~/.mozilla/ may seem more appropriate. Authors of different brands of Mozilla may want to have their programs to use separate directories and users may also want to use different versions of Mozilla with different config files. I wrote a patch (it is derived from the patch by Dan McGuirk ) that adds a new run-time option -configdir  that allows to specify a config subdirectory different from .netscape and also allows to change the compiled in default by changing one line of the ocode instead of hunting for all these hardcoded strings through all the sources. I am trying to keep the patch up-to-date with the current CVS sources and you should always be able to find them (divided into files according to modules) at ftp://ftp.cs.cornell.edu/pub/nogin/mozilla/"
0 "When I asked to give me a list of all Bugzilla bugs in Bugzilla, buglist.cgi gave me a list titled Mozilla bug list. I think you should drop Mozilla from the title"
1 "I liked the idea of sending the bug entry updates as a diff. And I think it would be especially nice to be able to personalize the diff options. For example, I would prefer to receive unified diffs instead of context ones."
1 "It would be nice if you could add some extra possible values to the Version field. I would suggest adding: - A CVS image with an extra field for the date/time of the last cvs update - An extra text field for the Other choice - An Unknown (with no extra fields)."
0 "The old Bugzilla had a nice feature - it had a link on the query page that allowed to see all my bugs (including those where I added myself to the CC list). It would be nice to have something similar in the new Bugzilla. A reasonable choice may be to add a checkbox to the query form with effect of showing only those bugs that have my e-mail somewhere (in CC list, in Description field - anywhere)."
1 "The query page ought to have a choice of either showing the bugs (like it does today) or showing activity in bugs (sort of like what the View Bug Activity link does now). I think the database tables we have now will support it (well, probably need to add an index or two to the activity table)."
0 "See URL given for sample HTML and test case. Again (similar to 311767) I expect we dont have the api yet in the content model, but ultimately its a style thing."
1 "This has driven me nuts ever since Communicator came out. Back with Netscape Gold the positioning was three seperate buttons. When Communicator came out they integrated it into one list button. I have wasted a long time just remembering to click on that stupid button to get the list of three other buttons so I can position my text. Could you please split it in to three seperate buttons on the button toolbar?"
0 "You should put in that GPF crash detector/reporter which is in Netscape 4.5b1 into Mozilla. Then you could set up a server side of the bug reporter which would delete multiple entries of the same crash."
0 "Something else you probably havent thought of is timebomb code. At some magical point, youll be releasing betas, and youll need a way to disable the engine after a particular date and time, or relative to the first launching of Raptor. This should be implemented in an XP way."
1 "The characters with french accents are not displayed correctly in the bookmarks list."
0 "In a style system where style sheets can cascade, it is important to support the CSS2 universal selector, *, as described in the CSS2 spec at the URL above, and with priority rules as defined at http://www.w3.org/TR/REC-CSS2/cascade.html#specificity This is needed in case a user style sheet defines, for example, a color or background-color for B, EM, DT, or some other similar element. The page author, to override this, would either need to define his color and background-color on every single element, or use the universal selector. This is why it is so important (but is only useful in the context of *cascading* style sheets.) ‰"
0 "This is a request for enhancement. Its overdue in mozilla, in the competition, and demanded by champions and users. When I or Chuck Simmons (chrlsim@futureone.com, netscape champion) or many others surf the wilds of the Internet, we turn JS off. Why? Paranoia, good sense, whatever -- it doesnt matter. Mozilla has not had its last security hole closed. This goes for JS, Java, HTML layout (remember the Danish form-type-change attack), netlib, etc. Then there are denial of service attacks to consider. OK, what can be done? This bug (really, RFE) asks that mozilla at least automate the disabling of executable content when surfing away from URLs that begin with host parts from a known-trustworthy set of fully-qualified domain names. That requires preference UI support, I suppose. Although with just the pref checking code in libmocha and the Java glue, and with a signed script that called navigator.preference, we (or anyone trusted) could construct a set your shields-up preferences page on mozilla.org, home.netscape.com, that acted as a web-server-based pref UI. So Mike, can you bug me about implementation, do the libmocha hacks (or find someone else to do them), then reassign this bug to raman for Java? We should figure out the pref syntax and value types first, make them common and extensible. After you and raman are done, we can give it to german for UI consideration -- but the web-based pref UI approach seems better to me. /be"
0 "Scroll wheel on newer MicroSoft and Logitech mice has no functionality. Works fine in most other win32 apps, where it manipulates the vertical scrollbar up and down. Am very used to it now and am missing it in Mozilla :)I am using the 9/17/98 daily build. -Matt Hargett (matt@darpanet.net) 9/24/98"
0 "CSS2 provides a mechanism for the author to designate a color based on aliases to system color defaults, rather than an explicit color name or hex value. Specified values are outlined at the above URL. Support for this would be particularly helpful to those trying to build an appliction interface with the native look and feel using only HTML/CSS."
0 "The setting of left, right, or both margins to auto should (respectively) right- justify, left-justify, or center a block whose width is less than 100%.‰"
1 "[This is just a reminder: peter, reassign this to kipp when you have implemented the css2 content attributes - thanks]"
0 "I have to wait forever for Tinderbox to load over my modem. There are really only two relevant pieces of information I go to tinderbox for 95% of the time. 1) wether or not the tree is open 2) wether or not any of the trees are in flames (which is really wether or not the tree is open when it says open). These two pieces of information can be made known much sooner (as in a minute or more) if you seperated them from the massive table, and just had that information in a much smaller table right above and flush with the massive table. thx"
1 "bugzilla cannot be easily port to non-mysql databases such as postgress or oracle. while things like bitint and tinytext are trivial to port; enum, timestamp, and datetime may not be easy. is there anyway to start moving the perl and sql parts towards sql92?"
1 "The inherit value is very useful for authors to develop stylesheets in cascades where they want to inherit values from the parent element. I dont think it is quite as important as the universal selector (bug #850), but it is worth implementing. ‰"
1 "There ought to be a button you can click that will make the window wide enough for the widest thing on a web page, but otherwise as narrow as possible (with a settable minimum, probably defaulting to 640)."
1 "according to css2, the relatively positioned object becomes a containing block which means we need to do the body-block wrapping stuff we do for absolute positioning. this doesnt help us do ILAYER, but at least we can make css2 people happy."
0 "A few things needs to be fixed, asap: 1. Documentation. 2. Line truncation to 80 characters. 3. Base64 encoding of binary data. 4. File handle management, passing files etc. 5. Clean up code."
0 "Some common problems are 1. Perl version, how to get binaries etc., how to compile etc. 2. Where to install LDAP SDK DLLs 3. Version of C SDK DLLs Having a complete package for NT would certainly help..."
0 "We need simple way of browsing the directory (a DN), its one of the most commonly asked questions. Something like $entry = $conn->browse($dn, [@attrs]); where the @attribute list is optional. This will effectively do a base search on the DN, with objectclass=* as the filter."
1 "So handle defaults like Base-DN, LDAP host, port numbers etc."
1 "In the 10/29 binary release, selecting a link which extends across multiple lines of text results in only the clicked line being selected. In Netscape 4, clicking a multi-line link turns all applicable lines to the active link color, specified in the body tag. Also, when loading pages that stretch for multiple screens, with several tables throughout the page, my Mozilla binary has to wait for the entire page to load before displaying anything on the page, even data not enclosed in a table. Its behavior could be changed to draw text and images on the screen if and when it has the horizontal spacing figured out, then continue drawing down the page. I realize this is just a developers version, and these are minor annoyances. I do generally like the new feel of the browser. Oh, and that green lizard icon is really cool. :)"
1 "allow Clear Location bar to have a text box associated with it to set the maximum number of URLS that can be placed here (0 being allowable as well - to turn off putting NEW URLS in this location bar)"
0 Allow multiple pop accounts under ONE user login name
1 "Netscape 4.5 Bookmarks and Windows 98 XMouse are incompatible."
1 "Currently, changes made via a document.write or equivalent are saved in a cache file and are printed in a wysiwyg: mode. If a user has cache turned off though, then this wysiwyg document.write change isnt sent to the printer. It should be what is on the screen is what you get regardless of the if cache is turned on or not. Also a developer has no way to check if the users cache is turned on or not. Regardless, the default should be to print a wysiwyg mode."
1 "when one has a radio button in a form, the outer part of the circle (the rest of the square), is white. This should be transparent, as it looks very non-professional on a non-white background."
0 "The view source removes all the line breaks and whitespace from comments (  ). The comments in this case are within a STYLE element. Perhaps comments within STYLE and SCRIPT should not lead to a change in color, but STYLE and SCRIPT themselves should? Handling CSS and JavaScript comments in the comment color would be even neater..."
1 "I do an optimized build with Direct Draw enabled. On 95, any area with opacity doesnt work, but on NT, it does. If I recompile without direct draw, the problem goes away. Both NT and 95 displays are 16bit color."
0 "The CSS specification do not detail how dotted borders should be drawn. Opera 3.50 draws them as zebra crossings when the border width increases. I think that is a better solution than nglayouts current square dots policy. Section (5) of the test page lets you test this."
1 "We need two methods to set/get options."
1 "From: kburns@netscape.com (Kevin L. Burns) Third, I added two new functions to Conn.pm. Both are almost completely the same as add(). The first one Ive called cc_add(). This is the same as add() only it returns $ret (the return value from ldap_add_s). I did this since LDAP_SUCCESS is 0. In an if statement in perl, 0 and false should work the same, e.g. if(0) if(undef()) and if(1 == 2) should all evaluate to false. Also, for a lot of reasons you probably dont care about, I never use an entry object, only a conn object. This makes your update() function break. To solve this I made a copy of cc_add() and changed the call from ldap_add_s to ldap_modify_s and called it cc_modify(). These were all trivial changes, but if you want them, they are below. ############################################################################# # Modify an object sub cc_modify { my ($self, $entry) = @_; my ($ref, $key, $val, %ent); my $ret = 1; my $gotcha = 0; $ref = ref($entry); if (($ref eq Mozilla::LDAP::Entry) || ($ref eq HASH)) { foreach $key (keys %{$entry}) { next if (($key eq dn) || ($key =~ /^_.+_$/)); $ent{$key} = $entry->{$key}; $gotcha++; } $ret = ldap_modify_s($self->{ld}, $entry->{dn}, \%ent) if $gotcha; } return($ret); } ############################################################################# # Add an object sub cc_add { my ($self, $entry) = @_; my ($ref, $key, $val, %ent); my $
0 "Today, when document.writeln() is called, it calls the parser with its text argument. Then, it calls again with plain newline. These should be combined so that the newline gets to the parser as part of the text. This will improve performance."
1 "At various times Mozilla will bring itself to the top and take the focus. In addition to being very confusing behaviour this has serious risks, I know for one Ive been in the process of entering a password when another instance of the browser came to the top, took the focus and got some portion of my password typed at it. If this is not exploitable (that is the focus never goes to anything that can recieve my typing), its at least annoying and should be corrected. Specifically, I dont believe Netscape should _ever_ come to the top/claim the focus even in error conditions. I notice this all the time when I have two or more browsers open, I enter the address Id like to visit in one, switch to the other to read the page there and as soon as the other page I requested is viewable it comes to the top!"
1 "The JavaSoft guys want to add a VERSION attribute to the object/applet/embed tags that allows the particular version of the Java VM to be specified. I think this needs to be done in a fairly general way, allowing not only the version of the JVM, but also the version of the plugin, the manufacturer, etc to be specified. The reason they want this is to facilitate going and getting the latest version of the Java Plugin if a particular applet needs some version-specific features. This relates to Alex Musils proposal for smarter installation of plugins that dont involve dumping the user at the plugin page, but instead install automatically. I think we also want to think about how we might support applets that belong to different JVMs simultaneously. It should be possible to install the JavaSoft Java Plugin, and the Microsoft Java Plugin, making one be the default, but allowing an applet to specify exactly which one they need to run. Maybe this relates to the clsid stuff in bug 255180. More Object tag stuff..."
1 "Gagan had a sweet idea to turn ftp:// layout and file:// layout and http:// directory layout into XML objects (is that the right term?). This would be cool for a number of reasons - but I like it because then the XPFE guys can use it to implement a sweet ftp client."
1 "Sometimes when Im surfing, I stumble across a site that wants to start up Java. On some sites Im willing to start up Java but on other sites (depending on what Im surfing) I dont want to pay the price of Java startup. My suggestion is to (hidden preference?) prompt the user whether to startup Java or not. If the user says OK (start Java), proceed as normal. If the user says no (dont start Java) then finish loading the page without Java (as if the user pressed the Stop button) or take the user back to the previous page. (The UE group may have an opinion on what happens when the user chooses not to start Java; you may chose to do something else...I leave it up to you.)"
1 "I think itd be great to have the ability to access my bookmarks from the right-click menu. That way, you could have a certain page appear in any frame currently displayed; great for cross-referencing material within a site."
0 "Sorting the bug list by importance should sort primarily on priority and secondarily on severity. The following change to buglist.cgi should do the trick: *** 285,291 **** last ORDER; }; /Import/ && do { ! $::FORM{order} = bugs.priority; last ORDER; }; /Assign/ && do { --- 285,291 ---- last ORDER; }; /Import/ && do { ! $::FORM{order} = bugs.priority, bugs.bug_severity; last ORDER; }; /Assign/ && do {"
1 "The other day the escalation people were trying to track down a specific bug fix in the ns tree. Since there wasnt a tag cut for this particular fix, it was like looking for a needle in the haystack. What would be useful would be some form of query mechanism that worked on the cvs checkin comments. That way you could do a search for a keyword or keywords and would return to you the comments and the files that had the keyword in it."
0 "(Should this be assigned to Content Model? Parser? Rendering? Viewer App?) cite, datetime, title, hreflang... should all be accessible to the user. See the quoted uri for some more information and for some ready made tests to cover this topic. Attribute Elements lang (all) title (all) cite Q, BLOCKQUOTE, INS, DEL datetime INS, DEL hreflang A, LINK href A, LINK longdesc IMG (has its own bug report) Currently, AFAICT only href is made available."
1 "(Should this be assigned to Content Model? Parser? Rendering? Viewer App?) The longdesc attribute of the IMG element contains a link to a file describing the image in detail, for those times where the image cannot be downloaded. The browser should supply some way of accessing this information. See also bug #1995"
1 "While a document is loading, reflow of the previous document is needed until the new document can be displayed. In more detail: 1) When the home page is loading (for those who have NGLAYOUT_HOME set to a non- local file), the viewer window should fill with something - probably white (or gray, if that is the default BG color). 2) When the window is resized or brought to the front while a document is loading, there is nothing displayed in the middle of the window, leaving whatever was there before. Drawing a blank screen would be better than what you do now, but you probably should draw the previous document, and allow movement within it (and allow clicking on a link to cancel the load in progress and start a new one). This would make NGLayout feel much more like a browser..."
1 "Once you get child selectors working, you can get compact DLs by putting the following in ua.css: DL[COMPACT] > DT {display: compact; } See the above URL."
1 "Link colors are always navy. Like on the Netscape Web Site the red links have navy blue underlines."
1 "We need a simple way to synchronize two LDAP entries. Perhaps something like $entry->synchronize($oldEntry); Well need to have some options/arguments to handle conflicts. Some alternatives are: 1. Always prioritize the old value. 2. Only get the old value if the new entry doesnt have the value already. This is a join of the two entries attibutes. 3. Like #2, but only if old entry doesnt have a value at all for the attr."
0 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #329783 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=329783 Imported into Bugzilla on 01/14/99 02:47) It would be very cool if a user could specify a .sig based on whether the outgoing mail was being sent internally within a company or externally. This perhaps could be accomplished by basing the .sig on if the from: domain corresponded to the to: domain. Examples: My internal .sig for Netscape Jim Race - x5410 Netcenter PD Quality Assurance http://webops/qa/ My external .sig for Outsiders Jim Race (650) 937-5410 Netcenter Product Development Quality Assurance http://home.netscape.com In this way, I can provide information which is of use to both audiences, all without changing prefs."
1 "This came out of bug 2254. When you re-size a page (or just load it initially), you can not do anything else until the re-size has finished. On a large page like this one, that takes quite awhile. This does not happen on Nav4.x. Using Jan. 14 build on win95."
0 "Form elements of type image need to be tab stopped & they need to have the cursor changed to a hand when the mouse is over them. The bug/feature is that were not treating input form elements uniquely. This creates an inconsistent feel when web browsing. For instance, go to the bugsplat login page. Its non-intuitive (web-wise) to click on something when theres not a hand there. IE supports this. Other similar sites would be the e*trade login site and Netcenters webmail login page."
0 "I know this is like the lowest priority ever, but is there any chance you could add the following to the viewer app menus? Report Bug link to http://bugzilla.mozilla.org/enter_bug.cgi?product=NGLayout Latest Build link to ftp://ftp.mozilla.org/pub/mozilla/nightly/latest/ Thanks ever so much."
0 "Currently the cookies.txt file gets deleted when you disable your cookies. An end-user has asked that we not delete the file. Here is the quote: Sometimes we want to stop netscape from receiving NEW cookies temporarily but we want to keep the cookies that we already have. This is REALLY annoying to have to go and undelete the cookies.txt file from the hard drive."
1 " You used to be able to call the mailbox from a Navigator window by using the command mailbox:// but this doesnt work in Communicator anymore. This needs to call up the Messaging Window - can this be put in.... thanks jones@netscape.com x4573"
0 "Include an enter new bug link at http://bugzilla.mozilla.org/post_bug.cgi Is there a way to include a link to enter a new bug at the page: http://bugzilla.mozilla.org/post_bug.cgi ? I find myself entering subsequent new bugs and would like to be able to quickly jump to the new bug entry screen (for the same project) without having to go read the bug I just entered and then click to enter a new bug. Thanks."
0 " The client wants to be able to print all the contents of his address book Currently there is no way to print the entire address book, the client does not want to use the export option, although exporting the address book to something like Word would be nice. ------- Additional Comments From granrose 10/29/98 11:22 ------- This is a commonly requested feature for Communicator, along with printing your bookmarks (and sorting your bookmarks by name). FYI, with 4.5 you can export the file to a comma-separated-value or tab-separated-value file which can be imported into many different apps (i.e. Excel) for printing, so that does give a workaround. But we should be able to do basic printing of the information from within Communicator, even if its just field:value with entries separate by a blank line."
1  I dont know if this is important or not, but it sounded interesting enough to record in bugsplat. Subject: Suggestion for Communicator ~ POP3 for OUTBOUND Mail via XTND XMIT Date: Sun, 29 Nov 1998 19:47:03 GMT From: jman@hal-pc.org (Carl Pearson) Organization: Houston Area League of PC Users Newsgroups: netscape.public.mozilla.general Howdy, All: Please excuse me if this is not the appropriate group. netscape.public.mozilla.wishlist does not appear on my server. Id like to suggest that Netscape have the capability to send email with POP3, via the XTND XMIT command, instead of just using SMTP. Reasons are varied, the main one being able to connect to your home server without dialing in (say, if you were on vacation in another city, using a friends account, etc.), yet till being able to send mail. Many providers do not accept SMTP unless youre dialed in, so as to limit spam. So far as I know, only Eudora Pro supports this feature. There, you simply leave the SMTP box blank in their setup screen. Thanx! -- My school colors were clear. We used to say, Im not naked, Im in the band. --- Steven Wright _________________________________________________________________ Carl Pearson   HFDS #070  It would be nice if the placement of the cc: and the QA contact: field can be switched since, normally, the user would only enter one name for QA contact, but multiple names for the cc: field. Itd be nicer to be able to see more of the cc: field names without having to scroll the text field. Thanks."
0 "Build: CCK Demo (from //Einstine/CCKDemo location) Steps: 1. Copied complete directory on the local drive. 2. Ran RunCCKDemo file. 3. Clicked Next on the Welcome screen. 4. The screens are overlayed on top of one over the other. Display Setting: 1024*768 pixels 256 colors Large Fonts"
1 "Test URIs: http://www.bath.ac.uk/%7Epy8ieh/internet/eviltests/link1.html http://www.bath.ac.uk/%7Epy8ieh/internet/eviltests/link2.html Expected Behaviour: http://www.w3.org/TR/REC-html40/struct/links.html#h-12.3 For this bug to be completely fixed, the UI must be up and running. However, as a start, NGLayout could actually understand and parse the LINK information, and then when it comes to implementing it in the UI, itll be just a matter of making links and titles available on a menu."
0 "We need to add better support for binary attributes, and also handling Base64 better. The LDIF module from Grahams LDAP package looks like a good example."
1 "the static html files are not update when the bannerhtml is changed. with more static files being added (two this week), a mechanism for a) placing a standard, site specific header and b) keeping it in sync would be nice. so, here is a patch for doeditparms.cgi which will edit the .html files and update the header. it does this by placing a meta-tag ( and ) in the html file. this is replaced with Param{bannerhtml} whenever the param changes. thus, each static html file needs to have these two tags inserted at the appropriate place. also, you need to edit the doeditparams.cgi program to do the updating. the @htmlfile array should be updated to list all static html files. here is the diff (BEWARE WORD WRAP IN PASTING!): [jay@bic prod]$ cvs -z3 diff doeditparams.cgi Index: doeditparams.cgi =================================================================== RCS file: /cvsroot/mozilla/webtools/bugzilla/doeditparams.cgi,v retrieving revision 1.4 diff -r1.4 doeditparams.cgi 65c65,74 < $::param{$i} = $::FORM{$i} --- > $::param{$i} = $::FORM{$i}; > > if ( $i eq bannerhtml ) { > my $ok = ChangeBanners(); > if ($ok ne ) { > print WARNING:  Unable to write one or more html files
\n; > print WARNING: Banner saved, but some html files unchanged.
