Closed Bug 428781 Opened 17 years ago Closed 17 years ago

NetError.xhtml broken

Categories

(Camino Graveyard :: General, defect)

x86
macOS
defect
Not set
major

Tracking

(Not tracked)

VERIFIED FIXED
Camino2.0

People

(Reporter: phiw2, Assigned: stuart.morgan+bugzilla)

References

Details

(Keywords: regression)

Attachments

(1 file)

XML Parsing Error: Location: jar:resource:///chrome/embed.jar!/content/global/netError.xhtml Line Number 60, Column 99: <link rel="stylesheet" href="chrome://global/skin/netError.css" type="text/css" media="all" /> works: 20080412-00 build fails: 20080413-01 build Home made build cvs checkout @ 20080412, ~17:00 PDT fails http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2008-04-12+00%3A00%3A00&maxdate=2008-04-12+17%3A00%3A00&cvsroot=%2Fcvsroot bug 292789 ?
Right from the error console: 8:40 PM: Security Error: Content at about:neterror?e=dnsNotFound&u=http%3A//phiw.phiw.phiw/&c=UTF-8&d=phiw.phiw.phiw%20could%20not%20be%20found.%20Please%20check%20the%20name%20and%20try%20again. may not load or link to chrome://global/skin/netError.css. --> sure is bug 292789, adding Benjamin to the cc Like Phil says (bug 292789 comment 101) > Oh, don't worry about us, we'll find out when you break us, same as always.
Blocks: 292789
BTW, FTP dir. listings also suffer from this: not able to load a stylesheet with chrome:// url In this case, the file is loaded as text/html and you can actually see the page. e.g ftp://ftp.mozilla.org/pub/mozilla.org/camino/ Security Error: Content at ftp://ftp.mozilla.org/pub/mozilla.org/camino/ may not load or link to chrome://global/skin/dirListing/dirListing.css. not sure if it is worth a separate bug.
[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9pre) Gecko/2008041302 SeaMonkey/2.0a1pre] (nightly) (W2Ksp4) I was not able to reproduce, neither the netError nor the Ftp cases. Are these issues Camino specific ?
Flags: blocking1.9?
xpfe-chrome-registration-only, which I guess means Camino-only these days, and no, it's not worth a separate bug: you have a single bug, that Camino's global chrome package needs a chrome:contentAccessible="true" attribute in the contents.rdf, and nsChromeRegistry.cpp needs to be taught about contentAccessible. http://bonsai.mozilla.org/cvsview2.cgi?diff_mode=context&file=nsChromeRegistry.cpp&subdir=/mozilla/rdf/chrome/src&command=DIFF_FRAMESET&rev1=1.321&rev2=1.322 which taught it about xpcNativeWrappers="true" ought to work as an example of how.
What does surprise me a little is that a missing or non-existent stylesheet triggers an XML parsing error. I thought bug 418391 had taken care of this.
bug 428747 Comment 3 and bug 428747 Comment 4 might be the simplest change and should fix any and all regressions from the fallout.
Keywords: regression
Target Milestone: --- → Camino2.0
(In reply to comment #5) > What does surprise me a little is that a missing or non-existent stylesheet > triggers an XML parsing error. It surprised me too. Venkman (in SeaMonkey (v2)) has the "XML parsing error" issue too: bug 428848.
Whiteboard: [Camino/Xpfe specific bug]
-'ing this as this appears to be something that needs to be worked around in Camino. I there is a core code issue here, please re-nom and specify what the issue is.
Flags: blocking1.9? → blocking1.9-
(In reply to comment #8) > -'ing this as this appears to be something that needs to be worked around in > Camino. I there is a core code issue here, please re-nom and specify what the > issue is. Bug 428747 comment 3 (and 4) indicate this should probably be fixed in rdf/chrome. That's core code, so renominating.
Flags: blocking1.9- → blocking1.9?
The XML parsing error is pretty much the same as what Martijn found in bug 428847 (and I guess the same applies to Venkman): the href URI in the PI contains a ':'.
Can you please explain why we can't take this in a dot release? Should this hold back everything else if it were the last bug?
Minusing, if you have an answer to comment 11, renom.
Flags: blocking1.9? → blocking1.9-
The consensus in bug 428747 seems to be that this should be changed in core; if that's the case, then Camino 2 is blocked on that change. If you are asking for the opinion of a Camino developer as to whether something that prevents Camino from using Gecko 1.9 should be a stopship bug, then the answer is yes. As for whether or not it could reasonably go in a dot release, without any idea of what the schedule for such a release would be I can't offer an opinion one way or the other.
Flags: blocking1.9- → blocking1.9?
Executive decision: rdf/chrome is camino-specific code. It's no longer core, because only camino builds or uses it. I think Camino should take charge of maintaining that code since nobody else uses it, and this bug does not need to block the 1.9 release.
Sounds reasonable; removing blocking request.
Flags: blocking1.9?
Attached patch fixSplinter Review
Switch to allow-all, per discussion in bug 428747.
Assignee: nobody → stuart.morgan
Status: NEW → ASSIGNED
Attachment #316287 - Flags: superreview?(mikepinkerton)
Attachment #316287 - Flags: superreview?(mikepinkerton) → superreview+
Landed on trunk.
Status: ASSIGNED → RESOLVED
Closed: 17 years ago
Resolution: --- → FIXED
Blocks: 428747
No longer depends on: 428747
(In reply to comment #10) > The XML parsing error is pretty much the same as what Martijn found in bug > 428847 (and I guess the same applies to Venkman): the href URI in the PI > contains a ':'. Thanks for the hint.
Whiteboard: [Camino/Xpfe specific bug]
verified with Version 2.0a1pre (1.9pre 2008041717)
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: