view source caused Back/Forward buttons to be disfunctional. After two clicks on back button (one to be sure), viewer app crashed w/ Type 2 error. Used today's binary mozilla-mac.

Pierre -- can you please try this and tell me where the crash occurs? A stack trace would be nice. Thanks.

I tried really hard but I can't reproduce it, using the 03/23 build on a G3/266. Reassigned to the reporter. Could you provide more info in your bug report? Some step by step instructions from the time you launch the application would be especially useful, as well as a stack trace of the crash (type 'stdlog' under Macsbug). You can reassign the bug to me afterwards. There is more info about bug writing guidelines on this page: http://www.mozilla.org/quality/bug-writing-guidelines.html

are we really pursuing the 'Viewer' failure here? The apprunner behavior is not to crash but it is inoperable. Currently it looks as if it's not hooked up. It triggers a 'browser reload', which doesn't work either. bug 3915 is the view source bug for apprunner.

Of course, we are investigating the crash in Viewer. Reading bug #3915, it looks like the menu items that are not hooked up yet, which is different from the crash experienced here.

judic, if we cannot reproduce with latest Mac build and both viewer and Apprunner, we are going to mark Verified/WorksForMe. claudius, please take care of this one. Thanks!

I received this mail from Judi last Thursday: --- This bug can be reproduced reliably. Here's more details: 1. go to an outside page. I'm working on a redesign for www.netaction.org, so that's as good a target as any. I haven't found an outside page that doesn't work--that domain, www.netscape.com, or other. (so far) 2. view source. Select a little text. Copy it. 3. click back button. Crash. Other things beside the back button will also cause a crash, but I haven't found them to be as consistant. Sometimes just copying the text crashes it. I'm not talking a lot of text--a few lines will do the trick. Yeah, I know there's no back--it's a new window. I didn't realize this the first time when I first submitted the bug report, but I figured it out the second time when I needed to recover a page that was accidentally erased from the server but was sitting in the Viewer window. --- After I asked her to take a stack trace under Macsbug, she replied: --- Ouch. MacsBug crashed my system three times now. It lists various things, my file control blocks, then Displays resource info, Calling chain using A6/R1 links, Back chain ISA Caller, Return addresses on the stack, Stack Addr Frame Addr ISA Caller (system freezes). I tried turning off my extensions but that caused MacsBug to crash with a Type 1 error (and froze the system). The MacsBug version I'm using is fresh off Apple's site: v6.5.4a6. --- It looks like there isn't much more that she can do. Claudius, could you try to reproduce the bug?

Hey! I can reproduce it now, or sort of: the application quits but doesn't crash. Thanks Judi. Reassigned to myself.

So I did some poking around....with Apr01 builds Apprunner checks out fine on all platforms. Viewer isOK on NT. Viewer exits on Linux and MacOS This bug is slightly different than originally reported, but it may be easier to handle now. To reproduce: 1. Launch Viewer 2. Drag-select (highlight) some text. 2a. On MacOS select 'Copy' from the edit menu. Expected result: You should be allowed to highlight whatever text you want. You shouldbe able to execute the Edit->Copy command. Actual result: On Linux Viewer quits upon selecting text. On MacOS Viewer quits upon choosing 'copy'. I've updated the summary field to be more precise. Should the component be changed as well also? Is an mjudge bug? Selection and Search? Upping severity.

*** Bug 2599 has been marked as a duplicate of this bug. ***

Update: Linux now seems to be just fine(Apr05), only the Mac(Apr05) is now quitting on 'Copy' (linux does absolutely nothing on 'Copy'). I'm told sfraser and mjudge may have done some work on this. cc'ing them and changing platform.

*** This bug has been marked as a duplicate of 4153 ***

marking VERIFIED as a dupe of bug 4153