\n; > print

$ok

\n; > } > 0 Commented out the part about customer information. If you need to see that, go back to the bugsplat bug report.> Communicator Professional Edition, Export, Physical Memory Available to Windows: 65,052 KB, 76% Free Tested on Windows 95, not on any other platform: Communicator 4.5 PR2 and M15, Windows 95 ------------------------------------------------------------------ 1 - I like being able to have long lines in my mail message wrap, however, found a small side effect which is a little annoying. If I save the message with wrapped lines the saved text message is saved with those lines wrapped. This has actually caused me to have to re-write a program that converts an emailed report to a format which I can easily import into Excel. The problem is annoying, however if I must choose between one or the other, I would choose to have long lines wrap. ------- Additional Comments From shartley 10/19/98 14:39 ------- Additional customer input on line wrapping....... (sorry!)... I remember a little better the problem I had with Netscape 4.5 pr2 line wrapping. I would like it if long lines could wrap without actually inserting newline characters into the message itself. I would also like it if the long lines simply wrapped to the width of the window rather than the line length set in preferences. In other words, if I view a message with long lines and those lines are wrapped to make it easier to 0 " RFE - need ability to save email without attachment The client wants to be able to save an email without the attachment. Message/Edit Message as New is not an option because the FROM address is lost ------- Additional Comments From granrose 10/26/98 10:47 ------- yoel - need more details. how about a test case with an example of what the customer is doing, what is happening, and what they want it to do instead. also, for customer-related bugs we have to put the customer name in the Summary so we can see what customers have filed which bugs. Once we have this info, reassign this to q_dse_client and we can send this to engineering. thx. ------- Additional Comments From yoel 10/29/98 6:41 ------- Customer is not using 4.5, but since 4.5 does not provide this fuunctionality either Ive used it as an example What the customer is doing currently: 1. Mail is in Inbox with attachment 2. MESSAGE|Edit as New 3. Delete attachement and Save 4. Move from Drafts folder to custom folder Problems with this current proccess: 1. Sender address is lost 2. Date & Time lost 3. Message-ID changed What Customer wants: The ability to delete the attachement from a message while in the inbox without losing any of the existing attributes of the message(i.e. sender, date & time, message-ID). - 1 (This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #340381 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=340381 Imported into Bugzilla on 02/04/99 17:01) We would like to see messenger enhanced. One of the key features I like about Netscape Messenger is the ability to sort message by thread. When I forward a message, the subject becomes [Fwd: ...] and its not recognized as being part of the original thread. Reply subjects get turned into Re: ... and those _are_ recognized as part of the original thread. 1 "<2/4/99. Transferred from Netscape internal bug system, bugsplat 331764. I commented out the customer name information. If you need this info, pls refer back to the bugsplat bug number.> Using the messenger client: 1. Click on message 2. Click on EDIT 3. Delete Message shows shortcut as alt-d ------- Additional Comments From granrose 10/28/98 09:48 ------- Im pretty sure this is something that can be controlled by the Netscape.ad file if doesnt want to wait. ------- Additional Comments From hannigan 10/29/98 16:14 ------- assg;d to rvelasco to see if it can be done as granrose suggests ------- Additional Comments From rvelasco 12/29/98 11:12 ------- Talked with Chris McAfee and Seth Spitzer and they warned me that making changes to this file may require the user to add a fix to the fix. The customer will just need to use the Alt-D to delete messages rather than the delete key. Resolving this as INVALID, best solution to this RFE is to leave well enough alone. ------- Additional Comments From yoel 01/19/99 16:18 ------- Im sorry but a doing a fix to a fix is not something that can be told to the customer. I need to know why this RFE cannot be done, is it that the coding is too exntensive. Is this something that can be latered? ------- Additional Comments From marek 01/25/99 22:22 ------- mass-setting to TFV 5.0 all enhancement requests against Communicator 4.5 ------- Additional Comments From rvelasco 01/26/99 11:49 ------- reassigning to 1 "IE offers a hover feature in the style sheets. Its rather useful. It causes a color text change when the mouse moves over TEXT. Any chance of adding this?" 1 "We like our QA department to screen new bugs and verify them before they are actually assigned to the initial owner. It would sure be nice if we could do this so we could verify that the bug is legitimate, and the information is complete. Then QA would assign it to the initial owner of the component." 1 "The bug owner and the TFV seem to be basic info for any bug. Id like to not have to click the link to get these values. There may be others that everyone can agree on... Thanks!" 1 "The different links types in HTML 4.0 are not implemented, they are specified here: http://www.w3.org/TR/REC-html40/types.html#h-6.12 Some useful examples: " 0 "lxr queries have a
tag in the form, but no
tag before the form is ended. They look fine in 4.5 (which auto-closes the center tag) but in gecko, everything stays centered, which I dont think was the intent." 0 "You do not support the :before and :after pseudo-elements, which are part of CSS2 generated content. This is defined in the spec: http://www.w3.org/TR/REC-CSS2/generate.html Generated content isa prerequisite to correctly implementing Q (see bug #1061) in the ua stylesheet, and would also be useful to QA in testing attribute support (in addition, of course, to being useful in its own right on the web!)." 1 "There should be hot keys for each button on the Wizard screens." 0 "It would be neat if tinderbox had a way to make a comment about a tinderbox machine (probably linked via a star in the name area of the column) so tinderbox owners can list relevant info about their machine, such as: contact info, libc version, compiler, and any other relevant software to the build (e.g. gtk version, make version, yaddayadda)" 0 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #343965 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=343965 Imported into Bugzilla on 02/26/99 11:33) 1) On a Mac when hovering cursor over IMAP mail folders, a folder status flag diplays so many unread of so many total. This behavior is not seen on the PC. Can this functionality be introduced to the PC client?" 1 "On the Navigation Toolbar, how about an option for full screen view?" 0 "From John Kristian: To make this work better, I propose to add this code (or something like it) to Entry.pm: sub get_LDIF_records { my ($self) = @_; my @record = (dn => $self->getDN()); foreach my $attr (@{$self->{_oc_order_}}) { next if ($attr =~ /^_.+_$/); next if $self->{_${attr}_deleted_}; push @record, ($attr => $self->{$attr}); # This is dangerous: @record and %$self now both contain # references to the same array. To avoid this, copy it: # push @record, ($attr => [@{$self->{$attr}}]); # But thats not necessary, because the array and its # contents are not modified as a side-effect of getting # other attributes, from this or other LDAP::Entrys. } return \@record; } OK with you? Got a better idea?" 0 "Support for CSS2 system fonts would be nice. These allow fonts from the users UI to be specified. See also bug 1004, relating to system colors." 1 "Apprunner: Layout is clamped to a fixed height and doesnt respect vertical resizing. Width is Ok. kipp says: I have fixes sitting in my tree that make the box model handling for block elements much *much* better. Having said that, I havet yet looked at the xul source for the top level apprunner chrome so I dont know if my fix will apply there. Im starting to think that they need to use a table to get what they want." 1 "ISPs are very interested in branding issues. They want what branding they are doing to show up in *current* profiles to the extent that is reasonable. That is they dont expect to have bookmarks replaced, but book marks added. They want home page to overwrite existing things. Security keys being added to existing databases and that sort of thing. The upshot is to pay a lot of attention to exsiting installs and to band them as well." 0 "Using 3/8 build on Win 95, Win 98, Win NT, Linux and 3/3 build on Mac 8.5. Open URL. Expected result: The first sentence should have a wider font than the second. Actual result: Font in the first sentence displays the same as font in the second sentence. This behavior is the same for all other values of font-stretch: narrower, ultra-condensed, extra-condensed, condensed, semi-condensed, semi-expanded, expanded, extra-expanded, and ultra-expanded." 1 "From Chris Saari: The idle_thead acting as a queue scheduler for I/O blocked threads and having a low priority are inverse goals. When I/O completes, you want the thread blocked on the I/O to run as quickly as possible. In UNIX systems, this usually means having threads waiting on I/O have a high priority, usually kernel level, but not be runnable until the I/O completes. Just before the blocked thread runs again, the priority is set back to its normal user level priority. How does this relate to NSPR threading on the Mac: Id like to have the I/O completion routine AsyncIOCompletion bump up the idle_thread priority if it fails to schedule the blocked thread itself because of a lock on the queues. The idle_thread itself can reset its priority back to normal when it is run. Thread priority is an issue that we may wish to address further. The current 5 level priority and round robin scheduling" 1 "It would be beneficial if the I/O buffering mechanism controlled thread blocking within NSPR instead of automatically blocking on all I/O. The benefit would be immediate I/O call return if the requested data is already buffered and in RAM. Ideally, the I/O stream/buffer mechanism would buffer at least the amount of data requested by the previous read to that I/O source. Under this model, the first I/O request would block and yield control to another thread, but subsequent requests of the same or smaller size would return immediately. This is a file ring buffer that I have partially implemented. However, this needs a global manager so that file coherency can be maintained across multiple reads/writes to the same open file. This would be a significant increase in concurrency, which is very low today if you have a small number of threads." 1 When creating a new list in an address book, Netscape ignores the order the list members were added, and instead seems to use a random order. Unfortunatelly, at times the order is critical; for example, when sending mail to Senators and their aides, politics dictates that the Senator email addresses appear prior to the aides. Please enter defect so that Netscape maintains, when creating or modifying lists, the order specified for the list members. From: Jason D. Blue Thu 3:13 PM Subject:Re: Case 99183 & 99173 To: smcarthur@netscape.com Stuart, FIRST, Regarding 99183: I believe I found the cause (or at least one of them). Try the following: 1.Create a new address book under 4.5. 2.Create five new cards in this address book, for example: a.a new card with a display name of McArthur,Stuart (without quotes). b.Create another new card, this one with a display name of Jason D. Blue. c.Create new card for John Public. d.Create a new card for Jamie Private e.Create the last new card, for John Blanco Losada. 3.Create a new list. 4.Drag to the list the above cards, one at a time, in the following specific order: e (John Blanco Losada) a (Jason D. Blue) c (John Pub 0 "It would be very nice if you could make grendelbot to inform us about Grendel horkages instead of SeaMonkey horkages." 0 "Ive copied text from bug 2235 (marked as a dup of this bug) which discusses this problem - it turned up with a user that was using the java-enabled javascript standalone shell and noticed very bad memory behavior. My explanation is at the bottom... roger, could you confirm my understanding of the problem? Mozilla presumably does register a branch callback, so this shouldnt be a problem with the browser." 0 "March 11 Seamonkey builds, Mac 8.5/G3, Linux 5.2, optimized, viewer March 10 win98 optimized, viewer 1. Launch Seamonkey and load test case page: slip/projects/marvin/html/pre_width.html 2. Width of text block displayed should be 20 characters; it is not. The pre width attribute is not being honored on any platform and appears to have not been honored on 4.51. While deprecated for 4.0 per spec it is still supposed to be supported for backwards compatibility." 0 "At least it needs to take a DN as the argument." 0 "On the test page you will see three sentences. The third sentence should be written from Right To Left. It is not. It is displayed normally or LTR To be fair, Nova does not handle this properly, nor does IE4. This bug is valid for all current (Mar10) builds of Seamonkey." 1 "Following on from bug #2808, it would be really useful if Mozilla could announce the MIME types of plug-ins the are available to it in the HTTP_ACCEPT header. It would then be possible to dynamically serve up different content, or point the user at a URL to download the required plug-in." 1 "PROBLEM: on all platforms, the element BASEFONT is ignored. im not sure if this is Layout, Parser, or Style... PLATFORMS: MacOS 8.51, WinNT 4, RedHat 5.2 TESTCASE: should be red sans-serif, size 7
should be red sans-serif, size 3 EXPECTED OUTPUT: view the testcase in IE4.x to see the correct output. the text on each line should describe its own style. ACTUAL OUTPUT: all text is the default font face, black, and default sized too. MOTIVATION: this is required by the HTML 4.0 Transitional DTD from the W3C." 0 "Consider a message, for example a bounce message, which has an imbedded message/rfc822. The user can click on the paperclip button, then double-click on the imbedded message/rfc822 to display the imbedded message. Unfortunately, the imbedded message cannot be forwarded. There should be some way to forward the imbedded message/rfc822, as if it were a top-level message." 1 "Using the following HTML snippet:
apprunner displays a text input control in the browser. However, when you click into it, behavior is different among platforms and in no case ideal. On Windows, the cursor shows up not at the upper left hand corner of the text area, but near the middle of the top row, and there appears to be a tab to the left of it (that is, you can press the left arrow key to go to the upper left hand corner). On the Mac, clicking into the text area places the insertion point about a space to the right of the expected upper left hand corner position, and you can press the left arrow key to go there. If its possible, please change this so that clicking into a textarea always places the cursor in the upper left hand corner of the area, and so that there are no tabs or spaces in the area." 1 "the split() method of String types is not supported, I think. It is described in the ECMAScript spec, section 15.5.4.8. See: http://www.ecma.ch/stand/ecma-262.htm The above URL contains an example of its use." 0 "How can I access a specific menu item, when its created from XML. What I want to do is set the status of the menu to checked when I create the frame." 1 "Michael Plitkins has just added APIs to nsIScrollableView to scroll the view by lines (or pages). The app can specify the definition of a line, and Steve Clark suggested that we use just-out-of-view line heights to decide how much to scroll, to ensure bringing a full line into view each time. Its a nice idea." 0 "I need to keep track of my teams progress, and report what bugs were found and fixed in their code each week. I was able to do this in the past using Bugsplat, which allowed queries for reporting, resolution and verification by date range. I dont see any way to efficiently use Bugzilla to do this. I tried to use the changed in the last n days feature, but that doesnt do the job because it isnt always (or even typically) the last n days Im concerned about. Also, it generates a slew of false positives (more than 90& of the total returned), since it includes any change to a bug. Could we get these fields added to Bugzilla, or is there some workaround?" 1 "Marking enhancement request." 1 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #335510 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=335510 Imported into Bugzilla on 03/26/99 10:21) cutomer contact: Frances Li fli@bear.com 973-793-5969` Problem: We need to create two Communicator shortcuts. One use the bear profile and the other use the default profile. When I had the default one up and then tried to launch the other shortcut, it didnt come up. Business: We need to provide a default profile for the users to do regular browsing. When they are in their home page, they can have a link that opens up another Communicator window that use the bear profile and bring them to the Web Shell login page. ------- Additional Comments From gshuss 11/12/98 12:46 ------- Frances, Currently Navigator does not support the second instance with another profile. In the future, additional instances with a different profile will be ignored and will use the original profile. Does this answer your question? Greg" 0 "The C (i.e. changes) links, on tinderbox are pretty meaningless because they may or may not represent what the tree actually pulled. If tinderbox pulled by date, then the C links could be accurate. This is assuming that files dont show up later because of the mirroring delay. The date should be at least as early as when the build began. (Perhaps 5 minutes earlier to factor in the mirroring delay). In other words, if a tree start to build at 10:32, then it should pull with a date of 10:32 (and maybe 10:27)." 0 "This bug report is created to document the contents of the NSPRPUB_RELEASE_3_1_BRANCH. This branch is rooted at the static tag NSPRPUB_RELEASE_3_1, which corresponds to the NSPR 3.1 release. NSPR 3.1.x patch releases will be produced off this branch. All checkins into this branch must be carefully reviewed and approved to ensure its stability. These include: - critical bug fixes; - Mac NSPR development; - new ports." 0 "This is an enhancement request, and not a bug. If the body tag specifies an invalid color as in these examples: or seamonkey renders the bgcolor as black. (communicator 4.51 and ie 5 choose other colors according to some process I cant divine.) If its possible and easy, it would be ideal to automatically switch the background color to white if the color specified for it is invalid." 0 The Local Mail component needs to be made aware of multiple accounts: - create a new local mail store for each account - make sure pop deposits mail from each account into the right inbox - expose this correctly in RDF Scott, Im CCing you on this one because youre the local mail guy - Im working on multiple preference stuff and Ill probably have my fingers in your code for this one. 0 "In our Mozilla QA Bug Writing Guidelines at: http://www.mozilla.org/quality/bug-writing-guidelines.html we suggest that bug writers use the following template when submitting a new bug. Client QA (and several CPD engineers who feel this would be extremely beneficial) would like to have this template be placed into the Description Area of New Bug Entry page. Could this be done in a way that folks can type in between the section titles and customize as needed? * DETAILED DESCRIPTION * STEPS TO REPRODUCE 1) 2) 3) * ACTUAL RESULTS * EXPECTED RESULTS * BUILD DATE AND PLATFORM FOUND * ADDITIONAL BUILDS AND PLATFORMS OCCURS * ADDITIONAL INFORMATION" 1 "Latest CCK Wizard Machine This is a request for feature enhancement. It may be beneficial to the CCK Wizard to incorporate Tool-tips in various areas such as: Animated Logo and Home Button URL. In doing so, we will be further supporting the ease of use aspect of the product." 1 "allow enhancement of the alert, confirm, prompt functions. For example, IE allows change of buttons names, change of default orders, change of icons, change of text in title bar, etc..." 1 "allow the user to access the User-Defined System Colors. This will allow users to programmers to create draggable OS-like popup dialogs, complete with active and inactive colored titlebars, that adapt to any color scheme the user has chosen. This is possible in IE. A discussion of how this is possible, can be seen at: http://www.webreference.com/dhtml/column24/" 0 "We in editor land would like to be able to control the COLS attribute on preformatted text from style. This will be used to control wrapping width in plain text mail compose." 1 "I would like to see a preference that allows for multiple browser windows (with different start pages) to launch at start-up. At least two. For example, I would then be able to launch My Netscape and Calendar Express when my browser launched, in addition to Mail and anything else." 0 "From: Olaf.Schneider@Materna.DE Hello Leif! Using the object-orientated Ldap Interface ive had the following problem: If the result of a search operation is greater then the LDAP_SIZELIMIT or LDAP_ADMINLIMIT the search Method drops the result and delivers the perl script an empty result. I have fixed this bug temporary with the following change (line 19-23 and 31-36 and in the function search URL). Maybe we should set a variable in the conn object so that the programmer can determine that the limit was exceeded. 00 ############################################################################# 01 # Normal LDAP search. Note that this will actually perform LDAP URL searches 02 # if the filter string looks like a proper URL. 03 # 04 sub search 05 { 06 my ($self, $basedn, $scope, $filter, $attrsonly, @attrs) = @_; 07 my ($resv, $entry); 08 my $res = \$resv; 09 10 $scope = Mozilla::LDAP::Utils::str2Scope($scope); 11 $filter = (objectclass=*) if ($filter =~ /^ALL$/i); 12 13 if (defined($self->{ldres})) 14 { 15 ldap_msgfree($self->{ldres}); 16 undef $self->{ldres}; 17 } 18 if (ldap_is_ldap_url($filter)) 19 { my $rc; 20 $rc=ldap_url_search_s($self->{ld}, $filter, $attrsonly, $res); 21 22 if (rc==LDAP_SUCCESS || rc==LDAP_TIMELIMIT_EXCEEDED || 23 rc==LDAP_SIZELIMIT_EXCEEDED ||rc==LDAP_ADMINLIMIT_EXCEEDED) 24 { 25 $self->{ldres} = $res; 26 $self->{ldfe} = 1; 27 $entry = $self->nextEntry(); 28 } 29 } 30 else 1 "Id like the Conn::search method to handle calls like search($base, $scope, $filter, @attr); This seems like a much more common/useful feature." 1 "Need a way to do things like search($base, $scope, $filter, DN-ONLY); or something like that. Other convenint shortcuts would be ALL for (objectclass=*) etc. Id like to look through all such cases, and make the appropriate additions to Conn.pm and Entry.pm." 0 "zooming needs to be finished. you have the code as i left it." 1 "fixed." 0 "tb_in_sf@my-dejanews.com makes this suggestion: by default, displaying the current user profile name in the Title Bar of the Messenger window. Otherwise its hard to tell what mailbox Im looking at - it just says Inbox - Netscape Folder. I tried renaming the Inbox.* files with a test profile but that didnt work (and seems like a really bad idea anyway). -- I like this idea - the messenger toolbar should show you your current context - at least the identity, maybe the actual account that is in use." 0 "Id suggest that the Summary be added to the list of required fields for new bug reports. Why: The only scenarios (from a two minute search through every browser bug in Bugzilla) in which the field is blank are when either: (a) someone accidentally types the information into the Status Whiteboard or URL field (b) someone is new to Bugzilla and really doesnt yet know how to use Bugzilla. By making this change: In scenario (a), Bugzilla would be providing a useful hey, did you really mean to do that? to the bug writer. In scenario (b), Bugzilla would be teaching the novice bug writer how to use Bugzilla. Also, it would slightly reduce time wastage on the part of Bugzilla users, who have to open up individual bugs that lack summary fields in order to determine whether its a significant bug or not. (e.g. after doing a query)" 0 "These diffs cover: o Reports for -All- are collected and displayed. o The no banner flag, which was weird, has been rationalized. o Default to links on, since it makes little visual but much functional difference. I never would have guessed the way to attach is to first enter the bug, then as a seperate step make an attachment. But Ill try it. -Bryce" 0 "PR_Opens PR_SYNC flag is not documented in the NSPR documentation at http://www.mozilla.org/docs/refList/refNSPR/." 0 "On Unix, NSPRs build system is based on GNU make. Build options are make variables (such as BUILD_OPT, USE_PTHREADS, etc.) specified either as environment variables or on the gmake command line. Specifying the NSPR make variables appears to be a significant learning obstable to developers of other products that have to build NSPR from the source, e.g., SeaMonkey. To them, the NSPR make variables are yet another build system to learn. They want to be able to just say gmake. Ramiro proposes that nsprpub/config/config.mk include user-specified makefiles (if they exist), in which the build options can be specified and the values of certain make variables can be overriden. The user config and overrides makefiles could be generated by the configure script of the product that uses NSPR. Then, the user could just type gmake in nsprpub. At the beginning of nsprpub/config/config.mk, we include nsprpub/config/my_config.mk, if it exists. The purpose of this makefile is to set build options, which are the input to the NSPR build system. At the end of nsprpub/config/config.mk, we include nsprpub/config/my_overrides.mk, if it exists. The purpose of this makefile is to override the values of certain make variables, which are the output of the NSPR build system. This is a general mechanism for bridging the build system of another product with NSPRs build system. It is not specific to SeaMonkey. Can I check this in?" 1 "seamonkey QA has written smoketests for seamonkey builds. These are already running (in Perl) on unix and win32. To create these particular tests (and others), we will need AppleEvents enabled. In particular (a naked minimum) we should be able to launch Apprunner, open a page, and quit." 0 "Enhancement request for INI File Editor: Provide the ability to set bold text and to change the default font size of text. This is useful primarily for creating the title of a wizard screen, so that the title stands out from the rest of the text on the screen." 0 "In bugzilla, when you run a query you get a pretty list. When you begin to look at the bugs on that list you have the option to jump to the First, Last, Previous, and Next. How about an option to jump back and look at the whole list instead of having to sort the thorugh the browser back button drop-down?" 0 "Ive always said itd would be nice to be able to create a bugzilla list by inputing bug numbers. This comes in handy when you want to make changes to a few bugs at once and you already know the numbers. instead of trying to cook up a query that is a near match and selecting from that, you could just enter in the numbers. Ive also noticed that Bugsplat seems to have this feature. Well how about us?!" 1 "it only works on linux and solaris, should be enhanced for other unixes. The LD_LIBRARY_PATH thing needs to be fixed." 0 "Id like to be able to upload a stack trace or test file when I file a new bug, not just later." 0 "This method will rename an attribute, e.g. $entry->rename(uid, oldUID);" 1 "This is a placeholder bug, so I dont forget. Our make system for xpidl has several explicit references to glib-config, because libIDL-config assumes that glib is located in the same place as libIDL. Andrew Veliath has fixed this for libIDL 0.6.8; when we require people to install a version >= 0.6.8 libIDL, we should clean up our code to remove the redundant references." 0 "This patch allows me to specify a timestamp for cvs checkouts, so I can grab the last known good date from the bonsai page, e.g.: MOZ_CO_DATE=04/20/99 10:25:49 PDT gmake -f client.mk checkout Note that there seem to be some problems if you specify MOZ_CO_BRANCH and MOZ_CO_DATE simultaneously, and Im not quite sure how to handle clearing sticky tags, so this might need a little more work, but for this purpose it works for me." 1 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #344968 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=344968 Imported into Bugzilla on 04/20/99 17:18) Customers need a way to print URL, title, date, ... on Unix platfroms like on Windows." 0 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #347577 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=347577 Imported into Bugzilla on 04/20/99 17:52) When you do a search on a parent folder or on a parent parent folder you get the results in a window which consists of several columns like: Subject|Sender|Date|Priority|Location So assume you have a hierarchy of folders like: taddei.mailbox.cern.ch INBOX sent-mail Foo1 bar1 Foo2 bar2 Bar3 foo3 when you search on the root, you might find messages into bar1 and foo3 and the problem is that the search will only display bar1 or foo3 in the Location column. This is incomplete because you would like to have the path being prepended. For users who have huge collections of several hundred of folders hierarchically organized this is very hard to find back under which collection is this or that folder displayed in the location bar. In addition it sould probably be clickable such that you could open directly the concerned folder such that you can jump in it. Here many people start to use the mail system as their PRIMARY information system and they expect a high level integration and ways to retrieve more easily information and its context in this case a folder. It helps them grouping and organizing their data more easily which will 1 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #348614 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=348614 Imported into Bugzilla on 04/20/99 17:56) The outgoing messages cannot be filtered and moved to the appropriate folder like the incoming messages. They can only be copied to ONE folder while sending. I believe if we expand the Message Filter capability also to the outgoing messages, it will be of great help in oraganizing the outgoing messages. For example, I have folder created for all the customers I work with. I keep all the correspondece with my customers in their folders. I have filter set up for my incoming messages such that the incoming messages from a customer go to its folder. But when I reply to the customers messages, all the replies go to Sent folder and I had to manually move them to the appropriate folders." 0 "When the build breaks because of Perl errors, this does not show up on the Tinderbox log." 1 "Incremental reflow causes a repaint of the entire window. This bug has come from http://bugzilla.mozilla.org/show_bug.cgi?id=3289," 0 "If bug 0002 is resolved as a duplicate of bug 0001 then bug 0001 automagically gets Bug 0002 has been marked as a duplicate of this bug and thanks to another cool feature the words Bug 0002 are a hyperlink to that bug. Unfortunately, bug 0002 itself gets only the following words: *** This bug has been marked as a duplicate of 4679 *** and therefore doesnt get to take advantage of the hyperlinking. IMHO all that has to be done is change the above line to: *** This bug has been marked as a duplicate of bug 4679 *** and everything should be golden. This comes in handy when you are looking at the dupe and want to look at what it is supposedly a dupe of." 0 "Patch to cut down the Windows temp directory getter code, and add code to get the Windows user home directory." 1 "build id: 1999042608 platforms this happens on: windows nt, mac os 8.5.1 (linux crashes before opening the file) to reproduce this, select open from the file menu and open a local file. result: the file opens, but in a new window. expected result: the file opens in the current browser window." 1 "For background, please see bug #3264. Because of this lack, nsIFileLocator is currently not available in viewer. This is not a big deal, but perhaps it is a bigger deal for embedding applications." 0 "First, congradulations Terry and Sam for an outstanding piece of software. Still getting to know it, but love it already. To business: A nice enhancement of Bugzilla would be to include the bug-writing-guidelines.html in the release. Bugzilla is kinda slight on documentation. IMHO ;-) Its up to you. Related bug: 4456" 0 "I have added the capability to add new components via CGI. The capability makes these changes: Add addcomponent.cgi Add doaddcomponent.cgi Modify editcomponents.cgi (CVS version 1.1) I am including the changes in diff format. I hope thats ok. I apologize if this is not the proper procedure for submitting patches. I browsed www.mozilla.org briefly, but did not have time to fully determine the best means to submit this. :-o The editcomponents.cgi modifications also include the one line modification I submitted in a previous bug concerning the link to editowners.cgi. diff -c orig/addcomponent.cgi new/addcomponent.cgi *** orig/addcomponent.cgi Tue Apr 27 22:19:43 1999 --- new/addcomponent.cgi Tue Apr 27 21:08:50 1999 *************** *** 0 **** --- 1,99 ---- + #!/usr/bonsaitools/bin/perl -w + # -*- Mode: perl; indent-tabs-mode: nil -*- + # + # The contents of this file are subject to the Mozilla Public License + # Version 1.0 (the License); you may not use this file except in + # compliance with the License. You may obtain a copy of the License at + # http://www.mozilla.org/MPL/ + # + # Software distributed under the License is distributed on an AS IS + # basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the + # License for the specific language governing rights and limitations + # under the License. + # + # The Original Code is the Bugzilla Bug Tracking System. + # + # The Initial Developer of the Original Code is Netscape Communications + # Corporati 0 "Ive had several folks if they can have a canned Show My Open Bugs link on the Main Page. What do ya think?" 1 "The option should take a URL and should print the page at that URL (preferably, but not necessarily, without opening a browser window) Example: apprunner -print http://www.apple.com" 1 "Do we need to do this for M6?" 0 "When loading a text/css file as the primary document, we ignore it. We should at least open it as a plain text file." 1 "I often want to check on bugs that I have filed with bugzilla. To do this I go the the search screen and look for bugs filed by me. While this is not too much trouble it would be a lot more useful if the main bugzilla screen had a link to that would search for each of the bugs I filed automatically. thanks mo" 0 "Project: Seamonkey Program: Apprunner.exe BuildID: 1999-04-29 Platform: Win NT Need new command-line syntax for Apprunner to load xul files in a separate window. They are loaded inside the standard navigator.xul when executed from command line: apprunner resource:/res// Example: 1. From the command window, run: apprunner resource:/res/samples/skindemo.xul Result -> Apprunner loads Skin demo inside Navigator.xul Requested Feature: Need command option for Skin demo to be loaded separately in its own window." 0 "Perhaps I grow mad. But. The cited page employs the mysterious
tag: MOC2-119
Apollinaris Patera
April 1999 Enforcer 5.0 and Ancestor 4.5 parse
