Closed Bug 56265 Opened 24 years ago Closed 24 years ago

can't view complete bug report

Categories

(SeaMonkey :: General, defect, P3)

defect

Tracking

(Not tracked)

VERIFIED FIXED

People

(Reporter: Brade, Assigned: rickg)

References

()

Details

(Keywords: regression, smoketest)

Attachments

(1 file)

I can't view a bug report with today's builds... mac/linux... branch or trunk. The bug report always gets chopped off after the dropdown list below the "Additional Comments:" textarea.
Same on Win32 2000101204.
add some keywords; this seems really bad for dogfood usage; I can't even Commit my comments using builds from today. :-(
Keywords: dogfood, regression, rtm
*** Bug 56266 has been marked as a duplicate of this bug. ***
this is odd. I downloaded yesterdays bits (which worked fine yesterday) and I'm still seeing the problem. maybe something in today's bits is corrupting *something* in my profile?
Yanking rickg's changes to htmlparser from yesterday fixes this in my tree: cvs update -j3.263 -j3.262 mozilla/layout/html/base/src/nsFrame.cpp cvs update -j3.319 -j3.318 mozilla/htmlparser/src/CNavDTD.cpp cvs update -j3.107 -j3.106 mozilla/htmlparser/src/nsElementTable.cpp cvs update -j3.44 -j3.43 mozilla/htmlparser/src/nsElementTable.h cvs update -j3.68 -j3.67 mozilla/htmlparser/src/nsHTMLTokenizer.cpp cvs update -j3.174 -j3.173 mozilla/htmlparser/src/nsHTMLTokens.cpp
Assignee: asa → rickg
Keywords: smoketest
I don't see it with an installer build from Monday
this regressed, was not on monday nor tuesday. Marking dogfood, this is killing QA.
Sorry, that should have been: cvs update -j3.319 -j3.318 mozilla/htmlparser/src/CNavDTD.cpp cvs update -j3.107 -j3.106 mozilla/htmlparser/src/nsElementTable.cpp cvs update -j3.44 -j3.43 mozilla/htmlparser/src/nsElementTable.h cvs update -j3.68 -j3.67 mozilla/htmlparser/src/nsHTMLTokenizer.cpp cvs update -j3.174 -j3.173 mozilla/htmlparser/src/nsHTMLTokens.cpp cvs update -j3.16 -j3.15 mozilla/htmlparser/src/nsIHTMLContentSink.h
*** Bug 56267 has been marked as a duplicate of this bug. ***
Here are some suggestions from buster@netscape.com: it would help if someone would look at the page in viewer and dump content, to verify that the content is getting through the parser correctly. if so, then dump frames and see if the frame construction code is properly building frames for the page. if not, then it's a rendering problem. if so, then it's probably a frame layout problem.
What is the bug number which causes this regression?
The checkin comment says this: fixed RTM++ compatibility bugs: 53011, 54117, 54651, 54834, 54840, 55095. sr=buster, r=buster, attinasi, harish, sfraser for various portions.
Here's the min test case: <form action="show_bug.cgi" method="get"> Show bug: <INPUT name="id" size="6" > <INPUT type="submit" value="Show"> </FORM> And the content model looks ok: html@02B62498 refcount=6< head@02B623A8 refcount=2< > body@02B81BD8 refcount=3< form@010C911C action=show_bug.cgi method=Get refcount=3< Text@010CBB60 refcount=3<\n Show bug: \n > input@010CB63C name=id size=6 refcount=3<> Text@010CAA10 refcount=3<\n > input@010CA99C type=submit value=Show refcount=5<> Text@010CA510 refcount=3<\n> > Text@010AC570 refcount=3<\n> > >
Status: NEW → ASSIGNED
I just removed my changes, and I still see the problem.
I think dist might need to get clobbered, rick. I didn't see the problem until i blew away dist and rebuilt with your changes. backing the changes out, forcing a rebuild in htmlparser, fixed things for me on linux.
I'm trying a full rebuild now.
This seems to me like a more descriptive testcase: <html><body> <FORM NAME="changeform" METHOD="POST" ACTION="do_nothing.cgi"> <SELECT NAME=resolution <OPTION VALUE="FIXED">bla1 <OPTION VALUE="INVALID">bla2 <OPTION VALUE="WORKSFORME">bla3</SELECT><br> bla4 </FORM> </body></html> Should show a selection box with three options (bla1, bla2, bla3) and then bla4 on a new line. Shows a box with three options: bla1, bla2, bla3 bla4 I don't know how to get 'viewer' and dump content.
Here are some bugzilla pages where this problem still shows up in build 2000101208 MTrunk Win2k http://bugzilla.mozilla.org/show_bug.cgi?id=30613 http://bugzilla.mozilla.org/show_bug.cgi?id=44437 Jake
Ok, this is bizarre. I have two windows showing the same page ( bug 44437 ) side-by-side. One displays only down to the "Resolve bug, change resolution to" and then the select box, which extends horizontally seemingly forever with no content below it. However, the second displays just as would be expected??? The window with the bad display was brought up earlier than the window with the good display. What the heck is going on???? jake
We're backing out the offending code now. Once we do, leaf will respin the dailies and life should be good. Sorry all.
this gets more interesting. using win98 with 101208, i have the same bugzilla-viewing problem until i log into my bugzilla account. now all is back to normal. could this be a server side issue, perchance?
the oneline fix seems to be working, committed, opening tree when other smoketests pass (respins on the way).
This should be fixed now; we backed out the one line change that caused the regression.
Status: ASSIGNED → RESOLVED
Closed: 24 years ago
Resolution: --- → FIXED
rick - what is the bug number assigned to that one line change which you backed out? Thanks.
*** Bug 56232 has been marked as a duplicate of this bug. ***
fixed on win32 and linux, waiting for mac before verifying this.
The cause of this regression seems to be bug 54834.
Thanks for the quick fix to address this bug. Even though this is "after the fact", we should still have a super review for this.
Verified on Win32 2000-10-12-13-mn6 branch builds.
Works for me: Platform: PC OS: Windows 98 Mozilla Build: 2000101214 M18 Trunk Build
Status: RESOLVED → VERIFIED
Sorry didnt mean to mark as Verified. Sorry for the spam
Status: VERIFIED → REOPENED
Resolution: FIXED → ---
it is still fixed.
Status: REOPENED → RESOLVED
Closed: 24 years ago24 years ago
Resolution: --- → FIXED
verified fixed on friday 13th commercial builds: windows 2000-10-13-06-Mtrunk & 2000-10-13-09-MN6 linux 2000-10-13-06-Mtrunk & 2000-10-13-09-MN6 mac 2000-10-13-08-Mtrunk & 2000-10-13-10-MN6
Status: RESOLVED → VERIFIED
Mass removing self from CC list.
Now I feel sumb because I have to add back. Sorry for the spam.
Product: Browser → Seamonkey
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: