365 bytes, text/html
27.49 KB, patch
|Details | Diff | Splinter Review|
5.49 KB, patch
|Details | Diff | Splinter Review|
Build: 2003-05-07-03 Platform: All Expected Results: Selection List Properties Dialog should appear with correct content What I got: XML parser error (undefined entity) Location: chrome://editor/content/EdSelectProps.xul Line Number 78, Column 83: <label control="SelectName" value="&SelectName.label;" accesskey="&SelectName.accesskey;" /> Steps to reproduce: 1) Open the attached test case which contains the select element in Composer 2) Place focus on the select element 3) Choose Selection List Properties from the Format menu. 4) Notice XML parser in dialog.
Severity: normal → major
Keywords: nsbeta1, regression
Looks like this has been broken on the trunk for months. The oldest trunk build on windows I can locate is 2003-02-02-04. The problem is occuring in this build so I'm not able to track the regression any further.
Assignee: composer → jaggernaut
Target Milestone: --- → mozilla1.4final
Keywords: nsbeta1 → nsbeta1+
This broke on 05 Oct 2002 13:44 when cmanske checked in some fixes for me. The rest of the fixes are in the patch in bug 45495.
Depends on: 45495
Neil--we really don't want to ship 1.4 with this regression; is there anyway we can remove the dependency on bug 45495? Can you split up the patch enough to address this specific issue?
Created attachment 123274 [details] [diff] [review] Just the access key changes This is attachment 109104 [details] [diff] [review] but with all the code changes removed. /me wonders if alecf's sr= still applies...
Comment on attachment 123274 [details] [diff] [review] Just the access key changes r=jag. Let's find out if sr=alecf applies :-)
Comment on attachment 123274 [details] [diff] [review] Just the access key changes sr=alecf
Attachment #123274 - Flags: superreview?(alecf) → superreview+
Comment on attachment 123274 [details] [diff] [review] Just the access key changes This low risk patch fixes some localization issues in editor form element dialogs.
we're past a l10n freeze for 1.4. This is the minimal patch needed to fix the XML error? jag, to reduce perceived risk, and get l10n approval, can you come up with a minimal patch, one that just defines the missing accesskey? (or, if we get rejected due to l10n changes, we could also remove accesskey="&SelectName.accesskey;" for 1.4 final. and then, for 1.5, land the entire patch, with all the new accesskeys) also, looking at the patch: <!ENTITY EditLegendText.label "Edit Legend:"> +<!ENTITY EditLegendText.accesskey "T"> shouldn't that be "t", instead of "T"?
Comment on attachment 123274 [details] [diff] [review] Just the access key changes this seems right for 1.5, can we get a minimal patch for 1.4 final?
Created attachment 123890 [details] [diff] [review] Patch for branch only If previous patch can't go on the trunk, this might work for the branch, although I haven't tested it.
Tested, works fine.
Comment on attachment 123890 [details] [diff] [review] Patch for branch only sorry for the delay, sr/a=sspitzer thanks neil (and jag) I'll let you guys decide to open a new bug or not, for 1.5
Comment on attachment 123890 [details] [diff] [review] Patch for branch only I've just checked in this patch. Once 1.5 opens this can be backed out and replaced by the previous patch that does the better thing.
Removing keywords to get this off the radar.
Keywords: nsbeta1+, regression
Previous patch checked in as part of bug 45495.
Status: NEW → RESOLVED
Last Resolved: 16 years ago
Resolution: --- → FIXED
Verified on the Macho (2003-05-29-08) and win32 (2003-05-27-08) trunk builds.
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.