as a valid line-break. Mozilla snorts and ignores it, refusing to issue a newline. Um. I suppose this may be considered a matter of syntactic fascism in the parser. Personally I believe that browsers should reinforce the writing of proper HTML/CSS by stingily withholding the render-magic for malformed script. But make of it what ye will." 0 "Heikki Toivonen from the DocZilla project suggested this - Unlike every other compiler on the planet, xpidl currently doesnt accept -Ifoo, but instead requires -I foo I learned it the first way myself, but it seems likely that xpidl will borrow $(INCLUDES) lines from other build systems, so it makes sense to follow other compilers there." 1 "Im not sure under which category or who this goes too but -- With the AOL/Netscape Instant Message window, it would be nice to be able to contact a server to check if a user is logged in or not. This would give a use3r the ability to lets say change an icon color if a user is in or not.(Check ICQ on how they do this? Might want uniformity since the idea is the same?) For example, with ICQ, you can have information on a WEB PAGE alerting people if they are logged in at the present moment and can chat with them if you wish. This would be a great feature to be included with Netscapes IM." 1 "Run WizardMachine. Go down the the new path until final screen comes up. Select BUILD INSTALLER button Result: Build starts up with no prompt to user and there is no way to abort the build." 1 "Run WizardMachine. Go down the the new path until final screen comes up. Select BUILD INSTALLER button Result: Build starts up with no prompt to user even if all pages have no new entires. Should be, at the very least, some kind of minimum configuration prerequisite before a build is allowed. Or have a minimum page configuration requirement before user can go to next screen." 0 When libmime parses messages looking for links, it looks for fully qualified URLs such as http://www.mozilla.org or mailto:phil@netscape.com. Increasingly, people are leaving the URL schemes off and just sending www.mozilla.org or phil@netscape.com. I think we should make our linkifying code more aggressive, perhaps keying off www for web pages and @ for email addresses. I dont know if theres a lot of danger of false hits here, but we could make this additional aggressiveness a pref if it seems dangerous. 0 "With this patch, build-moz-smoke.pl will output what a binary did to turn the tree orange -- rather than just telling you it crashed or quit, itll either say exited with status X or died on signal Y. The patch also cleans up some useless code in the binary execution: turning off buffering in Perl doesnt carry over after exec(), unfortunately. (Ive posted a comment about this on netscape.public.mozilla.builds and .unix.) (I wasnt sure whether this patch should go in bugzilla for the browser or webtools; I hope this is okay.)" 1 "WebNFS is not implemented yet :-( (-> what about Necko ?) (Question: Where can I suggest such features ? I id not found such a page :-( )" 0 "We want to be able to customize the browser windows title bar text by appending the Company Name value that the CCK user enters in the CCK Wizard (Customizing the Browser Part One screen) to the browsers title bar text. (Example: Netscape by Maxwell Communications). Not sure what Component category this bug belongs to, so I just took a guess." 1 "An option to view all errors found with the displayed document would be very useful for developers (especially since many do not bother to use validators, assuming that if it looks ok in their browser then it must obviously be correct). Such an option should report errors in the XML, HTML and CSS related to the document, as well as any problems discovered when trying to download related resources (eg, any HTTP errors like 402s when downloading images or 404s when downloading stylesheets). [ccing original contributors to the idea on mozilla-layout]" 0 "Noticed this while using the 5/11 build. Is the problem in table Layout, or somewhere else?" 1 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #327323 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=327323 Imported into Bugzilla on 05/11/99 15:52) 9/28 RTM Candidate - NT From Messenger, highlight an IMAP server and go to the menu File|Subscribe. A dialog called Commnicator: Subscrive to IMAP folders comes up with 3 tabs - All, Search and New. the first tab works - but it looks like Search and New were taken from the code to subscribe to newsgroups and never made to work for IMAP folders Search tab - The field labeled Folder: is always grayed out - so you cant enter the folder name that youre searching for. Note that if you select an news server instead it works. New tab - doesnt work either. All the buttons (including Refresh List) on the page are grayed out. The text above the folder list also refers to Newsgroups intead of IMAP folders." 1 "CCK Wizard screen Gathering Information screen (info.ini) has two checkboxes that need the following logic associated with them: CD-ready installer checkbox: If user checks this box, the wizard displays the CD Autorun customization screen (branding_page2.ini). Otherwise, this screen shouldnt be displayed in the wizard. Set default to checked state. Internet Setup checkbox: If user checks this box, the wizard displays the Customizing Internet Setup (branding_page3.ini) and Customizing Internet Setup - Points of Presence screens (AccountSetup.ini). Otherwise, these screens shouldnt be displayed. Set default to checked state." 1 "It would be very useful to have a not regexp search option for the Status Whiteboard field, like there is for the e-mail fields. An example of how this could be useful: Kipp has asked David and I to monitor his bugs while he is away. Our job is to remove duplicates and invalids. To make this possible, we need a way of distinguishing between new NEW bugs and old NEW bugs - that is, new bugs which we have already examined vs new bugs which we have not yet read. The easiest way to mark bugs as being read would be to add a consistent marker in the Status Whiteaboard field, such as [READ]. Then we could simply look for all the bugs without that marker to know which bugs we must next look at. Unfortunately, I cannot currently see a way of looking for bugs which do _not_ say a particular thing in the Status Whiteboard (or summary, or description, or url, for that matter). This feature would be a great help." 0 "It would be very cool to have some command-line queries for bugzilla, like SGIs pvquery. I dont remember the exact syntax of pvquery, but it had flags for the various options you might want to pass in in a bug query, for example (Im making up flag letters here) pvquery -Aakkana -P1,2 -snew,assigned -Rkostello -MM1,M2,M3,M4,M5,M6 -Sclipboard might get me all the bugs assigned to akkana, priority 1 or 2, status new or assigned, reported by kostello, milestones1 through 6, with clipboard in the summary (note I picked s for status and S for summary; alternately maybe S for summary and arguments without a flag match stuff in the summary?) The main thought is that a command-line query like this might be faster than going through a browser, and it would be easier to redirect output without dealing with 4.xs annoying reformatting when converting copied text to plaintext (though hopefully mozilla will soon be to the point where we dont need to use 4.x any more :-)" 1 "Remember the convenient way SGIs ptools revision control system was tied to the pv bug system, so when you checked in, if you put the bug number in the subject line, the bug was automatically closed with a note about the checkin? Wouldnt it be great if we could do that too? Put some code in the cvs checkin message, for example, if I put FIX 1234 or PARTIAL FIX 1234 in the bug message, then bug 1234 would have a note appended to it noting the date, the files and the rest of the checkin message, and if it was FIX rather than PARTIAL FIX, the bug would be resolved as FIXED at the same time. This would help QA, too, since theyre always asking developers to note when particular checkins were made to fix bugs, so they know which builds should have the fix, and were always forgetting to make that information explicit." 0 "Currently, on multi-user systems we do not recognize the security firewalling that is done to prevent users from harming themselves and others. In particular, on NT we force administrators to open up shared directories to general write access so their users can run our products. It would be very desirable to fix this such that profiles lived within the OSs concept of users. There are some proposals on netscape.public.mozilla.prefs which address this concept." 1 "Varada,this is an enhancement. We should clear high priority bugs away before starting on this. If we implement this, well need a way to determine the dimensions of a bitmap image. We may also need to figure out the dimensions of gif and animaged gif files since we also prompt for those." 1 "Thanks to waterson for this suggestion - xpidl should use #line, #file directives for %{ blocks. Maybe not for generated .h proper, as xpidl will munge the original .idl arbitrarily in creating the generated .h. But %{ blocks come directly from the .idl file, so we might as well point compilers and debuggers to the correct place." 1 "As I say it, it would be nice to distribute libjpeg.so.62 with mozilla. I do not know where I could get it." 1 "Latest Profile Creator dialog 1. Clean run of Apprunner (Delete Users50\, delete mozregistry.dat in C:\Windows\ prior to running apprunner.exe) 2. Proceed through each screen of the profile creator. 3. View the bottom of each screen Results: There is a horizontal scroll bar on each screen that is not necessary." 1 "Latest Profile Creator 1. Launch Apprunner on Clean system (delete Users50\, delete Windows\mozregistry.dat prior to running apprunner.exe) 2. Select Next button on first screen 3. Select Next button on the 2nd screen (prof. name & default profile directory) 4. View the fields on the Enter Name and information... (3rd screen of Profile Creator) Results: Missing SMTP, NNTP, POP/IMAP text fields. There is plenty of room to consolidate these text fields onto this dialog." 1 "The CSS2 outline property would allow widget focus to be drawn using spec compliant methods. There is a test page here: http://www.bath.ac.uk/%7Epy8ieh/internet/eviltests/outline.html See bug 1859." 0 "This one line patch ignores case in bug queries when using string search fields for short_desc, long_desc, bug_file_loc, and status_whiteboard. I will submit the patch as an attachment." 1 "You may not want this enhancement, but it seemed to me that you might want to whine about ASSIGNED bugs which have not changed. And also about REMIND and LATER bugs. So I expanded the concept of whining to include these. Note: The modifications were designed to minimize the differences between the official Bugzilla and my version. Thus the whinedays parameter was not changed to whinedays_new. And the whining was all left in whineatnews.pl. You may wish to change this if you accept the enhancement. I have some questions about the design and operations philosophy behind the RESOLVED REMIND and RESOLVED LATER bug states, but will address those on the discussion group." 0 "Latest Profile Manager (delivered 5/17/99) Assumptions: More than one profile has been created. 1. Open DOS box, go to the Apprunner\ directory, and type apprunner -ProfileManager to launch the Profile Manager 2. Select an existing profile in the pick list 3. Select the Rename button and proceed to rename the profile 4. Open MS-Explorer and go to the Users50\ directory. Results: The profile name listed in the pick list was updated. However, the profiles folder in Users50\ was not renamed. Expected: I expected the profiles folder in Users50\ to be renamed as well." 0 "For some odd reason, Im getting an error -48 (duplicate filename) when i call SetDragFlavorItemData() from my drag send proc. When I set the data as part of AddDragFlavor, everything works perfectly. This is just a placeholder so i dont forget, because I would really prefer to use the drag send proc." 1 "A good user interface for alternate and user stylesheets would be a very useful (and neat) feature. Alternate and user stylesheets were among the original design goals of CSS and an interface for using them has not been properly implemented in any browser. Ive written a document describing what I think should be part of such an interface: http://www.fas.harvard.edu/~dbaron/css/ssui/ The functions needed to implement part of the interface are already implemented in the style code, as I describe in the page above. Note that requests for this feature have been featured in WSP reviews of MSIE and Opera: http://www.webstandards.org/css/winie/#Alternate_stylesheet_UI http://www.webstandards.org/css/opera/#Alternate_stylesheet_UI" 0 "When I try to compile the 5/20/1999 CVS source of Mozilla for the Motif toolkit (against Lesstif 0.88.1 here on Red Hat Linux 6.0), I get the following compiler errors: nsMotifWidgetFactory.cpp: In method `nsresult nsWidgetFactory::CreateInstance(cl ass nsISupports *, const struct nsIID &, void **): nsMotifWidgetFactory.cpp:158: cannot allocate an object of type `nsWindow nsMotifWidgetFactory.cpp:158: since the following virtual functions are abstra ct: nsMotifWidgetFactory.cpp:158: nsresult nsIWidget::AddMenuListener(class nsIMen uListener *) nsMotifWidgetFactory.cpp:158: nsresult nsIWidget::SetVerticalScrollbar(class n sIWidget *) nsMotifWidgetFactory.cpp:158: nsresult nsIWidget::EnableFileDrop(PRBool) nsMotifWidgetFactory.cpp:161: cannot allocate an object of type `nsCheckButton nsMotifWidgetFactory.cpp:161: since the following virtual functions are abstra ct: nsMotifWidgetFactory.cpp:161: nsresult nsIWidget::AddMenuListener(class nsIMen uListener *) nsMotifWidgetFactory.cpp:161: nsresult nsIWidget::SetVerticalScrollbar(class n sIWidget *) nsMotifWidgetFactory.cpp:161: nsresult nsIWidget::EnableFileDrop(PRBool) nsMotifWidgetFactory.cpp:164: cannot allocate an object of type `nsButton nsMotifWidgetFactory.cpp:164: since the following virtual functi 1 "Unix printing should launch lp/lpr with a title (that banner pages get a more usefull info than job xxxx, for example). Maybe this can be implemented by printf-Style formattings: %t Page title (be carefull with the charset; maybe page charset must be converted to system charset, chars like must be escaped...) %f Page file (e.g. xxx.html) %u URL ---- The default should be lpr -T %t (for BSD systems like Linux) lp -t %t for Solaris (System 5 !?) OSes" 1 "Trying to install build M5 on GNU/Linux running on a 586 - following the M5 Release Notes at http://www.mozilla.org/projects/seamonkey/release-notes/m5.html The following is the log: martin@localhost:/home/martin > mkdir mozilla martin@localhost:/home/martin > mv mozilla*.tar.gz mozilla martin@localhost:/home/martin > cd mozilla martin@localhost:/home/martin/mozilla > gzip -dc mozilla-i686-pc-linux-gnu-M5.tar.gz | tar -xvf - martin@localhost:/home/martin/mozilla > mv mozilla*.tar.gz ../ martin@localhost:/home/martin/mozilla > cd package martin@localhost:/home/martin/mozilla/package > ./mozilla-apprunner.sh MOZILLA_FIVE_HOME=/home/martin/mozilla/package LD_LIBRARY_PATH=/home/martin/mozilla/package:/opt/kde/lib:/usr/local/rvplayer5.0 MOZ_PROGRAM=./apprunner moz_debug=0 moz_debugger= .//run-mozilla.sh: ./apprunner: No such file or directory martin@localhost:/home/martin/mozilla/package > ls PrimitiveTest libmimetype.so libremoturl.so SimpleTypeLib libmork.so libsecfree.so apprunner libmozdbm.so libsockstuburl.so bsdecho libmozjs.so libstubsj.so chrome libmozutil.so libwidgetgtk.so components libmsgbaseutil.so libxmltok.so flash-registry.rdf libnetcache.so libxp.so libabouturl.so libnetcnvts.so libxpcom.so libappcores.so libnetlib.so libxpt.so libcaps.so 0 "Yet another enhancement request... I just had a mid air collision with someone editing the same bug at the same time. While that was moderately fun (!), it should not have been possible. One way around this problem is to put a hidden field on the bug update form, containing the last modification date of the bug. When a bug change is submitted, it should then be refused if this field does not match the last change date in the database *at submission time*. That way, if someone edits the bug I am editing, and submits before me, I do not overwrite their changes. Instead, I get to read what the other person has written before submitting my changes. This should be reasonably easy to implement." 0 "Im seeing a lot of general cases where one bug cant be worked on or completed until another is fixed. In many cases, this problem is nested where several bugs must be cleared before a particular target bug can be worked on. For planning purposes, it would be extremely useful to have a depends on field where we could list the bugs that must be fixed before this bug can be fixed. Having them in a separate field would allow people to do useful queries such as bugs with no dependencies, bugs with dependencies, or all bugs depending on bug xxxx. An interesting co-feature would be to automatically bump the priority of prerequisite bugs when a dependent bug is higher priority (although I dont know if that would be politically acceptable around here :-)" 1 "Current state: Bugzilla marks bugs as duplicate, no futher mail is sent. Should be: Duplicate bug reporters should be CC:ed to the main bug (e.g. the first bug report about this particular problem), AND/OR if the bug has been fixed, all duplicate bugs should be marked as fixed and mails to the bug reporters should be sent that their reported problema are gone." 0 "One of the most missed features of nn<=4 SELECT element is that it is not scrolling to display the selected element when javascript sets selectObject.selectedIndex Now m5 does some scrolling but when MULTIPLE is set the scrolling is stopping one or two OPTION elements above the (newly) selected OPTION. I guess neither HTML4 nor DOM level 1 prescribe that the selected element has to be displayed but if you are doing some scrolling I dont understand why it is not done to display the selected element. So but Here is a control label" 0 "Build ID: 1999092508 Platform: all The UI for opening files and Web sites (URLs) has deteriorated significantly over the past several weeks. What has happened exactly? It seemed that we had the problem solved in early 1999: there was a single item in the File menu labeled Open; the keyboard accelerator was O and the shortcut was Ctrl+O. The dialog that it would bring up allowed you to type in a URL, or click Choose File to select a local file. Additionally, there was a check box for an implicit pref to open the file in a new window. The current UI is poor. There are too many things in the File menu, and the keyboard shortcut for Open File does not make sense. Ctrl+L was (if I remember correctly) used in previous Netscape browsers (pre 4.x?) for Open Location; pressing Ctrl+L and getting the file picker box is confusing. Have any final decisions been made in this area?" 1 "JavaScript needs a way of including script files into other script files, at least for the purposes of some of the heavyweight scripts that float around in the chrome folder. Example: If Im working on EdAdvancedEdit.js and want a utility function in another file, I have to find that file, load up the corresponding XUL file, and add a anywhere after a but the first is parsed as the closing SCRIPT tag which is wrong. In N4.x I have to include the HTML comment tags around the JAvascript in order for it to not parse the tag. I would think that mozilla wouldnt need the html comments. But even with them Mozilla barfs and shows the javascript after the first tag, instead of executing the alert command. I imagine this is a problem on OS = all. But I can only verify WinNT 2000020708 right now." 1 "It would be nice if keywords could be defined to have restrictions. Examples could be along the lines of: beta1 : milestone < m15 css1 : component = Style System This could help ensure keywords are used properly and are removed when they no longer apply. If a keyword is added and the restriction fails, the change should be refused and the changer should go back and fix it. If changes made break a restriction where a keyword already present, the change should again be refused. If changes made result in a restriction that didnt previously hold still not holding, a warning should be issued on the process bug screen. This is for legacy/transitional purposes since the person making the change might not be the one in a position to fix the restrictions. There could also be some whining emails sent by a periodic agent that checks for these as well. I dont particularly care about the interface for defining the restrictions, but if there was a web capability, the query screen selection could be used." 1 "I think a good feature to have would be reverse mouse scrolling for the mouse wheel such that scrolling up would really scroll down. This would be usefull for my 4 button mouse under linux because right now I can scroll up but not down with this 4th button." 1 "Dear Mozilla: I have been in communication with you about this problem literally for YEARS. I have received numerous messages from your staff assuring me that your Beta or final version will allow AUTOMATIC SORTING OF BOOKMARKS BY NAME, a terrible defect that goes back three years to the first 4.0 browser and that has yet to be corrected. I have not seen the Sorting option at all in Mozilla yet (even though your list of features mentions Sorting as one of the working options). As we come very close to the release of Beta 1, I cannot help but wonder just WHEN will this blessed event take place. When will I be able to automatically sort by Mozilla (and hopefully Netscape 5.0) bookmarks by NAME. I emphasize again: AUTOMATICALLY (by drag and drop), just I used to do it in Netscape 1.x, 2.x and 3.x years ago. It is my considered opinion as a long-standing user of Netscape that Mozilla 5.0 or Netscape 5.0 without this critical AUTOMATIC sorting of bookmarks is doomed to the garbage heap of browsers. This should have been fixed years ago and has no doubt cost Netscape millions of users and, I have no doubt, as much as I like Netscape, will cost Mozilla very dearly indeed. Only a madman would try to organize his thousands of bookmarks MANUALLY when the technology to organize them automatically has been available in Netscape itself from the very beginning. I have been tilting at windmills about this for months and months now. I do hope to see this Automatic Sorting of Bookmarks 1 When we fetch a message from the IMAP server, the server may have other things to tell us about, like the arrival of new messages. From a user experience POV, it might be nice if we set aside the unsolicited responses until the message had been downloaded, and then processed the unsolicited responses. This might make the product appear more responsive to the user. 1 "We should add this to some sort of debugging log file. From Steves document: To print out http headers, add the following to the SetHeader methods in nsHTTPRequest.cpp and nsHTTPResponse.cpp NS_METHOD nsHTTPRequest::SetHeader(nsIAtom* i_Header, const char* i_Value) { /* display headers as they get set */ nsString atom; i_Header->ToString(atom); char* s = atom.ToNewCString(); printf(-----REQUEST: <%s> <%s>\n,s, i_Value); return mHeaders.SetHeader(i_Header, i_Value); } nsresult nsHTTPResponse::SetHeader(nsIAtom* i_Header, const char* i_Value) { /* display headers as they get set */ nsString atom; i_Header->ToString(atom); char* s = atom.ToNewCString(); printf(-----RESPONSE: <%s> <%s>\n,s, i_Value); return mHeaders.SetHeader(i_Header, i_Value); } To print out the raw traffic received by the browser, add the following print statements: 1. mozilla/netwerk/protocol/http/src nsHTTPResponseListener.cpp after the call to ReadSegment in ParseStatusLine: printf(*** status = %s\n, mHeaderBuffer.GetBuffer()); 2. mozilla/netwerk/protocol/http/src nsHTTPResponseListener.cpp after the call to ReadSegment in ParseHTTPHeader(): printf(*** header = %s\n, mHeaderBuffer.GetBuffer()); 3. mozilla/netwerk/streamconv/Converters nsMultiMixedConv.cpp after the call to inStr->Read in OnDataAvailable(): printf(*** multipart=%s\n, buffer); To print out the raw traffic send by the browser, add the following print stat 0 "prlink.h should add the functions: NSPR_API(void (*)()) PR_FindFunctionSymbol(PRLibrary *lib, const char *name); NSPR_API(void (*)()) PR_FindFunctionSymbolAndLibrary(const char *name, PRLibrary **lib); According to the ANSI C standard, it is not portable to cast a function pointer to (void *) or back. Function pointers are allowed to be larger than data pointers. The casting appears to work on all of NSPRs current platforms, but on some of them the compiler generates a warning when the value returned by PR_FindSymbol() is cast to a function pointer. Deferring this to NSPR 4.1 would be OK." 0 "Patch should only take 5 minutes to evaluate and apply. Patch Description: This patch sets the filename on buglist.cgi, long_list.cgi, and reports.cgi output so that when the page is saved the extension is .html. Problem Solved: When saving Bugzilla output, a .cgi extension was used. Mailing the report to someone meant that they could not view the report as it was not interpreted as HTML. Problem Not Solved: This does not fix the problem when mailing the CGI output to someone (using Netscape Messenger for Solaris). There are a couple of other ways to do this which may solve this problem, but this patch was the easiest to implement." 1 "As stated in the comments above SplitEnumType in globals.pl: Assumes that enums dont ever contain an apostrophe!. This is due to the following line in sub SplitEnumType: while ($guts =~ /^\([^\]*)\,(.*)$/) { At our site, we tried to use Wont Fix rather than WONTFIX when we customized the Resolution values. While customization of resolutions may be considered not supported by Bugzilla, customization of severities, platforms, OSs, and priorities definitely are supported. These are also enum types, and SplitEnumType is used on all of them. It is not at all far-fetched that someone would want a severity such as Cant Run or Wont Run. We worked around the problem by avoiding the apostrophe, but I submit the following regexp for consideration, to force parsing up to the next occurence of ,, rather than stopping at an apostrophe in the middle of a value (not immediately followed by a comma): /^\(.*?)\,(.*)$/ I have unit-tested it on the following strings: my @testcases; push @testcases, qq{enum(,Fixed,Invalid,Wont Fix,Future Version,Remind,Duplicate,Cannot Reproduce)}; push @testcases, qq{enum(,Fixed,Invalid,Wont Fix,Good, Bad, Ugly,Remind,Duplicate,Cannot Reproduce)}; push @testcases, qq{enum(Cannot Reproduce)}; However, the above is not sufficient, due to MySQL quoting conventions, where an apostrophe is expressed as a double apostrophe in the output of desc (see attachment: demo of apostrophe in enum value). The extra apostrophe can be handled by changing 1 "In many Windows apps that have a tabluar layout (Excel comes to mind), if you double-click in between the column header, then that whole column will be resized to make all the data in the column visible. It would be nice to add this feature to Mozilla. I think this is related to the Tree Widget. -dave" 1 "so far the text form control does not show a right mouse click context menu, but I think it would be great if it does show one, including commands of cut/paste/copy/delete/.... itd also be nice for the url text control in the navigation toolbar to show a context menu. minor defect ... a single line text control shows its scroll bar when the user types something and then clears (backspace, delete, etc.) the whole thing, which is a bit nasty." 1 "... rather than waiting until Ive committed the initial report to go back & add in keywords. e.g. to mark perf, dogfood, etc. with the initial submission. Currently the Enter Bug UI doesnt have the keyword field. Sorry if this is a DUP; I have so many bugs to file right now before I leave on a plane I cant query." 0 "A few days ago I realized that it would be fairly easy to implement basic JS source editing in our editor: Take the selection, turn it into html via the usual route, then pop up an insert source dialog populated with the source of the existing selection (so the user sees a normal text area filled with the source). When the user clicks OK, just replace the existing source with the edited source. Filing this as a reminder so I wont forget." 0 "We find that the Component field of the long_list table output very frequently wraps onto more than one line, when it appears that there is plenty of room to fit it on one line. In my understanding, Target Milestone is a short field (not longer than Component), so it should work to give COLSPAN=2 to the table cells for both Component and Target Milestone. diff -r1.12 long_list.cgi 104c104 < print Component: $component\n; --- > print Component: $component\n; 106c106 < print Target milestone:$target_milestone\n; --- > print Target milestone:$target_milestone\n;" 1 "reports.cgi produces a list like this: X bugs for hacker@usually-netscape.com For those not actually on the list and those who dont (and dont want to) know whos working on what, it wont be bad if there was an option to also show the components for which the owner is responsible (if any). That way the summary reports might look something like this: X bugs for coder@work.com [XML XUL] Or maybe it might just be better to have a report on bugs according to components? Well, both wont hurt the user much..." 1 "At the moment, people leave their votes on bugs that have been closed. It would be good if we could remind them of this periodically. Not too much, since we dont want to discourage use of the voting system, say every 3 months or so. Basically the reminder should say that you have votes for a bug that has been resolved for x amount of time. For a resolved duplicate bug, it should say that votes arent transferred and why, and also how to transfer your votes to the new duplicate bug. You might want to handle the situation where the second bug is resolved as a duplicate of a third bug, or some other resolution. For all resolutions, it should explain how to remove the votes and prevent further whines. This could help keep the voting system vibrant, especially with all the bugs that weve seen Slashdotted, the voters never heard from again" 1 "Open link in new window - this cause to set active window to the new one. Ill try to explane what I want: W1 - window where is a link W2 - new window In W1 we want to open a link in a new window. Right click on the link (I hope bug 17921 will be rosolved sooner as in M20) -> open in new window. The W2 is now active, but I want to stay on the on W1, while W2 loads. But when I CTRL-TAB to W1 from W2 W2 becomes again active one during loading... This sucks - I must again CTRL-TAB." 1 "When you fill out a form and hit enter, mozilla should seek out and automatically press the Sign in button on the page, or the equivalent of that. For example, when logging into bugzilla, after entering your password, hitting Enter should press the Log In button, without you having to tab down to it or use the mouse to click on it." 1 "Right now on *NIX platforms when a user runs Netscape 4.x for the first time the default ~/.netscape profile dir is created silently and netscape starts (without any profile wizards). I think that Mozilla should act in the same way on *NIX platforms. When a user runs mozilla the first time a default profile is created (Im guessing ~/.mozilla/MozProfile/) without prompting and a less advanced user could start surfing right away. A more advanced user would of course still have the option of several profiles but the average user wouldnt need to understand them. I hope Im making some sense here, if not, ask me for clarification =)" 1 "Having created applet that require the SUN Java 1.2 (or higher) plug-in we have had some problems when running the applets behind strict firewall/proxy setups. The set-up is like: one Win98 client with no DNS or gateway; only IP number of a http proxy server. Running the applet in HotJava is fine, can fetch images and make urlconnections. In the plug-in, however, no connections back to the server seems to be allowed because the JVM is unable to resolve www.server.domain and it seems not to trust the proxy settings. In the general case this is unfortunate for many users setting behind corporate firewalls, completely shielded by strict proxy servers. Adding to the hosts file is generally not an option for most users and is at the least very clumsy. As an observation: Apples MRJ (which uses OJI) actually seems to trust the browser to handle URLconnections correctly. The main reason to file this report is concern about the limitations if the SUN security model is implemented and at the least to have clarified specs in this area." 0 "When you search the bugzilla database for recorded bugs you sometimes make less intelligent queries resulting in maybe 800 matches. This is something you dont know until the results page is done loading which can be really annoying on a slow connection. My suggestion is that the amount of matching bugs should be printed at the top of the bug list so that you can stop, go back and make a more specific query if you see that you got a couple of hundred matches" 0 "Latest ns cck installer build dated: 02-10-00 O/S: WinNT4.0 Steps to Reproduce: 1. Install latest build of CCK 2. Run Add/Remove Programs in Control Panel Results: Netscape CCK is not listed as a program that can be uninstalled." 0 "On Friday, Nelson Bolyard wrote: I have an IRC client that I run on Win98 and NT. Ive used it succesfully with a number of irc servers on a number of irc nets over the last year or so. But It doesnt work with irc.mozilla.org. When I connect to most irc servers with it, immediately after the TCP connection is made, I start to see messages appear in the status window. On some servers, these messages proceed to tell me my reverse DNS looked-up name, and my nickname. Some servers attempt to contact my identd (which I dont have, and they couldnt connect to it anyway), and then throw me off. Others tell me my nickname is not valid. And on many servers, it just works and Im in, and the rooms work OK. But all of them at least say SOMETHING to me. None of them just disconnect me with no messages at all. When I connect to irc.mozilla.org, no text whatsoever appears from the server when the TCP connection completes. 33 seconds later, the client reports to me that the server dropped the connection. Sadly, the server isnt giving me ANY clues about what it doesnt like about me/my client. I thought that perhaps someone with admin privileges on irc.mozilla.org might be able to tell me why irc.mozilla.org is unhappy with my client, perhaps by looking in some log file, or perhaps by telling me all the tests/hoops that I have to pass/jump-through to be acceptable to it. Lord suggested that you might be able to help me with that. If youre willing, and have any answers 1 "A sample link like foo should display a bar (like a tooltip) when mouse points to this link. Its a part of the specs (http://www.w3.org/TR/html4/struct/links.html#h-12.1.4) and the same thing works in IE4,5 and Opera 3.6. This feature is much more flexible with lots of JavaScript onMouseOvers, because I dont like that I cant see the target URL when I point to a link and status bar says a stupid phrase like go to blabla." 0 "In the upper left corner of the Create New Profile window there is a icon. Currently is a Windows Application Default icon. Expected: There shouldnt be an icon!" 0 "The URL parser should support RFC 2732 syntax. For example, it should handle http://[207.200.73.41]/" 1 "In the recent nightly builds the Search button is located alongside the URL bar. It looks isolated and unbalances the interface because it differs from all the other UI components on display. The original design had the Search button within the URL bar. I would request that it be put back there because this was an innovative solution that explicitly linked the search functionality with the URL bar and made excellent use of the available space." 1 "It would be nice to be able to sync your address book with a Palm organizer also under Gnome/Linux. Gnome has a CORBA-based Palm conduit architecture in place, see http://www.gnome.org/gnome-pilot/" 1 "Build: 2000021608, MacOS 8.6 The current design for the location bar includes a `Search button at the right end of the bar. This design has a major drawback: many, many users will think that `Search in Mozilla means the same as `Go in IE, and they will click on `Search whenever they enter an URL in the location bar. This problem will be particularly bad for users migrating to Mozilla from other browsers sporting `Go buttons (IE 5, for example). Presumably the Search button as implemented would have a fall-back to detect URLs and redirect to the URL instead of the search results, but this would have three problems: * the user might never realize that they were doing things the slow way; * the user might never realize the original purpose of the search button; * if the URL-detection and redirection was done by the search engine, rather than the browser, the extra redirection step would give the impression that the browser was slower than it actually was. Thats why Im reporting this as a bug rather than an RFE. SOLUTION A suggestion I made in n.p.m.ui was unanimously endorsed by those who responded. That is, for a popup menu to be placed to the left of the location bar. This popup menu would contain the following items: * Location: * Search the Web: * Search Usenet: * Whats related to: etc. (Hint: Extending this menu with sponsored items such as `Stock symbol:, `Shop for:, `Show map of:, and so on, would be a nice little money-earner for popular Mozilla di 1 Add an item Copy Article Location, which would put a news: URL for a newsgroup article on the clipboard, to the context menu in the thread pane for newsgroups. 0 "In the installer there should be an seperate Shortcut Icons dialog. The dialog should be presented after or together with the Program Folder option in the custom install. This dialog should give the user the following options: - Add icon to Desktop - Add icon to Quick Launch bar - Add icon to Start Menu Programs - current user - Add icon to Start Menu Programs - all users (Windows NT/2000 only)" 1 See bug # 17578 in Browser -- I would also like to be able to authenticate using Kerberos in Mail. IMAP supports this in the protocol and (I believe) both UW IMAP and Cyrus provide implementations through SASL. 1 "Back in the Navigator 1.x - 3.x days it was possible to customise the number of simultaneous connections to be made while downloading a website. This feature has since disappeared. It would be nice to be able to specify this again as well as other related options like downloading URLs on the same server as the main HTML document first (or not), prioritising visible objects (over ones that are down the page, or above the current display position), etc." 1 "Platforms: Windows (Consumer) 98 and later, Windows (Server) 2000 and later If Mozilla is ever to be able to *completely* replace IE on Windows, it will need to be able to do the Windows Update thing. This will involve several tasks. * Replace the `Windows Update IE hooks in the Start menu, Windows Help, and possibly other places, with Mozilla hooks. * Get Mozilla to pretend to be IE, whenever visiting the Windows Update site. * Reverse-engineer, and implement, whatever protocol is used to (a) determine what updates are available for a given OS installation, and (b) install them. * Constantly keep an eye on this feature, to catch up with whatever games Microsoft plays with the windowsupdate.microsoft.com server to try and prevent Mozilla from accessing it. This feature doesnt have to be built into the browser, however. A stand-alone Windows Update app, written in XUL and JS, could provide a better user interface then IEs Web version -- as well as having the ability to check for updates of itself (and Mozilla) at the same time as it checks for updates of Windows components. `An error has occurred in accessing the Windows Update server. This may be due to a change made by Microsoft to the server ... However, a new version of Mozilla Update is available, which may fix this problem. Would you like to download it now? Im going to mark this as helpwanted, because its really a job for a committed volunteer or team of volunteers." 