As far as I can see, CanonicaliseParams() never existed, despite it has been introduced in bug 156548. I checked in 2.20.4 and in 3.1, and I cannot find any function with such a name: Undefined subroutine &main::CanonicaliseParams called at /var/www/html/bugzilla/duplicates.cgi line 45. To throw this error, all you have to do is to go to duplicates.cgi?ctype=xul. This makes me think this feature never worked.

Did you mean $cgi->canonicalise_query()? See Bugzilla::CGI::canonicalise_query().

I see, bug 147833 removed CanonicaliseParams() from CGI.pl and replaced it with Bugzilla::CGI::canonicalise_query() on Oct 25, 2002. And duplicates.cgi started using CanonicaliseParams() on Nov 4, 2002; one week later! So bbaetz couldn't fix it as it wasn't checked in yet. And probably the reviewer of duplicates.cgi didn't upgrade yet and so still had CanonicaliseParams() in his installation. Said otherwise: this code *never* worked (even when it was implemented in 2.17.x) and nobody ever noticed it!! Probably a very useful feature. :-/ Does it mean we should kill it completely?

Bugzilla/CGI.pm was checked in (including canonicalise_query(), which used to be CanonicaliseParams() in the now-defunct CGI.pl) on 2002-10-25, half way through the review process for bug 156548. I strongly suspect Myk did a CVS update and then checkin without realising that a function he called had disappeared from under his feet. And it took so long for it to arrive on b.m.o. that everyone had forgotten it existed. Ah, well. Gerv

I for one don't object to removing features that are never used. If somebody wants to make an XUL interface for Bugzilla, that should be a separate project. (And was--Bugxula, a while ago, but it died.)

Actually this feature has always worked at its canonical location, it just hasn't worked from the redirect built into duplicates.cgi. Here's where to go to access the feature on b.m.o: jar:http://bugzilla.mozilla.org/duplicates.jar!/duplicates.xul Note: it can take some time for the tree to populate. Nevertheless, I doubt the feature is used very much on b.m.o or at all on other installations, since you have to sign it (or compromise the security of your web browser) to get it working, so I'm fine with removing it.

As the XUL version still works I don't think we should remove it. We should instead just remove or fix the broken redirection code.

Are we sure the XUL version still works? XUL has changed a fair bit since 2002. Gerv

I wrote: > Here's where to go to access the feature on b.m.o: > > jar:http://bugzilla.mozilla.org/duplicates.jar!/duplicates.xul Err, actually it's: jar:https://bugzilla.mozilla.org/duplicates.jar!/duplicates.xul Teemu wrote: > As the XUL version still works I don't think we should remove it. We should > instead just remove or fix the broken redirection code. I think it actually is worth removing it if it isn't being used, even though it still works, given that it's unowned and that we haven't made any other moves towards building XUL-based interfaces into Bugzilla. And I think Max is right that a XUL-based interface should be a separate project, with the core Bugzilla project providing just the standard HTML interface plus an API that other projects can use to provide alternate ones. Gerv wrote: > Are we sure the XUL version still works? XUL has changed a fair bit since > 2002. I don't think it's changed a whole lot, actually. In any case, the parts that the Duplicates Report needs haven't changed, so it still works.

I just tried it; I had to trust "America On Line" twice, and then every time I load a bug, but it worked. Gerv

> I had to trust "America On Line" twice, and then every time I load a bug That's because AOL bought the certificate that we used to sign the code. Alternately, you could enable codebase principals, use the unsigned version, and trust bugzilla.mozilla.org instead.

Comment on attachment 321158 [details] [diff] [review] patch, v1 myk said he doesn't have time to review this patch.

Comment on attachment 321158 [details] [diff] [review] patch, v1 I think we should also modify checksetup to remove duplicates*.rdf. Otherwise I'm fine with removing all this.

Remove data/duplicates*.rdf

Comment on attachment 321530 [details] [diff] [review] patch, v2 Looks good. Of course, old .htaccess files will continue to have the allow statement in them for duplicates.rdf.

(In reply to comment #15) > (From update of attachment 321530 [details] [diff] [review]) > Looks good. Of course, old .htaccess files will continue to have the allow > statement in them for duplicates.rdf. Yes, but that's not a big deal as they no longer exist anyway. And this is probably safer than trying to fix .htaccess ourselves, especially if an admin edited it manually.

Checking in collectstats.pl; /cvsroot/mozilla/webtools/bugzilla/collectstats.pl,v <-- collectstats.pl new revision: 1.65; previous revision: 1.64 done Checking in duplicates.cgi; /cvsroot/mozilla/webtools/bugzilla/duplicates.cgi,v <-- duplicates.cgi new revision: 1.62; previous revision: 1.61 done Removing duplicates.xul; /cvsroot/mozilla/webtools/bugzilla/duplicates.xul,v <-- duplicates.xul new revision: delete; previous revision: 1.2 done Checking in Bugzilla/Install/Filesystem.pm; /cvsroot/mozilla/webtools/bugzilla/Bugzilla/Install/Filesystem.pm,v <-- Filesystem.pm new revision: 1.30; previous revision: 1.29 done Checking in docs/en/xml/security.xml; /cvsroot/mozilla/webtools/bugzilla/docs/en/xml/security.xml,v <-- security.xml new revision: 1.19; previous revision: 1.18 done Removing js/duplicates.js; /cvsroot/mozilla/webtools/bugzilla/js/duplicates.js,v <-- duplicates.js new revision: delete; previous revision: 1.2 done Checking in template/en/default/filterexceptions.pl; /cvsroot/mozilla/webtools/bugzilla/template/en/default/filterexceptions.pl,v <-- filterexceptions.pl new revision: 1.112; previous revision: 1.111 done Removing template/en/default/reports/duplicates.rdf.tmpl; /cvsroot/mozilla/webtools/bugzilla/template/en/default/reports/duplicates.rdf.tmpl,v <-- duplicates.rdf.tmpl new revision: delete; previous revision: 1.4 done

Part of the deleted code in this bug began shipping again in the 3.1.x and 3.2.x series, but was removed (yet again) in 3.3.x and later. See Bug 501040 for more details.

(In reply to comment #18) > Part of the deleted code in this bug began shipping again in the 3.1.x and > 3.2.x series, but was removed (yet again) in 3.3.x and later. See Bug 501040 > for more details. This statement is completely wrong. You misunderstand what the version field means, as I explained in bug 501040.

(In reply to comment #19) > This statement is completely wrong. You misunderstand what the version field > means, as I explained in bug 501040. Yep, my bad.

Added to the release notes for Bugzilla 3.4 in bug 494037.