1 "Build: 2000021608 To reproduce: * Install Mozilla. What happens: * A `Mozilla shortcut appears on the Desktop. What should happen: * The installer should ask you if you want Mozilla shortcuts placed in any of the following places: - on the Desktop - in the Apple menu - in a category of your choice in the Launcher (with a popup menu to select the category, based on the contents of {System disk}:System Folder:Launcher Items:). The `on the Desktop checkbox should be checked by default, and the others unchecked." 0 "The usability of the sidebar could be greatly enhanced if the sidebar panels looked more like something the user is used to. They could look like the cards in a card file, as in the URL given. A small (~3-pt) bevelled border around the outside of the panel, contiguous with the border around the panel heading itself, would be needed to make the panel as a whole look like a cardfile card." 0 I have a non-Mozilla biff running. If I read a msg with Mozilla (or 4.x), Mozilla doesnt update the state to READ (not even SEEN?), so my biff keeps notifying me of new mail, which isnt new anymore. I have to close (!) Mailnews (or Messenger) in order to update. Please give me an option to syncronizise with the server, when Mozilla gets new mail. At least set SEEN plag, better also READ flag and expurge (delete mails on server, which I deleted in the client). 1 "Greetings -- the My Bugs link in the nice yellow box doesnt show Unconfirmed bugs. I think it should show unconfirmed bugs. I found it when I learned more information about a bug I submitted .. it surprised me when I couldnt find it through My Bugs. hehe. :) Thanks! :)" 1 "The very first time a given email address (user) files a bug, it would be great if they would automatically be sent a one-time-only thanks, welcome, and get more involved email message saying the following: - Thanks for testing Mozilla and for filing your bug. - If you havent already attached a simplified test case to your bug, please do; it will help QA and engineering to fix the bug sooner. - If you have the time, please join us in making Mozilla the best browser possible by helping to simplify bug reports in the BugAThon at http://www.mozilla.org/newlayout/bugathon.html - Please make sure your bug report complies with the QA bug writing guidelines at http://www.mozilla.org/quality/bug-writing-guidelines.html - [list of QA URLs worth reading] Implementing this would require having a one-bit flag for each email address in the database default to off when the account was created, and set to on (sending the message) the first time a bug was filed." 1 "IE4/5 allow to call elementReference.click() to simulate a mouse click on all elements which allow an onclick event handler, e.g. links, radio buttons, checkboxes. NN2/3/4 for some unexplained reasons have the click method only for radio buttons and checkboxes but not for links for example. I think NN5 should do away with this asymmetry and provide a click method for all elements it provides an onclick handler for. It is quite often that you want to click a link with js, till now you have to write something alike to get that working with NN4. That is unnecessarily complicated for a rather simple thing the browser DOM api should provide." 1 "There should be someway of attaching a menu to a button, such as is used in the Back and Forward buttons for 4< browsers. Discussion in netscape.public.mozilla.ui has suggested that the IE-style double button is preferable to the Netscape 4 style click and hold to display menu, click to do default action." 1 "In mozilla versions <= 4.x, the third/middle mouse button, when selecting a link, would open said link in a new window." 1 "This may conflict with the XP philosophy, but under Unices there are two buffers, and versions of mozilla =< 4.x would put selected text into the passive buffer (as nearly all X apps do). Similar behavior in 5 would be great." 1 "The entries on the owners page are rather sparse and confusing to the novice. I suggest two enhancements: 1. Add a description field to the owners database containing a sentence or two of explanation as to what the module is. 2. Possibly add a project field to the owners database that could cross-reference back to the projects on the projects page . Id be glad to help write module descriptions and to try and coordinate modules with projects." 0 "The Getting Involved page briefly mentions using cvs diff -u to prepare patches for submission, but I think many users will need more explanation than that. I suggest modifying the Getting Involved page to add more info, or (preferably) creating an entire new page How to Submit Patches. The page would describe two overall ways of preparing and submitting patches: 1. Developers with their own Mozilla CVS repository should use cvs diff -u and redirect output to a file: $ cvs diff -u >patchfile.txt Does the developer need to be at a certain level in the Mozilla CVS tree? If so, this should be added to the example. What about Windows developers; would they use this same command? If not, what? What about Mac developers? 2. Users who do not have their own CVS tree should use the diff tool or equivalent. The easiest case is for single files: $ diff -u oldfile newfile >patchfile Should they follow a particular convention for naming the files, for example, name the new file as the old filename appended with -new? $ diff -u foo.xul foo.xul-new Or should they rename the old file with a -old suffix and have the changed file have the original name? $ diff -u foo.xul-old foo.xul Which way works better for people applying the patch? How about Windows users? Where can they get a copy of diff if they dont already have it? From the relevant page of the Open Directory: http://dmoz.org/Computers/Software/File_Management/File_Comparison/Windows/ there appear to be several 1 "The font pull down (shows Ariel, variable, ...) does not have the same number of fonts as 4.x. Should have a larger list." 1 "Endico suggested that we get rid of the Whats Related panel, and add mozine to the sidebar for the initial load of the sidebar, prior to customization. The following attached file does that. It needs to go into /bin/defaults/profile/, but i dunno where that comes from before mozilla is built." 0 "I have played around with file and ftp url directory listings. I noticed that the ftp http-index-format was incorrectly formatted. The code in nsFTPListingConv.cpp uses a nsIDateFormat object which depends on the locale settings (sv_SE in my case). http://lxr.mozilla.org/seamonkey/source/netwerk/streamconv/converters/nsFTPDirListingConv.cpp#877 This date should be formatted as a RFC822 date, which it is in the file url. http://lxr.mozilla.org/seamonkey/source/netwerk/base/src/nsDirectoryIndexStream.cpp#226 So I took the code from the file url and modified nsFTPListingConv. See the patch I have attached to this bug." 1 "Although I am not at all a fan of IE, there is one feature which I think would be nice to have in Mozilla--full screen browsing. For those with small monitors, it would greatly enhance the browsing experience by allowing one to see more of the page. Would it be possible to create such a feature using XUL?" 1 "There has been much talk at MozillaZine about features that should be included in Mozilla, one of them is an autocomplete for the URL field. Its a general concensus, that I personally agree with, that the Netscape Communicator autocomplete method is awkward and not as useful as it should be, and the IE autocomplete method should be adopted: type a few characters of the URL will pop up a layer listing possible choices, which update itself as more more the URL is typed." 1 "Change Platform to CPU (or CPU type or family) or at least change: PC to i386, x86, or IA-32 DEC (which doesnt exist anymore) to Alpha Id like to see the others changed as well, it just makes way more sense DEC is now Compaq which also makes PCs. HP makes PC as well as UNIX boxen. SGI is switching to Linux in the future and Compaq may too. There no entry for AIX/PPC. Saying Linux/DEC, Linux/Macintosh, or Linux/Sun always seems really wrong For completeness Sun -> Sparc, SGI -> Mips, Macintosh -> PPC and then add an entry for IA-64 while your at it. I understand that this is something core to the database organization, but Im not asking to re-do the whole database, just rename the existing names." 1 "Build: 2000021608, MacOS 8.6 To reproduce: * Start the Account Wizard to create a new e-mail account. * Enter your e-mail address as `abc@foo.bar at the appropriate point. * Continue to the page for specifying the incoming and outgoing mail server names. What should happen: * Since its extremely likely that, given an e-mail address of `abc@foo.bar, the incoming and outgoing mail server names will be of the form `xyz.foo.bar, `.foo.bar should be placed in the incoming and outgoing server name fields automatically. This feature should be smart enough to update the server name fields if I change my e-mail address while still in the wizard, but *not* once Ive made any changes to the server name fields myself. (In the same way that the contents of the `Display Name field in 4.xs address book is automatically generated from the `First Name and `Last Name fields, *until* I make any changes to the `Display Name field myself.)" 1 "I IE you have the option to right-click on a webpage and choosing Show Links which will open a new window showing the links in the current webpage. This could be made into Mozilla too. Just instead of opening a new window a sidebar would appear. I willing to write JS code is I knew how to open a new sidebar..." 1 "It would be nice if it were possible to hook into the history mechanism in Mozilla so that an external program could be run for visited pages, etc. at the users request. This could be done to, for instance, save all visited pages in an external (and faster) database table for analysis, later viewing, frequency checking of domains, etc." 1 "The wheel mouse dont work with the main windows!!!!!!" 0 "Currently, servers are hard-coded in initHost . If ChatZilla were invokable with irc://host[/#channel]-style URLs, servers could be stored in bookmarks." 1 "In previous versions of Netscape the ESC-key was used to stop loading a page. IMHO this was a very useful feature. For all who use the keyboard to scroll and prefer to use the mouse as little as possible, or atleast try to move the mouse as little as possible, the keyboard functions was worth a lot. (TABing to links, ESC to stop load, ALT-Arrowkeys to go back etc etc) Will this be implemented in mozilla?" 0 "What I mean is the when a bug is marked duplicate, the reporter of that bug and perhaps even its cc list should be added to the bug that the reports bug was a duplicate of." 1 "THere is no easy way to login to bugzilla without entering/changing a bug. This makes it so that you cant get your saved querys immediatly." 1 Currently multiple POP3 accounts are supported, but only one SMTP host can be defined at any time. If an SMTP host was entered along with each POP3 account then it would simply be a case of selecting the correct FROM: address to select the appropriate SMTP and the reply-to address as well. 0 "I love the new mail format. It would help, though, for dependency changed mail to include the summary of the dependency bug. Right now I know Bug A blah blah blah (in subject) had its dependent bug XXXXX change state, but no idea what bug XXXXX is. While youre at it, a link to bug XXXXX in addition to the link to the unchanged bug A would be helpful." 1 "geturl.exe doesnt seem to notice when it gets 404 errors.. would it be possible to add an optional third parameter (after url and filename) for a file to put the headers in?" 0 "Suggestion: Shouldnt there be a publishing facility in the editor, which make it possible to publish documents to a Netscape Enterprise Server (Iplanet Web Server 4.0)?" 1 "Addition of a contextual menu item when text is selected on a web page. The item View Source at Selection would bring up the page source code with the web page selected text already highlighted. This assumes that the browser is using the internal source viewer and it has not been redefined as an external program. Not sure how complicated it would be to select the text in the view source window, but if the window could be brought up to the point of the select would still be benifitial to developers that have large complicated pages." 0 "Platform: ALL OS: ALL This is a feature request (Ill write it myself if you point me in the right direction). It would be nice if, on first starting Mozilla, that the user could be shielded from entering any profile information, and a default one is set up. The user could be called default or similar. This could be made a preference, just a case of editing a file (e.g. all.js) before startup. This would be of benefit where it will be guaranteed that there will be only one user EVER using the s/w on a particular machine. It is also useful for third party apps using Mozilla that need to by-pass this." 0 "Users without the capability to edit a bug should not be able to remove other users from the CC list, but should still be able to remove themselves. A good way to solve this would be add a checkbox for Email me when this bug changes. If the user has the capability to edit bugs, then they should still get the existing textbox like we do currently, as well, but if the user doesnt have that capability, or if they havent logged in, then the CC list can be included in a hidden field instead of a textbox, so that it still gets submitted with the update, but the users dont have the opportunity to change it. (This also helps to protect the privacy of people who want themselves CCd on a bug). Then the cgi can look at that checkbox and add or remove that persons name as necessary when processing it. (It would also have to be preset if their name was in the CC list when they viewed it, if they were logged in at the time)." 0 "The FTP client code needs to support the EPSV command on IPv6 connections." 0 "Bugs with different confidentiality settings should show up in bug lists with obvious distingushing marks, perhaps using different colors. This will help keep humans from inadvertently disclosing information about these bugs to inappropriate areas." 1 "It would be nice (especially since the new email tech sends individual letters instead of mass-mailing everyone) if the reporters address were included on any email about changes to a bug, even if the reporter isnt the person who changed it. (In the old email system you could just look at the To: or Cc: line)" 0 "Description is available at: http://www.wrec.org/Drafts/draft-ietf-wrec-wpad- 01.txt Short desc: Allows browsers to automatically locate a proxy-auto-config (PAC) file. Please support at least the Well Known Alias discovery mechanism. IE 5+ supports WPAD, and is *very* usefull for sysadmins. Please have this turned on by default. It should be pretty easy to implement, here is my take. In order for machine snoopy.mis.mycompany.com to find its PAC file, it tries these URLs to find the PAC. if DNSnameExists(wpad.mis.mycompany.com) try http://wpad.mis.mycompany.com/wpad.dat elseif DNSnameExists(wpad.mycompany.com) try http://wpad.mycompany.com/wpad.dat else no proxy All a company needs to do is create a DNS A or CNAME record for wpad pointed at their webserver and create a wpad.dat file which is their PAC file (setting the MIME type appropriately on the webserver for wpad.dat). Again, please have this be the *DEFAULT* behavior." 1 "I think it would be really neat if you kept track and displayed the percentage of people coming/going through mozilla.org using the Mozilla browser. That kind of statistic is relatively trivial (perl ... /var/log/apache/access.log) and I think it would be a great motivator to get this project into its beta stage. It would be nice to see how many people are actually using Mozilla (although the sample set would be biased for this particular web page), and definitely encouraging." 1 "It would be really nice, especially for vision-impaired people, if you could set the font shape, size, color, and the message pane background color for plain text message display. NS 4.7, e.g., displays messages on white background: better contrast than the black on grey in Mozilla. Linux build 2000.02.23.08" 0 "Netscape 4.x has the ability to load javascript source out of unsigned JARs by using the syntax: ----------------------------------------------------- ----------------------------------------------------- It would be nice if this would still work in Mozilla." 1 "Build 2000022408, Win95 Go to http://client/international. This is a page with 2 frames, the International Client Product Marketing intranet page. Choose Edit Page from File Menu The frames keep flashing, sometimes with text in them. Sometimes just the background color. In 4.x I actually choose edit frame but this option doesnt seem available in Seamonkey. Guess this is 2 bugs 1) the flashing 2) will we have edit frames" 1 I know of some German MUAs which prefix a replied-to subject with AW: (for AntWort) instead of Re:. Maybe there are similar things in other languages, and maybe it would be worthwhile to make threading by subject work for these just like it does for Re:. 1 "With there being so many things now that cant be seen unless you are logged in, if you have access to confidential items, what things you are allowed to change, etc, bug lists and bugs themselves should have a reminder at the top that you are not logged in right now, so your preferences and access privileges are not being taken into account when generating this list/viewing this bug or something to that effect." 1 "If you enter: file:///c|/temp in Netscape the type of documents are also listed. Something like: cp47_to_472ex.jar 7636 Kb Fri Feb 25 11:14:46 2000 Java Archive logebog.doc 962 Kb Tue Feb 10 18:32:14 1998 WINWORD File Please add this to Mozilla too!" 0 "so mirc has this feature where the task bar icon blinks when someone is talking about me in IRC. it would be cool if chatzilla had a similar feature." 1 "Im glad to see the application icon implemented in the Win32 version! It looks much better. In my opinion, I believe the icon should, however, follow the same exact (non-transparent) look as the M that lies in the top right corner of every browser window." 0 "Since netscape has been established as the leading porn browser (because of its handy close all feature, which is now called quit), it should also be set up to easily disallow a porn site from opening lots of new windows, each of which has javascript to open more windows (on open and/or on close events), etc. (Note: if that didnt sound like a legitamate request because of the porn, note that a buggy or malicious site might actually be _impossible to exit_ because it kept re-opening itself with copies, or might get eat up all of the computers memory and bandwidth by having each window open two new copies) Suggested solution: give each site an allowance of two pop-up windows. If a site tries to exceed this limit, or if a window that was created by javascript attempts to create a new window, bring up a choice box open; always open; dont open; disable javascript. Im sure this isnt the best solution, so please suggest others and/or options for this one. (The last option is in case a site loops until the user allows the window to open.) URL coming. Sorry in advanced for being too lazy find or make a non-porn site. Dont click on it if your browser doesnt like opening and closing lots of windows rapidly, if you dont like looking at porn, or if your parents wont take I was helping to alpha-test a cool open source browser as an excuse for having porn pop up on your screen where your little brother might see it. http://lust-lake.porncity.net/gimmea404error" 0 "Saw this suggestion on n.p.m.wishlist and thought it was a good idea.If the user drags across multiple table cells, mozilla hould behave like a spreadsheet and select the cells in a square with one corner at the cell under which the mouse button was fist pressed and the opposite corner at the cell under the pointer. This could be very useful for drag-and-drop/cut-and-paste to spreadsheets or the composer. Heres the relevant part of the original newsgroup post: From: Marc Newsgroups: netscape.public.mozilla.wishlist Subject: Ability to select columns Date: Wed, 23 Feb 2000 11:56:29 -0800 Lines: 21 Message-ID: <38B43B6D.7E9923F0@uvic.ca> It would be excellent if there was the ability to select from an X,Y coordinate to another instead of selecting the entire rows. This would make cutting and pasting columns of tables much easier, i.e. from a web page to another web page, spread sheet, or document." 1 "Request for enhancement: at the bottom of the mozilla window, display an icon indicating whether the page is valid html or not. I imagine small icons without bright colors, with a number (html version) and a green check mark for valid html and a red X for invalid html. When displaying a blank page, the area would be blank. When displaying an image, mozilla could leave the area blank or put a gif, png, etc. icon in the same location. Double-clicking the icon should explain why incorrect html is incorrect, and always give users the option of turning off the feature (along with instructions on where to find the option in preferences if they want to turn it back on). This feature would help mozilla testers figure out why pages are displaying differently in different browsers (is the html bogus, or is one of the browsers messing up?) If on by default, it would also encourage webpage builders to write html that displays as intended on all compliant browsers." 1 "When you change a bug and hit commit, process_bug.cgi appends information on the next bug in your list. This confuses me about every other time I add a comment on a bug, because I think I added a comment to the wrong bug. Suggested resolution: dont append anything. (as opposed to appending the bug you just modified, since the user can click back to bug .....)" 1 "I think it would be a good idea that votes for UNCONFIRMED bugs would not count into the quota of 5 bugs. This votes would be automatically deleted when bug is confirmed (maybe only when over quota?). I have voted for bugs that I consider very important for me and I cant help confirming new bugs :(" 1 "If you do local browsing hidden files are not shown. This should not be the case if the user has the Show all files option in Explorer to on. The Show all files option: Start Windows Explorer View -> Folder Options -> View -> Hidden Files (Show all files) This will set this reg.db setting: [HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced] Hidden=dword:00000001 More info: http://msdn.microsoft.com/library/psdk/bestprac/Topics/RULE40.HTM Expected: If the Show all files is on Hidden files should be shown!" 1 "Quit is too easy to mistake for close this window, and users might accidentally lose some work (page-finding) by hitting quit when they mean to close just one window." 1 "builds 2000-02-25. 0. Run on Mac or Linux. 1. Go to http://jimbob/trigger2.html 2. Select any of the WinProfile or WinReg scripts: a_winprofile_writestring, a_winprofile_getstring, a_winreg_createkey, ... 3. Trigger. OK. 4. Check logfile. Result: Install FAILS with -229 error (script error). Expected: Script aborts with error indicating that its not supported on these platforms. i.e. -202 (access denied), -214 (does not exist), or create a new error for not supported ..." 0 use nsIUserInfo to pre-populate the users fullname in the new mail account wizard. eventually, nsIUserInfo will be extended to pre-populate more fields in the new account wizard. note, the implementations of nsIUserInfo are not checked in yet. 1 "Greetings -- the Debian project requires each binary to have a man page viewable with man -- I dont believe mozilla has a man page yet. So, I thought I would offer to write a man page that netscape could distribute, and save the Debian project the trouble of supplying a useless manpage. So, I figured I would check with someone to see if a manpage was in the works, authored but not part of the Debian stuff yet, etc. (I have heard POD is very easy to use, easier than nroff any ways. Would POD work well for the mozilla project?) Thanks. :)" 1 "You want to have text you read in a proportional font but that makes some ASCII-art and tables hard to read if the writer assumed a fixed width font. I really like the way Eudora has solved this, with a button on the message window switching between fixed width font and proportional width font for that message only. That would be a great feature to have in Mozilla." 0 "When loading M13 on x86 platform/win2k, for first time: Profile creation window pops up. I want to see where it wants to put my profile. No provision for this. Browse lets you choose a NEW destination, but where is the default? If its just a subdir of mozilla, then Id leave it. But for all the end user knows it could be in c:\temp\fredsprofile I think this should be added to make it more usable. thanks ben" 0 HTML 4.0 specifies, that the source of a quote can be inserted as cite attribute. We could use that to show exactly, which quote comes from which msg, if we add e.g. cite=news://38BC14AC.AB10C963@bucksch.org to the blockquote tag (dunno, which scheme to use for mail). This would be similar to the References header, but with much finer granularity. With this information, we could implement a context menu item for the quote, which opens the original msg for exactly that quote. 0 "the file names for the single signon files are poorly named. for example, in my ~/.mozilla/sspitzer directory, I have: 51853843.p 51853843.u 51858695.k egads, can that be made more human readable, so people will know what it is?" 1 "per Dans comments in bug 26497. This only seems to be an issue on Linux. On Mac and Windows, even if the folder is read-only, XPInstall still downloads files to it. This is how the Mac & Win OS behave as well, so were consistent. Were also consistent with Linux, but install doesnt occur (w/ -225 error). 1. Make destination folder non-writable. e.g. on Linux, chmod 444 theFolder. (for this test case, create a folder called a_addtestfile and make it read-only). 2. Execute any xpi to it. Go to http://jimbob/trigger2.html. (a_addtestfile). Trigger. OK. 3. Check directory contents for install. (a_addtestfile) Result: On NT, Win98 & Mac, it lets you install, but not on Linux. Get -225 error. Cant list files in the directory. Suggestion: The only idea I can think of is install to a temporary directory. Dont rename installed files." 0 "I have posted this on n.p.m.ui and I found many people who agreed. I cant find the right component for this though (or should it be moved to the Architecture component?) With the current implementation of splashes we have the following problem: How should people who only download finished builds include customized splash screens??? How can you easily customize splashes (perhaps with skins) without rebuilding Mozilla? Perhaps an image library should be the first thing to be loaded and then display the splash as some image file. I personally think it should be a png file (splash.png?) and (if its possible) the executable file should be optimized so that it launches the libraries for png display first, then load the splash file, then continue loading..." 1 "basically, the Netscape installer currently has Navigator, Mail&News, SpellChecker, etc... Marketing has requested that there also be a Composer component. This would require that the Composer code be pulled out of having the Navigator (standalone component) require Composer. I dont know what (if any) dependencies are currently in place, but I just wanted it to be though of. Packages-win would need to be updated to deliver the new composer set of files too. Theres already a bug filed against the installer to deal with installing this new .xpi file (http://bugzilla.mozilla.org/show_bug.cgi?id=19139)." 0 "MSIE 4.0 and up with the QuickSearch add-on (comes with Browser Enhancements option for MSIE5 I believe) has a feature that I absolutely cant live with out. Basically, what it lets you do is add custom searching capabilities to the Address bar. What I do is assign a 2-3 letter combination for a search. For example, I did goo for the Google.com search. Then you enter the URL with %s where you want the keywords to be plugged in. For Google, I have: http://www.google.com/search?q=%s So when I type the following in the address bar: goo Mozilla Project It will load this URL: http://www.google.com/search?q=Mozilla Project (itll also convert the space to %20 too) This is extremely useful because all I have to do is figure out how a search engine organizes its search string and I can add anything to the Address bar for easy searching. For the W3C Validator, I have this with a w3 assigned to it: http://validator.w3.org/check?uri=%s&ss= And all I have to do is add a w3 in front of an URL to have it be validated. For Dictionary.com, I have: http://www.dictionary.com/cgi-bin/dict.pl?term=%s And then all I need to do is type: dc A_WORD_I_DO_NOT_KNOW in the Address bar to find a word quickly. This feature should be extremely easy to add, so Im crossing my fingures that youll add it. At least the ability to have this feature and add more searches by editing a text file if it would be too time-consuming to add a dialog box for easily adding new entries. Please consider 1 "Ctrl-O currently opens a file dialog, but I think it should open the url dialog. Netscape 4.x, linux: ctrl-L only to open new url dialog Netscape 4.x, windows: ctrl-O only IE (?): ctrl-L only (not sure about this) IE 5: ctrl-L or ctrl-O Mozilla: ctrl-L only; ctrl-O opens file dialog instead." 1 "it seams a sensible idea that the mozilla editor should be able to a wisiwig production of all the markup langages mozilla supports. not just a partial implementation of html..." 1 "The view source window has the text right up against the left edge without a pixel to spare. Looks weird. A little margin (maybe 10px or something) would fix it." 0 "Having a seperate Inbox tree (Inbox, Sent, Unsent, etc.) for each account is going to get really hairy for users with for than about 5 accounts. It would be nice if Mozilla could just use one Inbox tree for all accounts. Using individual Inboxes is not necessary to track which From: header to use on replies, just compare the To: or CC: header to known addresses in the account manager. If it exists, use that address as the From: and if it does not exist, use the default account." 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Macintosh; N; PPC; en-US) Mozilla/m13 BuildID: 2000022414 Suggestion: toolbar favorites should be intergrated in Mozilla, as in 4.x Reproducible: Always Steps to Reproduce: 1. Launch Mozilla 2. Look under the url bar 3. Notice that there is no toolbar favorites Actual Results: No toolbar favorites! Expected Results: An existence of toolbar favorites" 0 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Macintosh; N; PPC; en-US) Mozilla/m13 BuildID: 2000022414 Suggestion: Let the user decide the search engine they want to use within the sidebar rather than in the prefrences. Reproducible: Always Steps to Reproduce: Actual Results: Expected Results:" 0 "Incremental text searches in Mozilla would be (IMHO) a great usability enhancement for advanced users. What I have in mind, is Emacs-like isearch functionality that searches through a page as the user types in the search text character by character." 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Macintosh; N; PPC; en-US) Mozilla/m13 BuildID: 2000022414 Suggestion: Allow the option to copy image by control clicking on an image. Also, allow one to simply hold down the mouse button on the page to bring up the contextual menu rather than pressing control-click. Reproducible: Always Steps to Reproduce: Actual Results: Expected Results:" 1 "How do I enter bugs about localized versions?" 0 "Members of a group should be able to restrict the viewing of a bug to one or more of the groups they are members of, when posting the bug. Currently you have to view the bug report after you submit it before you can do this." 1 "It would be cool to be able to hide a given product, or a version of a product, so that a user would need to be logged in and have access to a certain group before they could even see that the product existed. A good way to implement this would be to add a groupset column to the versions table, and if there are no versions of a given product that the user would have access to according to that groupset, then the entire product is not even shown. Also, this groupset would be used for the default viewability of the bug itself (see bug 30187). Im playing around with this myself already, if I get it working Ill post the diffs here. If anyone has comments on this Id appreciate them though." 0 "Were running Bugzilla for our internal use and when we upgraded (we have to) MySQL to 3.23.x, Bugzilla started to print Opened: Y-m-d H:m at bug detail form instead of Opened: 2000-12-12 12:23. Its because function date_format is changed in 3.23, you have to use %Y-%m-%d instead. I dont know if its possible to fix this (like fetching MySQL version at start and send Y or %Y depending on its value), I just wanted to let you know." 1 "Two things about the sidebar. 1) When I click on customize, the form comes up messed up. On the lefthand side, the Add these panels to your sidebar window has pre-selected bookmarks, and the last one the list (at least initially) Sports and Entertainment is actually outside the white box and on the grey area; the scroll bar in the center appears to work and the list moves up and down; strangely, the words Sports and Entertainment remain on the gary as you scroll, revealing more items, including the real Sports and Entertainment; 2) the buttons along the bottom are not active, except for cancel, which means I cant do anything. And what I would most like to do is REMOVE the sidebar, because its annoying." 1 "Compare this: ------------- This bug depends on bug 29658, which changed state: What |Old Value |New Value ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED with this: nobody@mozilla.org changed: What |Old Value |New Value ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED Its easy to not be able to distinguish the two, and I often dont in a quick glance. One way you could distinguish them, and this would be useful in its own right, is a description for the dependent/ee. I currently have no idea what bug it refers to. The new system provides scope for doing this more easily. ---- This bug depends on bug 29658 [modify pkgcp.pl to merge all .xpt files into one file], which changed state:" 1 "Using build 2000030313 under Win98. Prepending the name of the proxy in the configuration with http:// will stops mozilla from accessing any site. So: www.proxy.com works http://www.proxy.com does not. Restoring plain name will make mozilla happy again. It seems to work just fine under linux." 1 "IE5 allows table rows and table cells to explicitly be appended to the end of the tablesection c.q. tablerow by passing a index of (-1). Although this is not in the Dom1/2 spec, its a good idea imho." 1 "(Note aside: I would say that creating a TABLE element should create a TBODY child, but I think Im wrong in that) When creating a table from the DOM interfaces, the creation of a THeader or TFooter should trigger the creation of a TBODY element, unless one is already present.(This is the current behaviour of insertRow() using M14 or IE5). IE5 will create a tbody at the moment that I add a header or footer through the Dom1. I think this is good." 0 "The Intellimouse Explorer (I scratched the label off of mine) has a forward and a backwards button that the user can access using their thumb. Microsoft built in compatibility so that it controls backwards and forwards in the Internet Explorer browser. I would like this same functionality in mozilla." 0 "I would really like to see my address book on the sidebar, like it is in Outlook express. This makes starting e-mail much easier. It would not take up space, since it minimizes to a tiny line." 0 "While I like the dependency tree I can get by going to the parent bug, its missing a lot of stuff thats necessary for bug management. Id like to be able to get a normal query result page for all the bugs dependent on the named parent bug. It would be OK if this were a link on the parent bug next to the existing link for the dependency tree. This enhancement would take the place of the current procedure of: go to parent bug, scroll down to dependency list, copy bug list from text edit field to clipboard, go to query page, scroll down to only bugs field, paste list into the text field, submit query. The other alternative is ugly, painful pseudo-keywords in the summary so you can query on the summary line. Setting this up is difficult and error prone since you must visit every bug and summaries sometimes get changed by others." 1 "Suppose I have a group of bugs to be marked confidential. How do I query to see if the whole group has been set properly? In addition, the change multiple bugs page doesnt offer this choice. It would definitely be easier to manage these things if the multiple bugs page offered it. I think its also useful to be able to query based on the confidentiality setting." 1 Communicator and currently Mozilla place a simple So and so wrote: above quoted text in replies and, I believe, forwarded messages. There are various circumstances in which it would be nice to have more information, most crucially dates. There are advantages and disadvantages to actual text-string customization; just a few checkboxes might be a better choice; there was no consensus in the newsgroup (n.p.m.ui) discussion on the subject. I am pasting below the pertinent parts of the newsgroup thread: -------- Matthew Thomas wrote: > Dan Ridley wrote: > > In fact, there are several elements of quoting that Id like to > > see improve: Id like to be able to customize the So and so > > wrote: text (for instance, So and so wrote, 2000-Mar-01 or > > So and so wrote, in thread Using ideas from Tog). > > Yes, thats one feature that users of other clients have > enjoyed, but I dont actually think its a good idea to > include it. There are two reasons for this. The first is > that presenting information other than the sender (and > perhaps the date) is usually a waste of time and readability, > because other such information is available from the message > headers and surrounding messages anyway (except in that rare > event where a person sends a reply to someone other than the > original sender or group). Is this actually a rare event? I do it quite frequently (think executive summary), but I may be unusual in this. I am also specifically thinking here of quoting multiple 1 "Request for enhancement: a customizable Page menu, with each item pointing to a an incomplete Form-GET url that gets completed with the URL of the current page. The defaults might include netscapes whats related, netscapes gist- in-time translator (moved from the View menu), and netscapes websitegarage html validator. It would be nice if this menu could be hierarchical, but Im not sure if people would want to arrange the menu by site (google, netscape, etc) or by what happens (html validator, translator, related pages found, etc.)" 1 "maybe it is the default of GTK, but I would like the checkboxs background color to be white (like the rest of form elements). With the present grey color they remind me of the former Motif checkboxes in checked state.." 1 "In your preferences, there should be 3 ways to customize message headers displayed. 1. Ignore the following 2. Display the following 3. Display standard only (To, From, Date, Subject)" 0 "Most programs can export to text files. Make it so that the address book import can pull up the fields of a tab or comma dillimeated text file and then allow you to till it which number field stands for what type of field (first name, last name, email, number, Pager, ect.)" 1 "AOL client support. (Your Netscape... AOL wont block you out!!) ICQ client support. (They own ICQ too!!!!!!) Jabber client support. Open source XML Messenger client. http://www.jabber.org/ MOZILLA 1st Multi-Service open source chat client!" 1 "My Bugs only shows assigned bugs. Need another button to show SUBMITTED BUGS ...so one can see what the status of the bugs he has submitted." 1 "M14 does not have any scrollbars, horizontal or vertical, when the preferences window is too small to show all the information on the right panel." 1 "RFE: An API for storing/fetching addressbook entries, e.g. an addressbook provider..." 1 "I find myself wanting to add myself to the CC: list for a bug an awful lot. Most of the time its either because I want to track the progress of the bug or because I added a comment and forgot to add myself to the CC: list as I did it. I dont want to spam everyone on the CC: list as I add myself to it, especially in the latter case where I already spammed everyone. If BugZilla didnt send out email notifications when only the CC: list changed (or when only the user added themselves to that list) I would never have to spam anyone just to track a bug. I know BugZilla knows when only one field has changed, because it now sends me email with only changed fields in it. What I dont know is whether anyone has any use for knowing when someone has been added to the CC: list. If there is a legitimate use for that information this RFE shouldnt be implemented. Otherwise it should. In either case, fixing bug 7345 would probably take care of this simpler issue, and fixing bug 7710 would take care of half of it." 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; Win95; en-US) Mozilla/m13 BuildID: 2000022820 I have a Logitech cordless wheel mouse. When I use it in other programs it works. IN Mozzila MP13 it doesnt. Reproducible: Always Steps to Reproduce: The Mouse wheel just NEVER works Actual Results: Not Applicable Expected Results: N/A" 0 "The text below was posted in the Communicator newsgroup. I believe we need a Go button on the Mozilla UI..... There are no easy ways around the described problem in Netscape 4.x so the user is best off using IE 5.0 at the moment. Mozilla should cater to the challenged..... ------------------------------------------------------------------------- Greetings, I am writing for a friend who is a quadriplegic. He is able to browse the web using a motion sensitive baseball cap (to move the cursor) and a straw (to simulate clicking the mouse) through the application Joymouse 4.1. The only real trouble he has is that if he is on a certain webpage and wants to view a higher level (for example, lets say he is at and he wants to view the higher level without the my_dog part,) he is able to move the cursor (with the baseball cap) up to the website window, click the mouse after the word pictures and then delete the words /my_dog. The next step to view the higher level (i.e., the ) would be to hit the Enter Key on the keyboard -- which, of course, he cant do because he is a quad. At this point he has to call someone into his room to hit the Enter key for him. Is there any way to add an Enter Key on the Netscape window so that he can hit Enter himself? Thanks for any help you can give. Please direct all replies to StAlfnzo@cruzio.com" 0 "I searched and searched and couldnt find a bug for this, so here it is. When Im managing bookmarks, either in Manage Bookmarks or the Bookmarks sidebar panel, I can drag-and-drop bookmarks all over the place, which is great. But the limitation with this is with dragging a bookmark to a closed folder. Expected Results: After a short delay (1 second?) while holding the bookmark over the folder, the folder expands and I can then drag the bookmark to wherever I want it to be. Actual Resules: The only choice that I have is to drop the bookmark on the closed folder, and the bookmark is made the last item in the folder. Build:2000030516" 1 "Mitchell wrote; > Dan: might there be a way to mine bugzilla for this sort of data: find > people who had contributed x patches, etc? If so, can you file a bug for > statistics 2.0? maybe itll stats x.0, but even so ..." 1 "(This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #85126 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=85126 Imported into Bugzilla on 03/07/00 12:04) I think this is an RFE, but its a really important one. I need the ability to encrypt a msg to someone with a cert which may or may not match the From line. Right communicator automatically picks the cert based on the From value. Since alot of people these days have mutiple email addresses and some even get modified as they pass through firewalls, having this one-to-one mapping is silly, not to mention unusable. I think there are two solutions for this. 1) autoselect the cert based on from line but have a select cert button in the security tab of the composer that allows the user to select a different cert. 2) allow the user to define a many-to-one mapping such that mutiple email addresses map to the same cert. ------- Additional Comments From repka 04/28/98 17:41 ------- Slightly modifying summary line; adding jsw to cc list, changing platform to All because the behavior is XP, and reassigning this to me, at least for the time being. Brian, I think that each place you say the From line you really meant the To line. Correct? Jeff, Id like to talk to you sometime about this requested behavior. Im thinking that the solution is really for the cert itself to be clear about what email addresses it is valid for -- as this is required for decent behavior on the receiving end 0 Im forever getting messages from people who use Outlook and put everything in a 1mm yellow font on a purple-and-pink image background. Having something in the View menu that said View as Plaintext and ran the message through the plaintext converter (or, if the message is already multipart/alternative, just take the text part) would totally make my day, and would be a huge incentive to use mozilla mail over 4.x. 0 "we need a nsIEventQueue accessor on the channels (specifically the transports)" 0 "Unix users could really use a pref for dont ever show me a font smaller than a given size (px, pt or mm; pt or mm is probably better than px). There are a lot of pages and mail messages that are just unreadable on linux. I heard a rumour that theres already such a pref for Mac; it should be xp (even on Windows, Id use that pref, but its on Linux that I really need it)." 1 "I am using a logitech cordless wheel mouse and the wheel works in all other programs except mozilla MP 14" 1 "A FTP listing should provide a parent directory entry like NS4 does (if there is one). Currently the only method to go up is manipulating the URI in the location bar." 1 "I cant get it managed to build Mozilla with dmake (Suns distributed make - a make which can run on multiple (!!) machines) due various errors. 1st problem is: -- snip -- % dmake dmake: Fatal error in reader: ./config/autoconf.mk, line 162: Unexpected end of line seen -- snip -- It would be nice to get this fixed because we have 14 Ultra*-machines which can be clustered to build the Zilla..." 1 "It would be of benefit if when one initiates the saving of a file, it be independent of the browser, such that if the browser decides that it no longer wishes to function from some variation of a segment fault (bus error, or what have you), the download will continue unabated. Keeping track of failed downloads and querying whether or not the user wishes to continue these downloads would also be good. A download component as such ..." 0 "Some suggestions for the midair collision warning: Since it states right in the submit anyway text that added comments will not be overridden, the midair collision warning should not happen if the other thing the previous person to edit the bug did was add a comment. Since it displays the diff right on the page, its pretty obvious that was the only thing done, and the collision detection should be smart enough to figure that out and not give the warning. Conversly, if the CURRENT edit is only adding a comment, this should also automatically happen, and not issue the warning. (on the other hand, it could be a duplicate post from the same browser, which is probably also detectable, and the warning should occur in this case). Also, instead of only Submit anyway and erase the other persons changes (except comments) and throw out my changes and revisit the bug, it should also have a choice to add my comment anyway, but ignore any other changes I made." 1 "would it be possible to adjust the html generated by the boolean charts and and or buttons are formated as several rows rather than one. example table will be attached." 0 "Latest CCK build dated: 03-07-00 O/S: WinNT4.0 Are the following keys needed?: 1. DefaultName = Mozilla 1. Organization = Netscape 2. Department = CCK _NewConfigName = (Is this key needed since the CustomizationList = is set to the same value? Also, need to remove the _ from the beginning of this key to be consistent with the others that are listed.)" 1 "This bug is a spin-off from bug 6585. If we cannot find a glyph for the character that we are trying to draw, it may be nice to download a new font for the user, possibly automatically." 1 "From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0) BuildID: 20000022820 the html 4.0 standard recommends to show the content within the attribute title of a a-tag as a tool tip. this isnt included in mozilla 5. Reproducible: Always Steps to Reproduce: 1. load the site 2. hold the mouse over a link 3. nothing will appear ;-) Actual Results: nothing Expected Results: a tool tip click here" 0 "I would like to see the default context menus that come up when you right click a link on a w/page to come up when i right click a bookmark in the sidebar. Therefore I would be able to copy the location, open it in a new window, etc." 0 "I would like to be able to rearrange and delete bookmarks [like in the bookmarks window] from inside the sidebar. --Can this be done by simply embedding a special version of the bookmarks window inside the bookmarks panel? --This would probably work best after drag and drop is implemented in bookmarks, therefore you wouldnt have to add anything to context menus or add an edit menu to the bookmarks panel. --Good idea?" 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; Win98; en-US) Mozilla/m13 BuildID: 2000022820 In order for the No Proxy For support to handle a mask for my internal hostnames, I must specify the domain name like this: .mydomain.com This works the same way in Netscape 4.72. In IE 4.0+, I must specify the domain name prefixed with an asterisk, like this: *.mydomain.com Should this syntax also be added to Mozilla in addition to the current syntax? This is more of an enhancement than a bug report. Reproducible: Always Steps to Reproduce: 1. Specify the domain name for internal hosts starting with an asterisk, like this: *.mydomain.com 2. 3. Actual Results: Internal host names cant be reolved by DNS properly. Expected Results: Internal host names be properly resolved and the appropriate hosts contacted. Again, this is more of an ehancement request than a bug report. M14 looks REALLY good! Excellent work!!!! Peace......" 1 "Will Mozilla for Linux eventually do font anti-aliasing? The party line from the X people is that it is impossible to implement at the X protocol level without a complete rewrite (breaking backwards compatibility in the process), and is something that has to be done by individual X clients. The problem is that no application that I know of does it. Anti-aliases fonts would make Mozilla MUCH prettier. The lack of anti-aliased fonts is the main reason that the Windows and Mac interfaces are so much cleaner looking that any Linux desktop." 1 "Hello there. :) Mozilla handles a click on the navigator button (underneath the status bar/progress meter, next to the envelope button) in the same manner as Navigator 4.x -- and I think it is wrong. When a user has two or more windows open, clicking on the navigator button does nothing (mozilla) or switches to a different window (navigator). I would much prefer it to run the File|New Navigator Window.. command. Perhaps the behavior should be prefable? :) Thanks" 0 "I use mutt as my email client and it has this cool feature where it can automatically arrange related emails into thread-o-logically correct order. To do this it requires the In-Reply_to header to be correctly set, which bugzilla doesnt do. The format of the header is: In-Reply-To: ; From email@blah.com on What would be good if it would set the In-Reply-To header to point to the previous comment for that bug so that they will be displayed as a thread." 0 "This would also affect other platforms which support some form of toolbar to switch tasks, so I have set OS/Platform to All. Currently in Navigator 4 I can have many windows open, so many that the buttons on the taskbar are too small to read (The have been known to get down to just the ships wheel icon). My request is that all windows in mozilla are shown by just one button on the taskbar. Clicking on this will show the most recently used window, and windows can then be switched using the Open Windows or Tasks menus. I would suggest adding a tick option to either both these menus, or the View menu reading something like Show all open windows on taskbar." 1 "This is a test bug. Ill mark it as a duplicate of the real bug report later. http://www.slashdot.org www.slashdot.org www.slashdot.org http://www.slashdot.org Bug #10000 Bug 10000 Bug #10000, bug 10000" 1 "Only an enhancement request XML Data Islands are a Microsoft IE5 feature. I wouldnt expect you to provide all MS extensions. However, this is a pretty useful one for thin clients which use http to transfer data as XML. The crux of it is that wed get a consistent way to get at the XML embedded into documents. This is a really exciting example of its use in IE5. I presume youll have no trouble finding details of MSs documented element." 0 "The current situation: If you have more than one bug group (isbuggroup=1), it gets alphabetized in the list, and theyre all mixed up with the access groups (isbuggroup=0), making them hard to tell apart without closely reading the descriptions. My proposed enhancement: I think it would be a little cleaner-looking in editusers.cgi when viewing the user if the bug groups were listed separately from the access groups in the list, perhaps with an
between them. This gets all the bug groups grouped together, so you can easily distinguish them from the access groups." 1 i try out most nightly builds, the profile manager does not display my old profiles, why not? i delete mozregistry.dat and the /bin/ folder, why couldn t i suppose a directory where an old profile is stored? 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; NT4.0; en-US) Mozilla/m13 BuildID: 2000022820 feature request: when u save a web page, it saves the url in the form of a comment in the first line other issues: Ctrl-H doesnt bring up url history cant get windows context (right click menu) when in the address bar Reproducible: Always Steps to Reproduce: 1.try saving a web page 2.type ctrl-h 3.try right clicking in address bar Expected Results: it doesnt do/support it currently" 1 "I think we should have a stop button in Composer.... there are some times when I right-click on a link, meaning to open in a new window, but open in composer by mistake. Ive heard this complaint from enough other people, I dont think Im the only moron whos done it. :) It would really be nice to have a stop button in composer to stop loading whatever page its loading." 0 "On Win32 2000030516 daily build: Actual Results: When max-width: is used, min-width: is ignored on tabs with little reason to be the size of max-width. Every tab renders the same size. Expected Results: XUL tabs should have a minimum and maximum size. If a tab can be the width of min-width, it should just like it is if no width style is used." 1 "I think someone needs to design a drop-down menu for the back button (it may need to be a separate small button next to it) similar to that of Netscape 4.X and MSIE 4.X-5.0. From my limited knowledge of JavaScript, this seems possible (to me). It would be a great feature for either websites that use long trails of links or for websites that redirect you to another site in 0 seconds, destroying the functionality of a back button. I dont know any XUL, so I wont be able to do this myself yet. Also, when Mozilla becomes Netscape 5.0, many end-users may become disgruntled that a feature they like in Netscape 4.X is downgraded." 0 "It is very convenient in IE5 to change the font-size from either a menu or from the toolbar. I would really like to be able to easily change my font-size on a site by site basis like I can in IE5. Also, there should be a preference to automatically override font-size px specifications, example being britannica.com. I hate reading there tiny italics font in articles, and no browser will let me override their sizes." 1 "With netscape under unix/linux clicking a link with the middle button would open the link in a new window. This was incredibly handly. Under windows with the intellimouse explorer, clicking the middle button enters the mouse into scroll mode, where you move the mouse and the window scrolls. Clicking the middle button again turns scroll mode off. Holding down the middle button and moving the mouse does scroll mode too, but scroll mode is turned off on mouseup. This is also incredibly handy. In mozilla under windows, clicking the middle mouse seems to do the standard unix paste, but it pastes it as a new URL even when Ive got the cursor over the text area! This is currently an issue in bug #29295 Heres what Id like to see: -if middle mouse button clicked over a link, open it in a new window -if middle mouse button clicked over an editable text area, paste from the clipboard as per normal unix behaviour (maybe a pref option here for unix/windows behaviour?) -if middle mouse button clicked over anywhere else, let the mouse driver take over and turn on scrolling (as per netscape 4.7 behaviour) This would make browsing much nicer, and I dont think it should be all that nasty to do. Thanks." 1 "I filed this as an enhancement mainly because 4.x does this too, so I assume that its an unwanted feature. Clicking the Component Bar at the lower right opens a new Navigator window only if one Mozilla window exists, and if more than one exists, it will switch to another Mozilla window. Id like to have it open a new Navigator window regardless of open windows." 1 "Layers doesnt seem to be implemented yet, URL provided is a page with a lot of linked layers and javascript combined. This file might be useful when developing and testing layer functionality. The page in question is usually a generated page from a database in an intranet, which results in inaccessible links in the provided URL. Im not completely sure that all the HTML on the page is 100% correct, but it works in Netscape 4.72. It doesnt work correctly in Internet Explorer, and that is a known bug (or feature if you like)." 1 "what about an about:demo or about:features? Think advertising - for a storefront - show off all the Mozilla coolness here for the few and new who want to see what Mozilla is capable of." 1 "It would be nice to support the ``data URI scheme, defined in RFC2397." 0 "If the extension of a webpage document isnt given in the URL, or if the extension in the URL doesnt match one of the recognized extensions for the content, mozilla should add an extension to make the saved file be the correct type to ensure that nothing gets lost. URL DEFAULT FILE TYPE http://foo.bar/ index.html Webpage (*.html, *.htm) http://foo.bar/index.html index.html Webpage (*.html, *.htm) http://foo.bar/index.htm index.htm Webpage (*.html, *.htm) http://foo.bar/cgi/cgi.exe cgi.exe.html Webpage (*.html, *.htm) http://foo.bar/cgi/cgi.cgi cgi.cgi.html Webpage (*.html, *.htm) http://foo.bar/cgi/image.cgi image.cgi.gif GIF Image (*.gif) depending on http://foo.bar/cgi/image.cgi image.cgi.jpg JPG Image (*.jpg) the image For each of those, there would also be an all files (*.*) option. Im not sure how much of this is specific to Win32. This should probably only apply to http and https." 0 "Now that Mozilla supports crypto in regular builds it would be nice to have a link off one of the menus to take you directly to the PSM code. This would make things easier for people installing/reinstalling the PSM code. My first suggestion is that the link be under the tasks menu. Then the debug menu have a link the the smoketests page. Currently the page is http://docs.iplanet.com/docs/manuals/psm/psm-mozilla/index.html It would be even nicer if we could get iPlanet to make a simple (redirected url) like psm.iplanet.com or www.iplanet.com/mozillapsm/ something that would always be the same. That might be safer then the long URL we have now. I know this is not the place, but saying here might help in big picture... It would be nice to have a link off a menu next to the one above to where ever Sun hides the OJI Java needed for mozilla. This again would be helpful to us all. (when the Java Plugin and Mozilla finally get along.) Assigning to mcafee@netscape.com as a start as suggested." 1 "The page should contain table cells in Webdings font but doesnt display that font though it is installed on the client. So I make a request for enhancement to improve Mozilla to display that font as it contains a couple of useful symbols which using css and dtml can replace a couple of image swapping effects. I couldnt find a font related component so I filed on layout being the most generic term. Please change that as appropriate." 1 "Show extra markup and Hide extra markup arent really enough. The editor also needs a way to edit the HTML directly. If bug #28809 (view source at selection) ever gets implimented, the editor would be able to go to the correct place in the HTML source to allow you to edit a specific part manually." 0 "The URLs on http://bugzilla.mozilla.org/describekeywords.cgi should be links because Im too lazy to select, copy, click, paste." 1 "It would be nice to have a -print commandline argument that would cause Mozilla to load a page and print it, then exit. It could function like this: mozilla -print [printername] http://some.lame.site/page.html The printername, when omitted, would default to lp. This flag would come in handy for printing HTML that is not well-handled by html2ps, such as tables." 1 "BuildID: 2000.03.13.03 When attempting to save a file, and returning to browse the web in the Mozilla window, the browser stops responding and the download halts. Window overwrites (like tooltips) are persistent over the Mozilla window and the d/l dialog. Reproducible: Always Steps to Reproduce: 1. begin downloading a file 2. browse the web in mozilla Actual Results: Browser locks up Expected Results: Save the file in a child process such that this could not happen (even if the process is zombied) or under a download manager that would allow continuing files implicitly and remember which ones failed." 1 "Im leaning a bit widely out, but heres my proposal: Use newsgroups for creation and storage of comments. Set up a news server f r bugzilla and you can give each bug its own newsgroup. Then use any open source web/new gateway to show the articles in the bug page like today (just that they are hierarchical now). You can use mozilla.orgs mail/news-gateway for the cc functionality (with the cc field in the bug page as frontend). Advantages: - Bugzilla commments are only linear and dont scale. Some bugs are very controversive and a lot of discussion is going on, in which case it is hard to track several subparts of discussions. - I sometimes menually quote, when I reply to bugzilla comments, and then wish to have the comfort of a mailnews-client. - When I write long comments (and get distracked and surf during writing), I fear, that Navigator or Mozilla could crash, and - open a new mail - copy and paste the comment to it - save the mail as draft This would be just a Save as draft or even an autosave with Mailnews. - I could use HTML for comment formatting, if I need to. - I believe, its the right thing. NNTP is the most widely used system for internet-based discussion and it evolved a lot because of that, much more than bugzilla propably can. The above mentioned advantages are just examples for that. Disadvantages: - Other fields: Its hard to edit the owner, state etc. with a mailnews client. You cant make the field only changable via 1 "The default Netscape middle-click behavior for links (namely, open this link in new window) seems to be missing in Mozilla and I sorely miss it. I dont know if this is a chrome or code issue, but I would LOVE to see it fixed. :-)" 1 "It would be useful to consider other ways to make roaming more easily possible in a LAN/WAN environment. Currently 4.x on Win32 is rather unfriendly to this, as you cant merely throw the user profiles on a network drive and pick up and go to another computer and easily get them, since nsreg.dat must contain the profile data, and it cant be moved from the systems local Windows directory. There are workarounds for this, but theyre rather clumsy and can be a deterrent on large deployments. I fear that Mozilla is going down the same path, given what appears to be stored in mozregistry.dat. The goal would be to create a better way to allow network roaming, perhaps possibly storing mozregistry.dat, or at least its profile location data, within the user profile directories. This request is similar, but not identical, to bug 9556, though the fix may be the same. This bug is branched off from bug 17048, where some prior discussion occurred." 0 "Connecting to an SSL server where user certs are requested but not required when the user has no client certs installed, PSM puts up a modal dialog box No Acceptable Identification. This dialog box annoys the user for no purpose and discourages sites from configuring their servers to accept client certs, since configuring their servers that way will annoy their large installed base of non-cert-holding users. If the server sends a fatal alert indicating a certificate is required, then such a dialog box would be useful. Otherwise, the server is perfectly capable of giving the user a reasonable error message, if such is warranted." 1 "Add a Bug created Date and a Bug Created in last X days attribute/choice to the boolean table section at the bottom. Please? :-)" 0 "Rename Messages pref pane to Sending or Composing and create a new pane, Display. Reorder pref. Currently, we have a message pref pane, which keeps prefs for displaying (incoming), sending msgs and quoting/forwarding. Additionally, the top-level mailnews pref pane also keeps prefs for displaying, font families and quoting formatting, which should be moved to Messages, if we had enough space, but we dont, so move them to the new Display pane together with the other display-related prefs in Messages. So, the new structure of the pref panes would be: - Top-level Mailnews - The unnamed box with Play sound etc. - Messenger Start Page - Display (notice, that all prefs are related to plain text) - For incoming messages (find new name) - Font, holds current Display plain text messages and articles with - Composition - Forwarding and replying to messages - Formatting, holds current For outgoing messages (Attention: Name clash with pref pane Formatting. Rename the latter?) Fixing should be easy (I can do that, if you give me enough time), but what do you think?" 1 "Subpart of bug #17048. Create the main code, that is invoked at startup and shutdown and does all non-protocol-related work. No UI, however." 1 "Subpart of bug #17048. Create UI for roaming access, mostly preferences. Possibily also login, ask the roaming access protocol owner(s) about it." 1 "Subpart of bug #17048. Implement the client side of the ACAP protocol for roaming access." 1 "Overview Description: Enhancement: New State for configuration manager(CM) When developer is working on a bug it is in the Assigned state. After making the necessary coding changes he/she the source in to source control. At this point the bug is not really Resolved because the testers cannot test it until the CM creates the official build from the source control database. It would be very useful if there were another state such as ready for build to reflect this situation. In this case the developer would mark the bug ready for build and after the build, the CM would mark it Resolved For backwards compatability, a better approach might be for the developer to continue marking bugs Resolved and the CM mark the Ready for Testing after the build. The testers would only look at bugs marked ready for testing Actual Results: current states do not include CM" 0 "Currently, the installer makes no attempt to upgrade an existing install. You can install to a new directory, or you can delete the old install and then re-install with the new version. We need to support the ability to upgrade. During an upgrade, the existing installed components need to be considered or the result may not be a working product. The main example would this: component A exists on the machine component B is being installed both A and B depend on component C In this case, if component A isnt installed at the same time as B & C, it probably gets broken. Information needs to be collected about whats already installed, and then this needs to feed into new dependency rules during the install. Cathleen, please do not mark confidential. I no longer depend on this internally." 0 "If an SSL connection is using encryption weaker than 90 bits, the lock icon should be displayed in the insecure (open) state." 0 Spin-off from bug #31047. See longer description there. Goal: 1. Recognize individual quotes (i.e. also quotes within quotes) in non-flowed plain text, so we *can* make them look like HTML quotes (with a bar instead of > s). The latter should be controlled by a pref. 2. Clean code up to make look configurable by stylesheets. Preferably, 2. also works for the recipient of quoted plain text msgs in HTML msgs, i.e. I always see quotes of plain text msgs the way I want, even if another Mozilla Mailnews user replied to them via HTML and sent the result to me. I have this already mostly working in my tree. I spent 1-2 hours browsing my mail archive and the recognition works fine. Maybe Ill attach some screenshots. --- What Im not sure about is, what we should recognize as quote tags. Currently, I recognize x spaces, directly followed by y upper case chars, directly followed by > or ], optionally directly followed by a single space. That is , because some mailers use something freaky as quote tags, e.g. VM (EMacs): >>>>> BB = Ben Bucksch wrote: BB> bla blabla BB> blabla I do recognize that, but 1. it is rare 2. it is not completely correctly recognzed in this case. The xxx wrote: line is recognited as a nested quote of level 5, i.e. would get 5 bars. 3. it increases the likelyness of failure. I did not see such a case in practice, but should it happen, it would be bad, because I remove the plain text quote tags in favor of the bar. 4.x seems to only recognize > , which 1 "For every HTML element that has an attribute title a tooltip displaying that title should appear when the mouse is over the element. This is especially useful for and . Note that bug 19425 demands this for and bug 27828 demands this for " 1 This is sort of an international usage issue. Currently Mozilla uses [author] wrote: when replying to messages. I propose it to be changed into [author]: Currently localized strings are exposed internationally i.e. Karl Ove Hufthammer kirjutas:/Irve skreiv:/etc. On newsgroups, for example, it can be quite annoying. By omitting wrote we achieve localized versions that no longer produce this kind of confusion. Jaagup Irve: is internationally understood regardless of the language pack used. 0 "We are frequently asked by frustrated users to include a README file with our binaires that explains how to start mozilla and gives minimum requirements. I think if the file explained how to start mozilla and had a link to the release notes, that would be adequate. For milestone builds that people contribute, it would be nice if the build system could also append build environment info to the readme listing what options they built with and if its not difficult, maybe which versions of system libraries they linked against. The buildID should also be listed here. Build creators should also be able to add their own info to the readme." 1 "Itd be nice to have a pref that would put a bound on infinite so animated gifs would stop automatically after the specified time. At least on Unix, older versions of the browser used a lot of CPU cycles for animations even when the window was covered or minimized." 0 "What needs to be in a distribution for either 1) running mozilla 2) developing components for mozilla 3) developing mozilla 1) should only contain those files, that need to be running mozilla 2) should be some kind of SDK, and contain all the files one needs to develop components and plugins. There should be some integration into a build system as well. Should there be rules.mk, autoconf.mk or autoconf.mk.in in lib? How about something like mozilla-config, cloning gnome-config? Then one could probably add most of the important stuff, and still let people choose compilers, optimization, debug on their own. 3) is of course the complete source. This is here just to distinguish headers like widget/public/nsIDragSessionGTK.h, which are currently in dist/incude, but only for internal use. This has been pretty unix-centric so far, what is needed for other platforms? Axel" 0 "There is no UI for enabling or disabling individual SSL ciphers. This is primarily useful for disabling ciphers that are too weak for the users security policy." 0 "The User Identification Request dialog box for selecting the client cert to send to a server should have a remember this decision checkbox." 0 "In the account manager, there is a single checkbox Use secure connection (SSL). Since SMTP uses STARTTLS to implement SSL, this should instead be three radio buttons, one each for Never, If possible, and Always. The current checkbox is ineffective anyway. It should be removed from the UI until it is implemented correctly." 1 "The name of the Cancel button on the User Identification Request dialog box should be changed to something like Dont identify yourself" 1 "Implimention of the publish command used in 4.x to save having to fire up 4.x or my FTP client just to upload a page to the server" 1 "The drop-down menus on the Back and Forward buttons (with most recently visited sites in each direction) in Netscape are very useful; it would be nice to see this functionality in Mozilla" 0 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; WinNT4.0; en-US; m14) BuildID: 2000031317 like IE5 mozilla shows icons in the location bar. Why not have this be a user defined icon like is possible in ie5. and maybe also make the location bar a list like object where you can choose previous locations Reproducible: Always Steps to Reproduce: 1. go to url 2. 3." 0 "This is not a bug, it is a feature request. I am posting it to remind everyone that this is one of the most important parts of DOM Level 2 that has not yet been implemented. Without this feature, many DHTML applications are going to have problems. DOM Level 2 specifies the method ViewCSS::getComputedStyle for determining what CSS styles are applying to an element. Currently in Mozilla, this style info can be determined via el.style, but only if the styles were set inline (e.g.
) or via Javascript (e.g. el.style.blah = blah). IE5 implements this functionality using the non-standard el.runtimeStyle... not a standard, but at least we can use it. Without this, developers are encouraged to use the
syntax for all CSS, rather than taking advantage of the cascading part of CSS. As an example, this means that DHTML app developers will be unable to write an XUL- like widget with Javascript by reading in style values from a big CSS file, then laying out a component according to its computed box model. Youve seen the benefits of this in XUL, so imagine how this would benefit developers who want to do the same with HTML/Javascript. I realize everyone is swamped with stuff to do, I just wanted to remind you of this feature so you can prioritize it when deciding what things you can and cannot finish in time for feature freeze. IMHO, this one is important. Read about getComputedStyle in the latest DOM WD: http://www.w3.org/TR/1999/CR-DO 1 "In testing Mozilla i often remove the entire mozilla directory and reinstall from the begining. The probelm is one has to re download the PSM from iplanet each time you want crypto. If we could have the PSM in a different place or be able to download the PSM installer to keep it local it would be very helpful in testing." 1 "Pressing escape while editing the URL in the location bar should cause the URL to revert to the URL of the page displayed. Currently a user must hit reload to figure out the URL of the current page. If a new page is being loaded, but hasnt started rendering, while escape is pressed (but the location bar still has focus), the first escape should stop the loading of the new page without reverting the location bar to the previous page. Pressing escape a second time should revert the URL. See also bug #29085 (escape key should stop page loading) - this bug might depend on 29085." 1 Now that bug 17074 is fixed, it might be worthwhile to think about a pref Default sort order. Before, it was annoying me that I had to click on the Date column when I first open Mailnews to get my emails in Newest-First order every time I start Mailnews. Now it annoys me that I have to set the sort order on all my 3 dozen mail and news folders. 1 "the option should make the browser accept cookies but delete them after the browser is shut down. additionally if this option is checked by the user, an additional panel could be activated where the user can list domains where permanent storage of cookies would be allowed. IMO this could lead to a broader acceptance of cookies which are a useful tool in designing web sites. [ when i let my imagination run wild this may even become the default setting for Mozilla .. :-) ]" 1 "Currently, both http and mailto links have copy link location on their context menus. For mailto, I think this should be replaced by: copy mailto link and copy e-mail address" 1 "When one has multilple plugins for the same mime type it is often difficult to configure which plugin runs. C(One has to play around with the dates etc...). It would be much easier if the about:plugins was in a drag&drop movable order, or if there where boxes where one could list 1,2,3,4, etc... and then submit it and the plugins would be in the order specificed. These are just two suggestions but there are many ways of handling this with a NICE user interface." 1 "It was there in M14, but the list is a lot shorter in version 2000031805." 1 "In situations where a vendor has a kiosk using the Mozilla browser, it would be nice to be able to disable JS features which change the state of the mozilla window. For example, it would be nice to disable open() so that consoles and console-based ads would not work as they generally tend to annoy/confuse/etc users, especially in situations where the user is not expected to be savvy. Disabling JavaScript is an acceptable workaround for the time being, however more granularity would be a wonderful addition." 0 "Id like to suggest adding an html4 keyword, to signify items that are required for compliance with the W3C html 4.0 specification." 0 "in 4.x the JS function document.getSelection() returned the currently selected text in a window. While this was a Netscape-only feature many websites make use of it. This does not currently work in Mozilla. This would be great thing to have for Mozilla. If this will not be implemented, the resolution of this bug should be a note added the developer release notes. It should include how this feature could be accomplished in a different way (using DOM or XPCom maybe?) Feel free to change the component/foward to other owner if JS Engine is the wrong one." 0 "Build: 2000031908, MacOS 8.6 The directory structure of a Mozilla installation is too `flat -- there are not enough subfolders, and too many files at the top level in each directory. Not only is this daunting (impolite, even) to those users who venture into their Mozilla folder, but it also makes life harder for those who want to skin Mozilla, because files are not organized nicely into subdirectories. An example or three. 1. The contents of the `Mozilla Folder for a Mac installation looks like this: Mozilla Mozilla AddressBook Mozilla Editor Mozilla Installer Mozilla Message Compose Mozilla Messenger Mozilla Preferences Mozilla Profile Manager Mozilla Profile Wizard Mozilla Select Profile chrome/ Components/ defaults/ Essential Files/ Plugins/ res/ Component Registry NSPR Logging Ideally, it should look like this instead: Mozilla Address Book Composer Mozilla Setup Compose New Message Messenger Preferences Profile Manager Profile Wizard Select Profile Essential Files/ All the other files should be in various subdirectories in the `Essential Files folder. 2. The chrome/ folder contains 22 items at the top level. This could be simplified by, for example, putting the messengercompose/ folder inside the messenger/ folder, bookmarks/ inside the navigator/ folder, layout/ inside the global/ folder, and so on. 3. The chrome/globa 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; WinNT4.0; en-US; m14) BuildID: 2000031908 In IE5, if you open a new browser window, the new window is a duplicate of the current one--i.e. it points to the same place and has the same history. In Opera, the new window function opens a new window at the home page, but there is a Duplicate Window option which allows opening a duplicate. Id prefer the Opera method, but one way or the other, it would be a handy thing for Mozilla to have." 1 "ALT text should display when moving mouse over images. This is probably a duplicate, but putting in a new bug so doesnt get lost in the shuffle. -Pn" 0 "From Bugzilla Helper: User-Agent: Mozilla/4.7 [en]C-AOLNSCP (Win98; U) BuildID: 2000031420 It would be handy to have the History of URLs Ive visited available to me as a panel in the Sidebar. Currently, it seems that the only way to view my History is from Tasks>Tools>History. Clicking on the Back button doesnt do it either (the way it works in N4.x) This buries a feature Ive come to depend on. Reproducible: Always Steps to Reproduce: This is a new feature request." 1 "There should be a way to go back to the most recent page in the last domain you were at. Maybe by double clicking the back button, or pressing shift left arrow as apposed to alt left arrow." 1 "Before the last build, you should speed up the code as much as possible. For instance, for the code: for (i==0;i<1000;i++) { a[i]=b[i]; } You should use memcopy instead. You should use hash tables when necessary instead of constantly sorting and searching. You can make a hash table template container class for easy inclusion in code. Use template container classes such as the ones in STL. You should make the code able to utilize multiple processors. You should use the quick sort with the ability to recognize the worst case scenario. Make sure you always pass large objects by reference unless impossible. Sometimes, a tree is better than a hash table for storing and sorting a large number of words. The root would contain a child for each letter that exists and each of those children contains a child for each letter that exists until you make the words in the document. They contain the location in the document where that word exists. If you need part of a word then you go down the tree for that part and then collect all children. etc... I think that when you finish the last build, you should release a beta version for a month for final testing. That way your first release can be 5.0 and it can work perfectly. :)" 1 "In communicator, starting at page A, go back and then load a new page, go back and then forward and you lost A. I propose a solution to that problem. Store all the pages visited in a tree. Dont have a history like IE5.0 . Instead, if you do the same thing, and held down on forward, you would see: new page (whatever its name is) page after that etc... ------------------ A other pages that you went to from this page The old pages such as A and any other that you went to and pressed back would not contain in the forward button any pages you went to from those. In the history, you would see a tree (Just like a directory tree) showing you the pages you went to and where such as: | page ||_A ||_some other page || |_fjjdls || |_fjdkls ||_new | | You could also be able to view the history by domain sorted by Name, Date, Type of file, etc... But pressing the forward button would do the same thing. The history should be dockable." 1 "Sometimes, people fill a directory with files such as: Clock1.jpg Clock2.jpg Clock3.jpg Clock4.jpg Hand_01_a.jpg Hand_02_a.jpg Hand_03_a.jpg Mozilla should have a smart browse feature (on a dockable toolbar of special features that lets you increment or decrement the number on a file to go to the next one. If a file is: Clock009.jpg: First it should try Clock010.jpg, then Clock10.jpg If a file has two numbers or more, then for the first file you press the button for on the file group it should ask which number you want to increment. It should also try incrementing such things as: Clock010a.jpg to Clock010b.jpg" 1 "Im not sure the component for mail. You should be able to put in mail filters that use script, just like Outlook. It should also have many of the features of outlook such as a calendar. You said you dont want to do a calendar, and just want to stick with a browser. Why not compete with outlook along with IE5.0? They are both built on the same theme. Whats wrong with making your product have more features? Isnt that the whole idea of the mozilla project? To have more features and less bugs? If you do a calendar make sure you have: Time slots for scheduling, along with a general field for tasks that dont have a specific time each day. Along with tasks, there should be a simple text box for notes for each day. You enter tasks from the main calendar screen by double clicking on a line and then entering info and alarm etc. much like outlook. Then it inserts that task on the day. You can click on a task and change its day. You can view all notes sorted by day at the same time in one text box, and double click on a note to edit its contents and date. Tues, Feb. 1, 1999: fjdfkds;lfadfj;lajdkf;lsjfkdls;fds Wed, Feb. 2, 1999: jfdk;ldafjak;lfjdal;fjfl I think you can make your mail client better than outlook. If not, then why include it?" 1 "Talback always crashed for Communicator when Communicator crashed (probably due to low memory cause of leaks in the software - but Im not sure). Anyway, make sure talkback is not tied into the Mozilla task so that doesnt happen again. Also, make sure Mozilla closes when you close all the windows. In Communicator, it didnt. I am going to download the Milestone to test it." 1 "The dark areaas should have a gradient fill to give a 3D appearance. The browsers original skin is too dark and grey. It needs some life and color." 1 "I love the animation in the upper right corner of the screen. I wish it ran all the time." 1 "Make sure you add the ability for skins to have animated buttons and other animated. The default skin shouldnt have animations though." 0 "Search results in my toolbar also overflowed the box. Please refer to my last bug. Also, you should be able to add your own engines. If you can already, make it easier to figure out." 1 "I know you havent made the menu shortcuts yet. F3 should be to find again because that is the usual in all programs." 1 "I know you want Mozilla to be fast, but files in the cache dir should really be 128 bit encoded. The reason is that children could have access to adult content that was viewed earlier by a parent. Maybe there is a way that you could make doing so fast. Cookie files should be encoded and the titles should not contain the name of the site because they might have objectionable material in them." 1 "you should be able to zoom in and out of a page. I have known many times this would come in handy." 1 "When the enter user name for a form dialog box comes up, you should be able to remove certain user names. Also, if you hit enter in a form textbox, it should submit if there is a submit button." 1 "I was going to suggest that. But I noticed you already have it :) There are some inconsistancies with tabbing in fields. I suggest having tab work the same and shift+tab tab inside a form. You can add that shortcut to the menu bar so people know." 1 "I think the browser should have optional smily face support just like AIM. That would be cute :)" 1 "I think there should be an option of whether you want the text cursor visible when you click on a page. It is helpful for knowing where the search will start from." 1 "Ive noticed on slashdot and the mozillazine forums that everyone asks how to get rid of the sidebar. Even though theres a menu item in View, no one seems to be able to find it! Rather than getting frustrated at stupid users, there should be some easy way for people to turn the sidebar off. Options include: * Ask if they want a sidebar when they set up a profile * A small text message in the sidebar itself describing how to close it * A big red X somewhere in the sidebar that does the equivalent of View | Sidebar." 1 "Pls reassign if component is incorrect. We should have a PRINT icon on the Navigation Toolbar as we have in other versions of the browser." 1 "I enjoy boberb@rpi.edus bugs a great deal. Please have him file more." 1 "You should be able to also select Beta, or Release1, Release2, etc. in the target milestone because it is getting near the end and some enhancements will just have to wait until the second Release." 0 "The old netscape plugin API allowed overloaded javascript functions for plugins. IDL does not support function overloads. MS works around the overload issue in its IDL implementation thru the use of the VARIANT datatype. XPIDL does not allow methods which use native types (for example, nsXPTCMiniVariant) to be scriptable. A potential work-around for this might be to support var args in the mozilla IDL implementation (and in the javascript engine if its not already there...). Or make an exception to the XPIDL native type rule for nsXPTCVariant. If a work-around is not available, legacy web content which calls into any of a plugins overloaded javascript methods will not work in mozilla." 0 "Build: 2000031908, MacOS 8.6 To reproduce: * Go to any Web page which has a multi-line chunk of text. * Triple-click in the middle of a line of text. What would happen in a perfect world: * The block (paragraph, DIV, list item, etc) gets selected. What is probably supposed to happen at the moment: * The current line (as determined by word wrapping with the current window size) gets selected. What actually happens: * The current line gets selected, except for the first character in the line." 1 "Ever since the link outlines were introduced into the mozilla browser project (whereby, on clicking on a link - it is outlined with a dotted line), they have not functioned properly (as indicated by existing filed bugs). The original Netscape Browser works beautifully without outlining links. I would have thought that such an addition interferes with w3c standardisation. With all the work put into achieving a consistent interface for Mozilla on any platform (i.e. making work easy for designers) I would have thought there would have been a drive to put power into the hands of the designer. But to enforce a law of outlining links undoes the usefulness of CSS control over visited-links... and many other dynamic HTML possibilities. Surely the designer is crippled? There was a day recently when I downloaded the latest build of mozilla and it ran without link outlines... this was a beautiful day." 1 "At first IDLoc should be implemented, then perhaps tumbler and later other types. We should be able to use a lot of the XPath code in the XSLT module." 1 "Go to a website with frames. You should be able to right click in a frame, pick Print Frame from the menu and print only the contents of a specific frame, instead of the whole web site. (This feature exists in IE)" 1 "We want to be able to verify signed scripts created by the Sun JDK jarsigner tool. Two things stand in the way: our libjar cant read zip files created by the Sun tool (bug 20512), and PSM signature verification fails." 1 "There is a RFE for identification of current profile (b23260) that requires this feature, we also need this to be able to do things like say Bookmarks for Ben Goodger etc. This is, in essence, a pretty name for the profile. If the back end for this is supplied, I can hook up a front end for it via the Profile Wizard." 1 WARNING: Serious re-design proposal ahead: Currently, it appears that any mail downloaded is placed immediately into the users inbox mailbox while it is being downloaded. I propose that instead, we place all newly downloaded mail into a special mail folder. It should have the following properties: 1) There should ALWAYS be one of these folders for each existing e-mail account. It should be created automatically if it does not exist for some reason. 2) The user should not be allowed to delete this folder. 3) The user should not be allowed to move mail into this folder, only out of this folder. 4) Any mail remaining in this folder should be moved into the inbox whenever the user asks to download new messages. There are several benefits of changing to this design: 1) When we impliment message filtering, automatic transfer into folders, were going to have to have some way of determining the set of messages that have just been downloaded. Having a special folder would make this much easier. 2) It makes it much easier for the user to see what messages have been newly downloaded. 3) It makes downloading all messages from the pop server an atomic event, as far as the real message folders are determined. If all messages arent downloaded properly, then the client wont tell the pop server to delete the messages. If the client crashes in the middle of getting messages, then you wont get duplicate messages in your inbox. Disadvantages: The only one I can think of is performanc 0 "Currently, we have: NS_IMETHODIMP nsLocalFile::GetDiskSpaceAvailable(PRInt64 *aDiskSpaceAvailable) { NS_ENSURE_ARG_POINTER(aDiskSpaceAvailable); return NS_ERROR_NOT_IMPLEMENTED; } That would tend to explain our crappy error handling around out of disk space conditions. :P Is this something that no one ever got around to doing? (as in, could I just do an obvious implimentation?) Or is this something that is sadistically difficult to do in a cross-platform manner?" 1 "Reproduce: 1. Make bug A depend on bug B 2. Resolve bug B as duplicate of bug C Actual result: Bug A is updated with the blocking bug B marked resolved. Expected result: Bug A is updated with blocking bug B removed and blocking bug C added." 0 The Mail/News Account settings item is not available in Message compose windows, but that is where you wind up after some errors, such as missing SMTP server, that tell you to change the account settings. It would be easier to find what you need if it was in the active windows menus. 0 "The cookie manager should allow you to automatically add a site to the block list when you remove the cookie. Maybe this should be the default behaviour. The reason is that in most cases, people are going to go through the list of cookies. For the cookies in the list that they want to remove, they arent likely to want to have the cookie re-appear later." 1 "I dont know what this command is called (I dont know Java) but I know its part of Java. Its the command that moves the browser instance to the top of the stack of instances. And by me, thats the single most annoying thing a browser can do. (Hotmail is a good example; it pops its browser to the top *every* time you follow a link.) At present I can turn off Java entirely, but I dont want to do that, just turn off this one command." 1 "Its just some comments about sidebar, pages, sidebar bookmark, etc. 1) I think general people dont uses too much bookmark. 2) A lot of people is going to use sidebar as bookmark. But at same time its not intended right now. Its intended to have a particular kind of pages specially designed for the sidebar. Just like another micro web. Its interesting but I am not sure how people will use. Lets see. 3) Open Page Bookmark - a suggested idea. Since I liked the sidebar style because of the bookmark effect an idea is: A kind of open bookmark. Its should be like the sidebar tabs, but in fact is a version of the thing for full size pages. So people can navigate through their pages using the tabs. Add new tab, put the URL in the TAB directly. When openning a new browser window, maybe should open a new tab. Not sure - just some comments about tabs etc. Comments appreciated. More comments: It right now, remembered to me the idea of AOL 5.0. Because each web window its inside the AOL browser window. With each browser window like a collection of tabs inside the main browser window should provide an environment like AOL browser." 1 Replying to emails does not quote the entire message (i.e. the header with the time and date and from: to: fields). There is, in addition, no option to modify this, which might solve the problem (like a quote header option). 1 "I didnt think this belonged under the umbrella bug for keyboard shortcuts in the Browser component, so here it is: Interested in seeing 4.x-style keyboard shortcuts in mailnews, for accessibility and ease-of-use. Here are the ones I use often: spacebar to next unread message if at end N/P to next/previous message n/p to next/previous unread message +/- to expand/collapse threads m to mark read/unread tab to switch panes (did this move to another key?)" 0 "(PC/Win98; didnt check other platforms yet) The size (in KB) reported for newsgroup articles in the Size column is frequently much too large. For instance, a five-line Usenet article with no attachments showed up as 489KB. Some articles report realistic sizes (i.e. 1-3KB), others come in anywhere from 50-700KB. The error is not proportional to the apparent real size of the article; I couldnt discern any obvious pattern from the UI. I reproduced this on two semi-randomly selected groups: rec.autos.makers.vw.watercooled and alt.tv.law-and-order." 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; N; Linux 2.2.12-20 i686; en-US) Mozilla/m13 BuildID: 2000022916 (The User-Agent is wrong. Im using Mozilla/m14) If the element of an html document contains an unclosed tag, the rest of the html document fails to display. On the one hand, an HTML document with an unclosed <title> doesnt conform to the spec, and so mozilla shouldnt have to display it correctly. On the other hand it would be robust behavior to assume the <title> tag closes when the </head> tag is seen. Reproducible: Always Steps to Reproduce: 1. Create a document with <title> tag in the header, but no matching tag. 2. Load the document in mozilla M14 on linux Actual Results: No page content is displayed. The debug output to the console gives the following feedback: FindShortcut: in=http://www.math.arizona.edu/~jsac/test-title-bug.html out=null failed to set the page title. Document http://www.math.arizona.edu/~jsac/test-title-bug.html loaded successfully Document: Done (0.499 secs) Expected Results: Assumed the tag when it saw and displayed the contents of the page normally. The following html demonstrates the problem succinctly: </head><body> <h1>You should not see this.</h1> </body> </html>" 1 "Request to implement the editres protocol under X11/Unix This protocol is required to implement high-level control of Mozilla from other applications like input devices for disabled people (speech recognition)..." 1 "User/Passord dialogs could offer more information about: - to which domain will the password actually be sent ? - will the password be sent over a secure connection ? this would improve the overall security of web applications. As SSL may do strong encryption the weak point in web sites seems to me where the user is guided from the non-secure area of the site to the secure (SSL protected) area. when it comes to sensitive data, as login information, the user should be shure the browser sends it to the right domain and over SSL." 1 "The Mozilla RDF/XML parser does not support the aboutEachAttr rule [6.8]. I believe the correct way to support this is to add a aboutEachAttr derivation datasource that can be added to filter queries. It will wrap another datasource, and will look for a magic aboutEachAttr meta-property (which the parser will need to create) whenever a property lookup is performed. Building this support directly into the parser will not work: although it would correctly parse RDF/XML, it would not properly deal with post-parse updates; e.g., via JS. Furthermore, any other datasource that is aggregated with the datasource containing the aboutEach would not correctly reflect the semantics of the containing element." 0 "We currently do not support <RDF:li parseType=Resource|Literal> in rule [6.30]. Supporting parseType=Resource should be fairly straight-forward; it is simply a shorthand syntax for adding properties to an anonymous, inline resource. parseType=Literal is a bit more difficult because it allows arbitrary XML to be embedded in the body: wed want to treat this as a string, not as a content model. The parser may have issues with that." 0 "When confirming bugs I often change component. It would be nice if we could reassign to component owner at the same time. What is needed is a confirm checkbox below the Reassign bug to owner of selected component radio button, exacly like the checkbox below the reassign but to [input] radio button." 0 Id like to assign a priority to received msgs following certain criteria. That way, I can predefine the order, in which I read msgs. This could be implemented like the following: 1. Add a msg attribute Rating or Priority of type integer. (This is /not/ the priority, that th sender sets.) Of course, the attribute also has to be added to the tree view in the thread pane. 2. Support this attribute in filters. At least let me increase or decrease priority, maybe additionally give the option to set the value absolutely. Additional Comments: It will propably be sufficient to add the attribute to the local database (and not store it in the msg header on an IMAP server) in order to avoid to modify msgs. But this means, this feature only helps rating new msgs (INBOX or news), not the archive. 1 "From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.0; Windows 98) BuildID: 2000032405 mediocre: at the Garden of NOX (see URL Field), the tables containing the News Headers (1..3) contain <td> fields that display properly in IE4/5 and Netscape 4.7, but will not display properly using the width=1% specification if the <td> contains more than one image... instead it wraps the images. Is there any workaround? Reproducible: Always Steps to Reproduce: 1.Load the page :? 2. 3. <td width=1%> <img src=images/layout/scroll/rightend.gif width=9 height=30 align=absmiddle> <img src=images/layout/scroll/rightcap.gif width=11 height=30 align=absmiddle> </td>" 0 "From Bugzilla Helper: User-Agent: Mozilla/4.51 [en] (Win95; I) BuildID: 2000022820 about:mozilla is one of the most needed enhancements to mozilla, because its just so darn cool. As of M14, though, it appears not to exist, so I cant impress my friends with my easter egg skills anymore. Reproducible: Always Steps to Reproduce: 1. Type about:mozilla in the location bar. 2. Watch as nothing happens. 3. Go get a coffee and reflect over the seriousness of such a problem. Actual Results: It did not display the customary about:mozilla page (..roiling cloud of vengeance.., etc.) Expected Results: It should have loaded a small page containing this code: <HEAD> <TITLE>The Book of Mozilla, 12:10








And the beast shall come forth surrounded by a roiling cloud of vengeance. The house of the unbelievers shall be razed and they shall be scorched to the earth. Their tags shall blink until the end of days.

        from The Book of Mozilla, 12:10" 1 "Allow text files to be displayed with URLs as hyperlinks (that probably means run them through mozTXTToHTMLConv). The user should be able to toggle this option easily, perhaps in a way similar to toggling extra markup in the editor." 1 "Give the user an option to open an open all links from a webpage in a new window dialog. The dialog should display the number of URLs in the document and then allow the user to choose how the windows should be opened (any combination of (minimized, back, front) and (normal, maximized))." 1 "What about an about dialog in the installer. Normally in other installers its possible to click on the icon just left of the Setup in the upper left corner and choose About. This dialog could honor the XP install engine..." 0 "With bug #24514 (which is strongly rlated), Dean Tessman found nsLookAndFeel::GetMetric (in mozilla/widget/src//nsLookAndFeel.cpp). This is a nice function - all of these values are candidates for similar bugs. (At least, its concentrated...) It would be nice, if there they would be sat in XML files (XUL or XBL; dont know, whats appropriate)." 0 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; WinNT4.0; en-US; m14) BuildID: 2000032409 The sidebar should have a close button (e.g. an X) next to the existing customize... button. Clicking on this would have the same affect as the View|Sidebar menu option but would be more obvious. Reproducible: Always Steps to Reproduce: none needed" 0 "THis is an RFE spun off from bug 33012. Im adding the original reporter to the cc; line so he can track this RFE. Current Behavior: On par with NN4.x and IE4, if you have a multiline text box that is extending below the bottom of the visible browser content and you type below the visible area the browser does not automatically scroll the content pane to keep your typing visible. Requested Behavior: Typing in a multiline text box which extends below the visible page would cause the browser to scroll down sufficiently to see the text you are typing. Just in case this is vague here are simple steps to see this behavior. 1. Scroll up to the comment box in this bug report 2. Scroll to a place where only the top half of the comment box is visible on the page. 3. Start typing at the top of the comment box. 4. Type several lines until the text you are typing is off the screen 5. Note that the main browser window does not scroll to keep your text visible." 0 "When a close box is clicked on a window, the first thing we should do is disable the ability for javascript to open new windows. If the user is closing, they usually want to get out of the damn thing, not see another stupid window. I think the onUnload handler is the one that needs to be fixed. Somebody sent me the idiotic url above. If you go to the page and then click the go back button a pop-up window appears (that says Wait!). If you close that window, another one appears. Annoying." 0 "The Cookie Manager is great. One additional feature Id like to see though is the ability to directly add/delete/edit cookie entries in the Cookie Manager. Say I install Mozilla on a new PC. I know I want to block doubleclick.net. Id like to be able to go into the cookie manager and manually add it as a blocked cookie site, rather than having to keep browsing until I hit a doubleclick.net cookie and reject it then. Another useful feature would be to be able to export the cookie manager black list to a file and then be able to import it into a different machines copy of Mozilla." 0 "Sinners repent! Give us a preference to open or NOT open a new window when we do File, Open new file. Opening a new window uses windows resources and is annoying to some users." 1 "The user should have the option of allowing Mozilla to remember the master password. On a single user maachine, such an option would be convenient so that one need not type the password every time one wishes to check email." 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 [fr-FR] (Windows_98; I) BuildID: 2000012008 When I see pages with TAHOMA or others fonts under WIN 98, The Mozilla browser display fonts, but not smoothed. Smoothing would be appreciated ! Good luck ! You could also include a second shortcut to relod pages than CTRL+R, many people are used to F5 key. Reproducible: Always Steps to Reproduce: 1.Launch Mozilla 2.View a page 3.The fonts arent smoothed Actual Results: Fonts arent smoothed Expected Results: Fonts would be smoothed" 1 "I was trying to see if a dict: protocol URL would work from a web page, and found out that nsScriptSecurityManager::CheckLoadURI in nsScriptSecurityManager.cpp is preventing me from doing so. Same thing happens with finger URLs, at least with the M14 source. (I noticed that this was updated since then, will have to pull changes!) I was wondering if instead of (most of) this routine, if each nsIProtocolHandler descendant could have a routine in it to answer yes or no to loads, rather than requiring changes to nsScriptSecurityManager.cpp every time a new protocol is added that should be loadable from web pages. (not that I see finger: or dict: urls much in web pages - but the ability would be nice!)" 1 "There should be a minimize button on the save file dialog button to allow the user to minimize it. As it stands now you have this dialog floating there that you cant shove out of the way while downloading a large file." 1 "The save file dialog boxs progress bar should have a border of some kind around it so you can better see the file progress." 0 "As per IRC mention: I have a Mozilla launcher on my GNOME panel which starts Mozilla and puts all the usual messages you get when starting it on the command into a file. It would be wonderful if I could select a url from a terminal and drop it onto the launcher and have Mozilla open on that url. This doesnt seem to happen atm, and you said to put that in here." 1 "I think this is the right component. My RFE is basically this: When you select a file to download, it could start downloading that file to a temporary file (or memory up until a certain size) while you are picking a directory to place it in. After you pick the directory, Mozilla would continue the download in the directory you picked - obviously there would be a short delay. Although this isnt of major importance now, it might be good for the final glazing of the product. Whats great about this is if you forgot the directory, etc. You wouldnt waste any time even if it takes you 3 minutes to find the directory. Also, small files (a few Megs) would already be downloaded by the time you clicked OK." 1 "SPAM web pages used to be just porn sites, but now there are many sites that use evil SPAM techniques. When you go to the site, you dont like it (of course) so you click back. Then, when you are supposed to go somewhere else, you end up back at the site and another window opens having another site in it. When you try to close that window it brings up another, etc. What Mozilla needs is a button to disable a malicous page. 1) Disable javascript, refresh Metas, redirection ability, etc. on that page. 2) Remove all changes the pages made to program information. SPAM is a big problem, and pages have been taking advantage of the vulnerability of browsers to trick users. Now, those pages could be disabled on command." 1 I talked about this in the Mozilla irc chat, and someone talked to newsgroup admins about this. The admins said it would be a good idea. I got permission to post the RFE. Refer to the web page for information about SPAM and 3rd party cancels. There is so much SPAM on the usenet newsgroups. There is one way to fight back, by 3rd party cancelling. That means that you are not the poster and not the admin, but you can still cancel by writing a correct header and sending it to the newsgroup. Mozilla should have a way to automate this. Of course, there is ettiquete in deleting posts. You must have good reason and tell people why. (The reasons cant be personal). Some common reasons are crossposting to too many groups, and off-topic SPAM. Although I am not experienced, it took me hours to delete a few postings (never tried again). A feature like that should cut it down to a minute or so. Of course Mozilla should include some information (IE ettiquite and risks - like being banned) and make them click Yes they have read it before cancelling is allowed. It should tell people the newsgroups that are made for practicing cancelling such as alt.sex.cancel and alt.cancel. It should also point to the newsgroups that give info about cancelling and web pages about it. (Forget the newsgroup names) 0 "Heres an idea I think would be really freakin handy. Every browser since the stone age has had back and forward navigation, but theres another movement thats common, and could be automated...up. Say your at... metalab.unc.edu/foo/bar/baz.html *click up* metalab.unc.edu/foo/bar/ *click up* metalab.unc.edu/foo/ *click up* metalab.unc.edu/ *click up* unc.edu/ Moving one up in the directory tree is a common task, and to: - go the address line - highlight the current file/directory - delete it - hit enter is a pain, when it could easily be automated." 1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; Win98; en-US) Mozilla/m13 BuildID: 2000022820 Because there is no way to choose how many pages back to go. I hit the back button repeatedly to go back 7 pages. it crashed. Reproducible: Didnt try Expected Results: gone back several pages maybe make it right click to a menu that lets you choose which page to go back to." 0 "start up a fresh mozilla go to file:///c:/ open a few directories deep with the arrow, not by clicking on their names read a text file or something in the opened directories click back all state is lost, your back at c:/ with nothing expanded... this is not expected behaviour" 1 "Position on a page is not saved, when going back to it via back the button. This is Bad. On very long pages, this is Very Bad." 1 "We talked about this in our group meeting last week... We should add an accessor to nsIChannel to allow us to set channel argument used in the event sink callbacks. Right now, down deep in the transports, they do something like this: mUserListener->OnStartRequest(this, mUserContext); but if they could do this: mUserListener->OnStartRequest(mUserChannel, mUserContext); it would allow us to avoid a whole layer of nsIStreamListener implementation in the channel implementations. (Look at nsFileChannel for an example.) Note that this change might have be done hand-in-hand with one to change the way load groups are managed since the OnStopRequest often removes the current channel from the load group in addition to passing the OnStopRequest notification upstream." 1 Reproduction, way 1: 1. Open the HTML composer 2. In the content widget, press return and enter From all. 3. Send the msg to yourself 4. View the incoming msg Reproduction, way 2: 1. Send youself a msg with the following HTML file attached: stupid sendmail From all Important: From all must be at the beginning of a line. 2. View the incoming msg Actual result: Youll see >From all Expected result: From all Additional Comments: I know, that is for some stupid UNIX mail programs (which use the mbox msg database format). However, theres no need for >, a space is completely sufficient and invisible, if the HTML is displayed (as long as not in a
         element). BTW: Why is > used in plain text and not space?
        1 "In bug #6085 the middle mouse button was changed so that middle clicking a link in the browser window would open the linked page in a new window. The same thing should happen when you middle-click bookmark in the bookmark pulldown. Akkana should know the details about how to make this happen. (also see bug #33758 for the same report about sidebar)"
        0 "Linux, build 2000032909. When using an http proxy, the statusbar always says contacting , where  is localhost in my case. It would be more helpful for mozilla to tell me what site Im contacting, as I already know the proxy Im going through."
        1 "Perhaps this is intentional behavior but... In IE if you click a link then, once you get to the linked page, you hit the back button, you are right where you were when you hit the link. In all versions of Netscape and Mozilla Ive ever used this feature is missing. Especially on large pages having to manually scroll back to where you were can be a real pain. Can IEs behavior be added at least as an option?"
        1 "In IE the default behavior is smooth scrolling where alhtough the page scrolls 10 pixels or so, the page is drawn in 1 or 2 pxl incraments so that the page glides up rather than just jumping. I find it much easier to keep my place on a page, particularly with lots of text, with this behavior. Netscape/Mozilla scrolls in 10 pxl or so incriments but I dont beleave there is a smooth scrolling option."
        1 "From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows 98) BuildID: 2000033008 I have a ui suggestion that I think is simple but very important. When you click on the url field it should automatically select the current url so that one can immediately start typing and overwrite the existing url. Presently you must click on the field, then move the mouse with a button down to select the field and THEN type the new url. This is very annoying. Both MSIE and Netscape work like I propose. There is no reason not to have it be like this. Reproducible: Always Steps to Reproduce: 1. Click on url field 2. Note that it doesnt select the field text so that typing will overwrite. Instead it inserts. Actual Results: See above... Expected Results: See above.."
        1 "There should be an option to make all application windows and dialogs resizable. This makes it easier to find bugs in XUL layouts that rely on assumptions about fixed sizes. These layouts will probably break with i18n or alternate skins; window resizing is a much easier way to find and diagnose some of these bugs (this was suggested by Hyatt in a thread in mozilla.xpfe some time ago). Furthermore, I (among others) think that every window should be resizable if the user wishes to do so and resizability does not create UI clutter. In Windows, resizability currently requires no additional visual elements. So, apart from its usefulness in testing, I would like this pref to become a real feature. Unfortunately this pref needs to be implemented per-platform, as the current XP widget APIs do not allow eBorderType_default to be combined with eBorderType_resizeh. I have completed an implementation for Windows. It adds the preference nglayout.debug.all_windows_resizable. It seems to work, and reveals a number of interesting bugs that I will get around to filing during/after the skinnability push."
        1 "It would be great if one could create links to folders and bookmarks and the bookmark database, these links could be placed eg. in the personal toolbar folder and represent a small extract from all the bookmarks the user has created, advances: - you don t have to change all bookmarks, because you get no double or triple (or more) entries - you can organize your bookmarks in a logical order and the most often used items can be linked for access, instead of having a part of the bookmarks in the personal toolbar folder (as I have at the moment) an the rest in another folder (or the case above, redundant bookmarks and folders)"
        1 "From Bugzilla Helper: User-Agent: Mozilla/4.72 [en] (X11; I; Linux 2.2.14-15mdk i686) BuildID: 2000032306 I went to my ftp server and logged in. Its great cause IE cant even display ftp contents. Any way I went to upload a file and there wasnt a function on the menubar to do so. Reproducible: Always Steps to Reproduce: 1.Log into ftpserver 2.Try to upload a file 3.Usually File, Upload File <- that doesnt exist Actual Results: There is no way to upload a file."
        0 "... for contact information of the Mozilla community. This information should be used by all other mozilla.org tools; dependencies for specific tools will follow. The server should be accessible world-wide, so I could access it from e.g. Mozilla Mailnews. There should also be an easy way for users to update information about their own (or most of it - not permissions of course)."
        1 "My primary email address for Mozilla work is . If I had a cvs checkin account and checked in, it would should up in bonsai (e.g. blame or log) as mozilla... We should soon have a LDAP server for mozilla.org, which can also store a nickname for users. Use this instead."
        1 "Would be nice, if we could have a myname instead of just myname in the contributor list. mylink (and propably also myname) could be any URI and should be configurable by the contributor without much involvement of mozilla.org. This could be relatively easily achieved by creating field on the soon-to-be created mozilla.org LDAP server, which takes mylink. This field should propably be only used by credits (at least for now), because I might want to present different links depending on the audience (the world (credits) or the mozilla community (e.g. Mozilla-internal webtools)). Maybe call it public_link or so."
        1 "In Netscape, if you follow a link from a long document, then use the back button, you are (often) returned to the same place on the page where you left off (ie, the location of the link). I think this is enviable behavior that saves a lot of time and frustration when working with long documents. This could be implementable by inserting a special target in the cached copy of the page."
        1 "When scrolling with the mousewheel on a page, scrolling annoyingly stops if the mouse happens to come over a textarea. This is unusual behavior; in IE 5.5, scrolling will continue unless the TEXTAREA has enough lines in it that the vertical scrollbar is enabled, at which point the scrolling will then take place only within the TEXTAREA. If the textarea is empty or its text is less than the size of the box (i.e. the vertical scrollbar is not needed to see the rest of the text), scrolling of the page continues as usual. Reproducible: Always Steps to Reproduce: 1.) Place the mouse in a spot in a site where it will pass over a TEXTAREA that is empty or does not have a lot of text in it if scrolling occurs. 2.) Watch the scrolling of the page irritatingly come to a halt. Actual Result: page scrolling stops Expected Result: page scrolling continues On a related side note, if a normal textbox has the focus and you put the mouse over it, then try to scroll down with the mousewheel, scrolling will not take place. This is completely unneccessary since textboxes cannot support multiple lines anyways, so the mousewheel would never need to be used."
        0 "I know by this stage in development, this doesnt have a chance of being changed, but still I think it should be said . . . Am I the only one who sees the barber shop-esque progress bar at the bottom as not only ugly, but useless? Its just like Netscape 4.xs in that it doesnt truly show the progress of the page, but rather just if the page is still loading or not. A person can just look at the animated icon in the toolbar if they want to know that much, or at the actual text in the status bar, or just at the actual page itself. It seems useless to have a progress bar that doesnt truly show the progress. Why not implement one thats more like Internet Explorers, with a progress bar that - when it reaches the end - actually signifies that the page is done? Just my 2 cents."
        1 "The most use of onset and onget is currently for converting properties to attributes, like the checked property on checkboxes:        I think this is going to happen a lot, no, make that *A LOT*. It would very usefull to make this a functionality of XBL, like:  or  Which makes sure setting a property or using setAttribute are equal, and getting a property and getAttribute are equal too. In other words: this will automatically insert the onset and onget values like above. If the inherits attribute works dynamically (which it should IMHO), it would replace _all_ the onset and onget attributes in xulBindings.xml, including the textfield!"
        1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; Win98; en-US; m14) BuildID: 2000032308 The Site X wishes to set a cookie... do you wish to allow it? dialog needs to have a button to show the contents of the cookie. It would make it possible to make decisions to accept or reject cookies based on their content (such as the cookies that dont expire until 2038, or that set personal information about the user into the cookie) BEFORE accepting the cookie, rather than having to go open the cookies.txt file later. Reproducible: Always Steps to Reproduce: 1. Visit any site that sets cookies AND that you havent chosen to disallow cookies from. 2. 3."
        1 "One feature that Netscape Navigator/Communicator has (and IE lacked) was that when you right-clicked on an image in a webpage, the context menu that came up would display the image name next to the View Image menu option. This was a very convenient feature, yet it seems to be missing from Mozilla. Although this may seem like an insignificant feature, I think it was very useful. Plus I dont think it should be too hard to implement? I hope. :)"
        1 "It would be nice to have a way for the Profile Manager to load profiles from a remote server. Possibly even requiring that a remote profile be loaded on startup if a bit is set in a config file. This would be useful in university computer labs to provide tailored settings for each user on startup. Something similar to IMSP."
        1 "I suggest you do as Acrobat reader for windows: let the user srcoll the homepage by grabbing on to it ANY where he likes and just drag around - where there isnt a link or a button. The user could maybe select if he wants to scroll or select/mark text/pics etc. for each mouse button - or maybe a mode for holding down shift to mark text with the mouse instead of scrolling or visa versa..."
        1 >------- Additional Comments From selmer@netscape.com 2000-04-03 13:12 ------- >David, this sounds similar to something Id like. How hard is it to be able to >filter based on whether an address is in one of my address books? Itd be >really handy to be able to say If [Sender [v]] is in [Personal AB [v]] then do >something.
        0 "Dave, could you fill in the blurb & SWAG on this task?"
        0 "There should be a button to add/remove me from the CC list instead of having to type in my e-mail address in over and over again. Red Hats version of bugzilla http://bugzilla.redhat.com/bugzilla/ supports this."
        1 "is it possible to add a pause button to the ftp download dialog? often, when I am at home over a slow link, I start downloading and half way through, want to stop and do something else more important that takes more network bandwidth. I wish I could stop my download, do my other thing (like get all my new mail what ever) and then resume my download."
        0 With a slow server, polling a news server for new headers can take a significant amount of time. I notice it in 4.x and I find it annoying. A pref would be nice. As I see it, this is one of those areas where user preference is important.
        1 "Need to add some exclude-list for servers which do wrong things with http/1.1."
        1 "From Bugzilla Helper: User-Agent: Mozilla/4.7 [en] (X11; I; Linux 2.2.14-15mdk i686; Nav) BuildID: Netscape 6 Beta Bookmarks are used frequently. The bookmark button should be large so that it is easily targettable (recall Fitts Law, for you UI designers). Function before style. Reproducible: Always Steps to Reproduce: N/A Actual Results: N/A Expected Results: N/a N/A"
        1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; N; Win98; en-US; m14) Netscape6/6.0b1 BuildID: 2000033112 Netscape 6 Preview Release 1 does not have a button near the URL text field that produces a URL history drop-down list. This a standard (and MUCH USED) feature of current releases of both Netscape Communicator and Microsoft Internet Explorer! Reproducible: Always Steps to Reproduce: [Missing function -- cant reproduce] Actual Results: No result. Expected Results: It should be possible to obtain a history list of recently accessed Web pages via a button adjacent to the URL text box. It should be possible to click on the items in this list, to go directly to those Web pages."
        1 "This could be considered a duplicate of bug 14137. Currently, you can add a watch to receive emails sent to another bugzilla account. This can be used for when the watched person is on holiday, etc. I am using the feature to watch the work of fellow developers, so I get notified on all bugs related to the Zope Studio project. This gets unworkable as more people start contributing to Zope Studio and uncover and report more Mozilla bugs. Some of these people may also be notified of bugs I have no interest in. Adding a watch on keywords would solve this particular problem for me. Many keywords are related to specific projects or areas where QA needs to give special attention (like the beta1 and beta2 keywords). I think this should be relatively easy to implement, all that is need is a query for accounts that have a watch on the keywords specified in the bug."
        1 see http://help.netscape.com/kb/consumer/19980630-3.html for more details. it seems like we implement altmail easier in the new architecture than it was to implement in the old architecture. but, of course, no one has the time to spend on it. starting this bug for amiga@eudoramail.com (Don McCuiston)
        1 "This is a feature request which we started to implement in Composer but then realized that we werent really implementing the feature in such a way that stylesheets could be easily manipulated. Due to time constraints we will need to pull out the existing functionality so we can better polish other areas of the product. This bug is to implement the necessary UI and backend functionality to allow stylesheets to be added, reordered, and deleted from a Composer file."
        1 "The printer prints a separate page for each frame of my frames-based web-mail program. The specific program that I found this behavior in is IMHO, which is available at: http://www.lysator.liu.se/~stewa/IMHO/download.html I would like to only print the active frame, and that does not seem to be an option. Thanks"
        1 "On the Select Components dialog in the installer it says: Total download size but Ive already downloaded the whole Mozilla.... Shoudnt it be Total install size ?"
        0 "Feature: It would be nice to have a radio button choice on the preferences that would allow one to choose when using the editor/composer between using logical tags (i.e.,  and ) and physical tags (i.e.,  and ) when composing to produce bold and/or italic text. If logical tags were chosen, the B button would produce ... and the I button would produce .... Some of us always use logical tags in writing pages."
        0 "NSPR defines three I/O methods, PR_AcceptRead, PR_TransmitFile, and PR_SendFile, that are intended to exploit fast system calls on some platforms, for example, the AcceptEx and TransmitFile functions on Windows NT. However, only the bottom layer (i.e., the NSPR layer) can take advantage of these fast system calls; all the upper layers need to emulate these functions. It turns out that the emulation of these functions is the same for most upper I/O layers: - PR_AcceptRead: PR_Accept followed by PR_Recv. - PR_TransmitFile and PR_SendFile: memory-map the file, then write the memory-mapped region to the socket. Instead of having each I/O layer emulate these I/O methods, NSPR should provide the common, emulated implementation of these I/O methods, which the I/O layer implementers can just plug into the PRIOMethods table of their layer. Because normal NSPR clients dont call these emulated functions directly, these emulated functions should be exported as private functions and declared in private/pprio.h."
        1 (This bug imported from BugSplat, Netscapes internal bugsystem. It was known there as bug #89725 http://scopus.netscape.com/bugsplat/show_bug.cgi?id=89725 Imported into Bugzilla on 04/06/00 16:59) Entered by debsmith - 08-Aug-97 10:45
         Platform: Win95 Browser: 4.01 
         _______________________________________________________
         Netscapes current behavior of putting up a
         big red invalid signature warning on emails with unrecognized certs is
         possibly a bug, or at least a crying need for a feature improvement.
         
         Can we work with Netscape to get this fixed in a future version?
         
         Here is the behavior requested:
         
         Green stamp (Ok) - all ok
         
         Yellow stamp with question mark (Untrusted) - the signature on the message is 
         correct, and all certificates in the chain are valid, but theres no trusted 
         root.
         
         Red stamp with the universal not symbol (Invalid) - any one of the 
         certificates in the chain has been marked specifically as untrusted, one of 
         the certificates in the chain is expired or has an invalid signature, or
         there is a CRL entry for one of the certificates in the chain (when netscape
         starts using CRLs).
         
         ------- Additional Comments From jfriend 10/08/97 14:55 ------- 
         
         I understand why they are asking for what they are asking for - it is the way 
         the end-user thinks of the problem. I also understand that its not as simple 
         as that because a fourth possibility is that the message has been mucked with 
         to 
         make it look like the yellow or red case. Even 
        1 "Overview Description: If would be great if clicking the navigator in the lower right corner could open more than two navigator windows. Steps to Reproduce: 1) Start browser 2) Click the open new browser icon 3) Click it again and nothing happens Actual Results: No more navigator windows can be opened with this browser icon. Expected Results: A new window should open every time the button is clicked. Reproducibility: Everytime Build Date & Platform Bug Found: Mozilla/5.0 (X11; U; Linux 2.2.12-20 i686; en-US; m14) BuildID: 2000040616 Additional Builds and Platforms Tested On: Slackware 7 Additional Information: This was also a limit in Netscape Communicator 4.x"
        1 "A thing which would be very nice, is a drop-down menu like the one in 4.x, so you can easily get to a page were you have been before. Maybe even make it so the user decides the criteria for which the urls are being chosen."
        1 "From Bugzilla Helper: User-Agent: Mozilla/4.72 [en] (Win95; U) BuildID: 2000040617 When I installed this build (the first I have in a while, since I took a small break away from testing mozilla), I noticed that the bookmarks sidebar shows two columns of the same bookmarks. These bookmarks were imported from the 4.x profile. Its a bit confusing, and a bit annoying to have two columns of the same bookmarks. Reproducible: Always Steps to Reproduce: 1. Install 2000040617 and import the 4.x profile 2. Look at the My Sidebar, Bookmarks tab 3. The previous builds that Ive seen didnt show the same bookmarks twice, in two columns, in the My Sidebar. I dont know why it does now, but I dont think its a good idea."
        0 "The download progress percentage should be displayed in the menubar of the Progress Window, if not at all times then at least when its minimized (assuming a minimize button is coming). This is a nice feature of IE that allows you to see how far along a download is without having to bring the progress window to focus."
        1 "On a page like the query page (http://bugzilla.mozilla.org/query.cgi) with so many listboxes, its often hard (if not impossible) to tell which one has the focus at any given time. This may cause confusion when, say, youre filling out a long form. The same goes for comboboxes; the Windows standard is that after an item in a combobox is chosen, the background of the item remains blue and a focusrect is typically drawn around the border of the box as well. Mozilla provides no visible focus indicator after youve chosen an item in either a listbox or a combobox."
        0 "In all 32-bit versions of Windows, the GetCapture() and ReleaseCapture() API calls are used for standard buttons to produce the following effect: if the mouse is pressed down on a button, the button - as expected - is given the appearance of being pushed, or flattened, downwards. If the mouse pointer is then moved around the screen with the left mouse button still held down, the button is returned to its previous state - it is no longer flattened, or pushed downward. However, if the mouse is again then placed over the button (with the left mouse button still held down as it was initially), the button again gets the appearance of being pushed. Wow...that took quite a bit of text to explain a small Windows standard In any event, the standard behavior of Mozilla HTML buttons is to simply keep the button pushed down if the mouse pointer, with the left button still down, is moved off of the button. This feels awkward, sloppy and just overall wrong in Windows."
        0 "It seems rather useless to me to have that little grippy for the Taskbar. Pressing it only gains you a nominal amount of screen real estate; the difference in screen space when its collapsed is so barely noticeable that you might as well just keep the taskbar open. Then again, I guess if you find the buttons along the bottom distracting, this is helpful. I cant really come up with a better idea, save for having no grippy at all. Perhaps pressing it would just hide the taskbar entirely (it could then be reopened from the View menu), though this wouldnt conform to the standards of other Mozilla toolbars (especially the ones at the screens top)."
        0 "Changing the Sidebar option on the View menu to My Sidebar keeps better consistency (as the title of the actual frame that opens is My Sidebar) and is also a bit UI friendlier in terms of the average-Joe."
        1 "Here is an idea that would make people happier and have a faster browsing experience. Mozilla should figure out which websites are visited most frequently and download the graphics that always appear on the page like the title and things it also could download graphics that dont always appear but that are used commonly. Mozilla easily has the fastest rendering capabilities around but this would make it even faster thus making the end-user happy. Just thought you would like this idea I came up with."
        1 "When the browser fails to load an image or loads a corrupted version of an image, there is only way to make the browser try again is to right-click:view- image, then reload. Much preferable if the context menu for the image contains Reload Image or Show Image depending on the status of the image. Show Image: - when an image has has not been loaded either because the browser has not started loading the image or because the last attempt at loading the image failed at while starting or half-way, this is a way of asking the browser to try again if it has given up or try harder, if the browser is trying, but the image is taking a long time to come over a slow network. Reload Image: - when a corrupted image has been loaded, but the browser was not aware of, the user can intervene with a reload of just that image."
        1 "One of the things Ive never liked about Netscape is the way you set bookmarks when compared to Internet Explorer. In IE, when you add a bookmark, a dialog box pops up asking the user what they want to call the bookmark, and what folder to place it in (or to create a new folder to place it in). But Netscape makes it more of a hassle, making you go edit each new one you create from the Bookmark Management screen. So IE basically skips a step, and lets you get the placement and naming of the bookmark done when you first make it. IEs way is much more convinent and causes the user to spend less time configuring their bookmarks. Hence, I suggest Mozilla develop a similar scheme for adding bookmarks."
        1 "I really hate the pulldown on the query page that says use the same sort as last time. I dont remember what I used last time. I dont see how its relevant to what Im doing now. Maybe other people love this, but it just obscures whats going to happen so I am forced to ALWAYS pull this down just so I dont waste 5 minutes getting the wrong listing. Hiding the selection this way is the absolute opposite of helpful. Please fix this so it just pre-selects whatever I used last time. Then I can see whats happening and immediately determine whether I can go ahead or need to change this setting."
        1 "I have come up with a few ideas for better skin management. First off we need a skin selector and customizer. If, a skin selector were implemented that once the skin was chosen and confirmed that the browser would change on the fly to the new skin. In Microsoft Plus! and Plus! 98 you can take bits and pieces from different installed themes to configure the look of Windows, this is what I mean by a skin customizer taking bits and pieces of different skins and configuring Mozilla to take traits from the designated skin for the option such as the buttons in the Browser look like the Navigator Classic skin but the colors follow that of the fruitstripe skin. Finally we need a standardized skin packaging format that the skin selector recognizes and installs easily. The ease of use is what this whole idea is all about after all so the feature would have to be user friendly to use. This makes for a better skin interface system and gets rid of alot of the confusion and non-standard installation issues that exist now."
        0 "In show_bug.cgi, it would be nice if there were an and accept bug (change status to ASSIGNED) checkbox under the Reassign bug to radio button. This would allow someone to steal and accept a bug in one move, and would cut down on email notifications."
        1 "From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0) BuildID: 2000033112 Background Page with watermark didnt work. See clear difference betweeen Netscape and IE5. Reproducible: Always Steps to Reproduce: 1.Start Netscape 2.Type www.xs4all.nl/~famtie/test1.htm 3. Actual Results: background page with watermark did in IE5 But it didnt work in Netscape 6 (M15) Its clear difference between Netscape and IE5  ^^^^^^^^^^^^^^^^^^^ Its added this bgproperties = Fixed This mean Background with watermark. It would be beautiful. This website is designed by Frontpage 97 / 2000."
        1 "First, on our firewall, we have to change the default port that AIM uses, otherwise it will not connect. Unfortanately, in IM for Netscape 6, IO cannot find a way to change this. When I try dialing in on another ISP, It will verify my name, and I get to step 4, transfering, and it sits there. I am sure the second is a known problem, but the first needs to be fixed."
        0 "Build: Pulled from cvs 10-Apr-2000 Textfield widgets placed inside a menupopup dont handle events properly. For both single and multiline textfields - they render correctly but text cannot be entered into them. Reproducible: Always"
        1 "When you use the installer it can take quite a long time to actually install Mozilla. Why not show some informative image with the capabilities of Mozilla. Meaning that while the Install Progress runs show so images and text about how fast and wonderfull Mozilla is. The old netscape472 installer had that. I hate just sitting and looking at the Install Progress dialog..."
        1 "It I compose a new mail and then decide to change my sender address (having created muliple accounts of course) and select a different one from the select field, it should be possible to change the attachment too (if the corresponding has an attachment defined) BUT, it should not automatically change, perhaps there are people who want s to change their sender address but keeping the attachment my idea is, that a checkbox could appear after having selected the new address where I can check if the attachment should be changed too ( [ ] change attachment too? ) , after mozilla has changed the attachment, it should disappear again"
        1 "It s not possible to configure where my deleted Mails should go. I want them stored on my local folders account too, but I cannot find an option to configure this."
        1 "Both NN4 and mozilla have a preference to disable stylesheets. Though signed script can read preferences to check whether stylesheets are enabled or disabled there a often requests by script writers to give a(n unsigned) script a property to check. I suggest to introduce a property to the navigator object, it could be named navigator.cssEnabled for instance or navigator.styleSheetsEnabled Such a property is particular important for Mozilla/NN6 as you could get that info with a cheap trick in NN4 (insert a positioned ided test element and check whether document.elementID is defined) but this doesnt work with Mozilla/NN6."
        1 "This is related to the discussion in bug# 22342 (also 34752, and 35069) regarding whether the Open File dialogue should use the same window or open a new one. Please add an Open New Window? flag to the Open File dialogue to allow me to choose whether to use the same window or open a new one. You can default the flag to Yes every time, or better yet, allow me to set the default behaviour. Note that, in Composer, if the user chooses to use the same window, it could trigger a Do You Want To Save? dialogue. In that case, the user must have the option to cancel out of the Open File dialogue altogether, so that he stays in edit on the original file without saving. Similarly, I would like the File|New menu in Composer to have separate choices for a new Composer Window and a Blank Page, with the latter simply clearing the current window. This option would make me very happy. Ive always found it extremely annoying that, when I finish with one file, and I want to edit another one, Composer insists on opening a new window, forcing me to close the old window, and reposition the new one where I already had the old one. No other editor makes me do this (IIRC). Actually, if you want to do something really cool, allow me to drag file names from the Open dialogue into the window of my choice (without closing the Open dialogue). That way, I could quickly scan through a whole series of files to find what I want. As an experiment, I just tried dragging file links from the Netscape browser 
        1 "During the Friday Mozilla Developer meetings QA breakout session, one of the overwhelming complaints was the number of Netscape 6 PR users who are reporting bugs into Mozilla. One possible solution to this was to put an interim step in between clicking on the Bugzilla Helper (and possibly the regular bug reporting form) and the actual form. Specifically: 	- Sniff the user agent 	- If the user agent is a commercial build, pop up an annoying dialog to the effect of, If youre reporting a bug against Netscape 6, please use the Netscape Feedback Center. Only bugs reported against Mozilla builds will be accepted here. (Ideally, one of the buttons should enable the user to be redirected to the Feedback Center) 	- If the user agent isnt a commercial build, load the requested bug form. However, if lots of Mozilla contributors use Netscape 6 (whether now, or 5 months from now) as their dominant user agent, rather than a Mozilla build, this could annoy a lot of people. Thoughts?"
        1 "We do need span (like in tables) for XUL grids ebfore we can re-create most of the common dialogs, properties dialogs and prefs panels among others. This is a pre-requisite for skinnability for NS 6."
        1 MCT: poster is a request for a Cc of followups to /this/ news post. MCT: nobody is a request not to have a Cc sent for followups to /this/ post. This header isnt in a RFC, but it is supported by more than dozen other newsreaders and is under consideration for USEFOR.
        1 "RFE for an up button that chops off the last part of the URL. It would probably be off by default. The main reason Im posting this as a browser RFE instead of asking skin developers to do it (aside from the fact that im a bugzilla-vote whore) is that there should be standard code inside mozilla that the skins can call. Theres no point in each skin developer creating the functionality separately, especially since they would probably each do if differently. Potential problems: - How would it handle a site where /foo/ redirects to /foo/index.html? See also the following slashdot comment and its replies: http://slashdot.org/comments.pl?sid=00/04/08/1110208&threshold=-1&cid=148 (more permanent url) http://slashdot.org/articles/00/04/08/1110208.shtml#148"
        1 "The attachment that follows contains several examples of data: urls taken literally from RFC 2397. There are several problems with them in a current Mozilla build. 1. The plain footnote doesnt work at all. Its supposed to show up as a link to text/plain data. 2. The gif image doesnt show up. 3. The greek data shows up correctly as a link to a new page, but isnt URL-decoded. 4. Nothing happens to the vendor-specific data. (This is *probably* just because unknown media types arent currently handled, but I thought I should point it out.) Its not clear to what extent this is related to bug 30822. The image/jpeg data URL in that bug displays correctly, but the image/gif in this one does not. All of these links work to some extent in Communicator 4.7, though URL-decoding doesnt happen there either. This is a current Mozilla tip build, compiled on FreeBSD 3.4 with gcc 2.95.2."
        1 "This next suggestion (autosave) is a suggestions to both forms and to the MailNews compostion window. I think Ive had this happen to me before and that is that I am typing an email (in the mail compostion window) and then the brower crashes (or a power failure), I have to type everything over again and I often find that if I have to type something over again, my heart or creativity wont be fully into the composition the second time. I suppose I could always press save (in mail compose) but if I am engrossed in my writting, I might forget. The program should do that (autosave) for me. The same thing with forms but more can go wrong with forms, since it may also be a brower crash, power failure but may be a result of a server (cgi script error) or I forgot to put in a field in the form, and am asked to press the back key, and to find the contents of what you wrote may or may not be there. This is similar to the new wallet feature (even it doesnt offer autosave, at least not yet). The wallet feature to me seems a bit to permanent since wallet offers the ability to save the name address, credit cards (saved from a form then a fill to similar forms) so we dont have to enter stuff again (like at amazon.com & other sites). No what I am talking about is only temporary as the contents of the form that I am typing now. Once it is actually sent (assuming there are no errors in the cgi server script), I wont need the contents that I have written anymore so what
        1 "This next suggestion (autosave) is a suggestions to both forms and to the MailNews compostion window. I think Ive had this happen to me before and that is that I am typing an email (in the mail compostion window) and then the brower crashes (or a power failure), I have to type everything over again and I often find that if I have to type something over again, my heart or creativity wont be fully into the composition the second time. I suppose I could always press save (in mail compose) but if I am engrossed in my writting, I might forget. The program should do that (autosave) for me. The same thing with forms but more can go wrong with forms, since it may also be a brower crash, power failure but may be a result of a server (cgi script error) or I forgot to put in a field in the form, and am asked to press the back key, and to find the contents of what you wrote may or may not be there. This is similar to the new wallet feature (even it doesnt offer autosave, at least not yet). The wallet feature to me seems a bit to permanent since wallet offers the ability to save the name address, credit cards (saved from a form then a fill to similar forms) so we dont have to enter stuff again (like at amazon.com & other sites). No what I am talking about is only temporary as the contents of the form that I am typing now. Once it is actually sent (assuming there are no errors in the cgi server script), I wont need the contents that I have written anymore so what
        1 "When you press fx the Get Msg button the text on the button moves along with the button. That should not be the case. The graphics on the button might move (seems to be pressed down) but the text should not move. Thats also the case the mail buttons in Netscape 4.72"
        1 "It would be nice to have a dialog that shows every link on a page, sorted by extension or MIME type or URI or link text. It might let you select links to download to a directory, optionally mirroring the directory structure of the website. To go a long with this, a download manager like Operas with a pause/resume/schedule feature would also be useful. Another option might be to open all of the selected links in new windows, which might be useful on a news site to read all the news stories, or on a clip art site to expand all the thumbnails, especially on slower connections."
        1 "It would be nice if lynxcgi, lynxprog, and lynxexec URL schemes worked in Mozilla. Or a work-a-like."
        1 "Clicking on an URL in a mail/news should open a window instead of reusing a random other browser window."
        1 "Currently (20000412) lines are not soft wrapped in the view source window. As a result, the users has to scroll both horizontally and vertically to read the source. I suggest soft wrapping the lines for better readability. If indicating hard line breaks and soft line breaks is important, some sort of visual clue about the type of the break could be provided."
        0 "I had a need to easily tell from a list of bugs which ones had been flagged with needing a group bit set in order to view them and which ones were visible to the public. The attached diff adds a * right after the bug number of any bug that has a nonzero groupset. Theres probably a few ways this could be done... one idea I was toying with was changing the color of the summary or something more subtle like that. But this works if you like it enough to check it in."
        0 "It appears that there is no mechanism to limit the number of history entries in nsSessionHistory. This could be classified as a non-trivial memory leak since history entries take a fair amount of memory with urls, hash tables, LayoutHistoryStates etc."
        1 "When you expand a thread by clicking on the twistie, the thread pane should scroll so that as much as possible of the thread is visible. To reproduce: 1. Open a newsgroup 2. Expand a thread whose thread leader is at the bottom of the thread pane Actual result: The thread will expand invisibly, below the lower edge of the thread pane. Expected result: The thread pane should scroll up until the last message in the thread is visible, or the thread leader is the topmost thread in the threadpane, whichever comes first. Linux build 2000.04.11.09"
        1 "If you are on a long web page, then click a link then click BACK, it doesnt return to the same scrolled position as you were at before (like NS4.x)."
        1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; N; Linux 2.2.14-15mdk i686; en-US) Mozilla/m13 BuildID: 2000022916 simple, beautiful, obvious, you know you love it!!! No more need to have start on last viewed page option, just click on the back button and youre where you were. Reproducible: Sometimes Steps to Reproduce: 1.start a new session 2.now where was I last time?? 3. Actual Results: womble around like deranged buffoon. throw fits. start fight with neighbours. Expected Results: back-button click results in happiness, marriage saved, children run joyfully in streets"
        0 It should be possible to move and copy once created addressbook cards to other address books
        1 "The rendering of Word Documents should be worked into the browser. It would have to work like an adobe acrobat document (.pdf) does itwould automatically render in the browser. However if someone tried to open a document from their hard drive it goes into Word Viewer or Word if they are installed."
        0 "Add a XML-RPC client component, sos we can call remote server procedures."
        1 "In situations where something entered is not found in a list of possibilities (email addresses, keywords, etc), it would be nice if a fuzzy search could find a list of what the user could have meant and ask: Keyword dontest not found. Perhaps you mean DONTTEST, or DONTTEST2? where the possibilities are hyperlinks that allow the user to choose an alternative. In cases where there are multiple failures you would get one screen after another (this would occur rarely anyway)."
        1 "When using system colour names from CSS2, such as ButtonFace and ActiveCaption.. they are not updated in the browser if I change my colour scheme, unless I restart the browser."
        1 "Internet Explorer 5 and Netscape Communicator 4.7 are displaying a list of the last visited sites when I click on the Back-Button with the right mouse button. This features is already not implented in the Mozilla Browser but very useful I think."
        1 "From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; N; Linux 2.2.14-15mdk i686; en-US) Mozilla/m13 BuildID: 2000022916 The master password feature is neat for important passwords - you can have different passwords without having to remember them all. But lots of sites now ask for a name and password, even when there is no security involved. For example, I have a password at FT.com and slashdot. So if someone finds it, they can read FT and post to Slashdot in my name - no big deal. There should be an option to remember site passwords without having to type in the master password. The user can be trusted to use this responsibly - its his/her data at risk. Reproducible: Always Steps to Reproduce: 1. get asked for password 2. type it in Actual Results: netscape offers to remember password no option to disable master password Expected Results: should be option to disable master password for this site it might be possible to set this behaviour as default using a dont show me this again button.... I do have an awful lot of pointless passwords!"
        0 "A variation on RFE bug 35835, copy URL option on right-clicking bookmark in manage bookmarks window: This would be even more useful on the Bookmarks menu (any of them)... in any case where the user knows what the page is, doesnt need to see it, but does want to copy and paste the URL into a mail message or web form. As it stands in NN 4.72, even though it is wasteful, it feels easier to open a new window, select a bookmark, wait for it to begin loading, copy the URL, hit [Stop], close the new window, and paste, rather than going into Edit Bookmarks..., navigating the tree there, opening the properties dialog, and so on. Either way, it is a multi-step process that involves closing windows in the midst of what ought to be a simple copy-n-paste. In contrast, navigating to the bookmark through the normal cascading menu and then selecting a Copy URL item on a context menu would allow the user to go directly from the copy to the paste, with no extraneous actions."
        1 "Overview Description: A way to install PSM on a filesystem that doesnt have write permissions as a normal user is needed. Steps to Reproduce: 1) As a root (Administrator) user unpack mozilla and make sure it is owned by root 2) Start mozilla as a normal user 3) Try to install PSM Actual Results: Mozilla segfaults Expected Results: Either give a dialog box that says no permission to install or allow the user to enter the root password to install it. Reproducibility: everytime Build Date & Platform Bug Found: 2000041617 Linux i686 Additional Builds and Platforms Tested On: any OS with directory permissions Additional Information: Possibly move the default install directory for PSM to the profile directory?"
        1 "At the moment, engineers often attach `[RFE] to the Summary lines of bugs which are already marked as `enhancement. The argument is that this makes them easier to see. Therefore, the current means of highlighting enhancements in bug lists is not working as well as it needs to, and an additional (automatic) method should be used. I propose that color be used in bug lists to indicate the severity of a bug. This would make it a lot easier to scan a list for RFEs -- or, for that matter, for blocker or critical bugs. I suggest the following colors: * blocker = #cc0066 * critical = #cc0000 * major = #660000 * normal = #000000 * minor = #003366 * trivial = #336666 * enhancement = #006600 These choices would need to be checked with color-blind users to ensure they were all readable. It would be best if this was implemented using a style sheet (
        , etc), rather than as a hard-coded FONT tag, to allow for easier changing of the formatting in other Bugzilla installations. Note that this bug clashes with bug 28884, which suggests using different colors for confidential/non-confidential bugs." 1 "When constructing any table - as so: ..etc.. the CELLPADDING attribute is not displayed. As it is part of the HTML4.0 standard, and is supported by just about every other browser, I think it should be included!" 1 "It would be nice if Mozilla would load the last page visited at the same vertical position that you were at when using the back button. Ex. When you reading a report that is 100 pages long and decide to follow a link while reading the 75th page, it would be desirable for Mozilla to jump back down to the same vertical position in the document when the user presses the back button from the link." 0 "The jar protocol should be able to read jar (or zip) files within other zip files, for example jar:jar:http://server.com/outerArchive.zip!/innerArchive.jar!/ file.txt." 1 "In NS 4.x theres an option to expand all the threads in a message folder. This would be really nice to have in Mozilla." 0 "See bug 36145 for the Mac version of this bug (which came from a thread on mozilla-xpcom). Erik, please close this if its not relevant to Unix for some reason. troy implemented a version of nsRenderingContext::GetWidth for Windows that measures text in larger chunks, for performance reasons. This needs to be implemented on Unix too." 1 "When considering the amount of confusion that could be caused by someone who loads up his or her own chrome with cryptic (or invisible?) buttons, I think it would be a good idea to have a button/selection which would switch over to the default chrome. If you walked up to your friends computer and couldnt see the buttons or didnt know which was which, you could then click a menu item reloading the chrome that shipped with Mozilla. A least common denominator type of thing so that no matter where you you will have the familliar default inteface available." 0 "patch to mozbot.pl to add the ability to check for files on a ftp site , and add tally data to the tinderbox listings [10:21:20] New files @ ftp://ftp.mozilla.org/pub/mozilla/nightly/latest mozilla-source.tar.gz; 10:49:29] Tinderbox status from http://tinderbox.mozilla.org/showbuilds.cgi [10:49:29] SeaMonkey : [Mac Debug Clbr: Success] [Mac Opt Clbr: Success] [Mac Opt Depend: Success] [Win32 VC6.0Clobber: Success] [Win32 VC6.0Depend: Success] [shrike Linux Clobber: Test Failed] [shrike Linux Depend: Success] [speedracer SunOS/sparc 5.6 Depend: Success] 7 success, 1 test failures, and 0 horked." 0 "Have the QA Contact field be editable and appear in New Bug submission form When submitting a new bug report, the QA Contact field does not appear in the new bug form. Sometimes the QA contact may differ than the default QA contact that gets assigned. Being able to change it at bug submission time would reduce bug email notification. If someone knows who the QA contact should be, it would be less steps to change it. Suggest: QA contact field be added with the text (Leave blank to assign to default component QA contact) Thanks!" 1 "from ./ posting: when opening a new window, got to the current site not the home page seems like a good idea/option to me. adding it to my list of Things To Do." 1 "Ive edeted the sidebar tab images to be antialiased. I think they look much better. Here they are: http://www.rpi.edu/~frantb/mozilla The four images go in the global/skins/default directory." 1 "I dont know if this is possible on all platforms, but... RFE: A SIGSEGV etc. (e.g. crash) on plugin-code should be catched instead of crashing the whole browser. Any suggestions how to implement this (on WinXX??, Unix) ? How does GDB do the trick ?" 1 "One of IEs major advantages over mozilla is that it loads faster by loading once, when the operating system starts. While you could complain that just makes the operating system take longer to load, it actually is advantageous overall, because it only needs to be loaded once (provided it doesnt crash). Id like to see a new feature added to mozilla to make it stay resident, even when all browser windows are closed. When the user tries to open a mozilla window, it would instead call the new-window code from the part thats already loaded, reducing the apparent loading time by perhaps a factor of five. The prefs would probably look something like: [x] Keep mozilla resident after all mozilla windows are closed. [ ] In addition, load mozilla into memory when my operating system starts. The system would have to be pretty robust: (1) it should unload correctly if the browser crashes (2) the user should be able to kill it by clicking on something and selecting quit and unload in case (1) doesnt happen. for win32, a system tray icon would work. (3) it would have to be able to handle multiple versions of mozilla, either by simply preventing them from intermingling or actively unloading unneeded versions in order to conserve RAM." 1 "Id like bzip2-compressed tarballs of the builds (milestones at least) and source as well - when using a 56K-modem line to download saving 500-1000 KB per milestone really would feel nice. And it shouldnt be to hard to change any building scripts to do this." 1 "I was told a while ago to say this by itself(it is currently said in less detail in another bug of mine), and I didnt find a matching bug in Query. A problem with plugins on Comm. 4.x I noticed that some plugins had different protocols than others. For instance, the Quicktime plugin had different embed properties than Liveaudio, and when it overwrote it, some web pages didnt work. Different protocols for plugins that play the same file type require sniffing the plugin to determine what should be done. I think there should be a base set of embed properties and javascript commands for common file types such as MID, MP3, MPG, WAV that might have multiple plugins to choose from. For instance, a WAV player plugin should always have javascript commands to start, stop, get current location, pause, make a playlist, loop music, play for X amount of time, and start at a certain location, among others. Then, a plugin software can add its non-standard commands to the list. Of course, for plugins that do such things as show .pdf files, there is no need because there is only one plugin. Although there is a standard for HTML, there is no standard that I know of for plugins, and there should be. If the plugins are forced to be a certain way for NS browsers, then I do not see why they will not do exactly the same thing in Microsoft Browsers - so this will cause a standard to exist for both browsers. (Maybe my logic is flawed, though)." 1 "I have hundreds of bookmarks in my BM folder, and its growing. Id like to be able to count the number of bookmarks in the current bookmark window." 1 "I do not really know based on which libraries mozilla displays graphics, but one should be prepared when the JPEG2000 format is released and mozilla should support this standard as early as possible" 0 "As mentioned by a poster on todays Slashdot.org article about the M15 release, I believe it would be a good idea making the tarballs available with both bzip2 and gzip compression. The smaller size of the bzip2 compressed tarballs would be of particular benefit to those people who wish to try Mozilla but are on modem lines where they have to pay for their online time. The Linux Kernel is an example of something available in both .tar.gz and .tar.bz2 formats." 0 "Were using bugzilla as both a bug-tracking and system administration trouble-ticket tracking system. The below diff allows default product and component to be set via command-line switches. The switches are: -p=PRODUCT -c=COMPONENT This comes in really handy if you specify different procmailrc files based upon which email address things are sent to. We have various queues for network operations issues, development, QA, etc. Hope this makes it into release! 724,731d712 < if ( /-p=[]?(.+)[]?/ ) < { < $DEFAULT_PRODUCT = $1; < } < if ( /-c=[]?(.+)[]?/ ) < { < $DEFAULT_COMPONENT = $1; < }" 1 "waterson, hyatt sez yo da man for overlays now ;) XML errors in XUL overlays cause assertions (in debug builds, and I assume crashes in optimised builds), rather than XML errors to be reported. This is somewhat problematic from a XUL debugging point of view ;) Could we get some kind of accurate error logging to stderr? Even a youve probably fucked up your XML message is better than a cryptic crash or cannot dereference NULL COMPtr msg. ;)" 0 "From bug #18619 --- Description by Phil --- Some of the early UE specs showed a To/From column instead of the 4.x-style Sender column. Although I dont know for sure, I assume that the idea is to look at the From header, and if its the same as the current identity, show the To (or Newsgroups?) header. This has the advantage that it works when youre keeping a whole thread (both sent and received messages) together. ------- Additional Comments From mozilla@bucksch.org 2000-02-18 04:33 ------- Phil, I like your idea to compare the from with the email address of the current identity. I always keep threads subject-based and the 4.x behaviour annoys me. Will we have configurable columns? If yes, I suggest to offer the described combined To/From field (default) and additionally separate to and from field, for that case, the proposed logic fails (which may well happen in some advanced cases). I suggest for the combined column to not only compare with the From address of the current, but all identities. TkRat has this feature implemented." 0 "Assuming, we will have configurable columns, I suggest to offer the combined To/From field* (default) and additionally separate Recipient and From fields. The user would then be able to choose any combination of them using the generic column configuration methods. Relevance: The logic of the combined column might fail in some advanced cases. *the current and 4.x behaviour or the intelligent combined column proposed in bug 36489." 1 "the feature missing is already a buggy behaviour in bug 36494 - one could use mozilla to have a look over a bunch of pages in a defined period of time, example: you re web page owner you need to know if your webpages are reachable or not you could use mozillas shedule function to let mozilla visit these pages every minute or so to enable this, mozilla has to make difference between unreachable pages (see bug 36494) and updated pages, an additional check box would be great [ ] check availability" 0 "Download window should have the following options: - Close (window should not be automatically closed by default) - Retry (retry download if it fails, optionally asking when not possible to resume at previous position!) I think these two are really important. Misc: it would also be nice if the download URL and File could be modified and a new download started. (see also bug 25995)" 1 "Split up the description window for a new bug into following parts: Reproduce: Actual result: Expected result: Proposed fix: (seen at messages from ducarroz@netscape.com) This could help bug-reporters to specify what is needed. (Especially newbies, like me... P-) ) It might also be usefull if these parameters would initiate an auto-query (before the bug gets submitted) to stop bugs of becoming duplicate entries." 0 "I dont know if its an issue or its wittingly, but if you go to cvsqueryform.cgi page with module=, it prints the module twice in the listbox. This fixes it (although not much): --- /home/mozilla/webtools/bonsai/cvsqueryform.cgi Tue Nov 2 00:32:49 1999 +++ cvsqueryform.cgi Fri Apr 21 00:18:02 2000 @@ -103,6 +103,7 @@ # Print out all the Different Modules # for my $k (sort( keys( %$::modules ) ) ){ + if ($k eq $::FORM{module}) { next; } print