Sumo stuck, wooot. Let's cvs remove the in-product help content. We should CVS remove en-US first, obsoleting the l10n files, and then kick l10n in a rush. The files are not built anymore anyway, and they just look like a huge amount of work for localizers working towards fx3.0.x.
This is following up bug 423486 and bug 425159. This patch actually removes the help files in browser, it removes their entries in browser/locales/jar.mn (which had been ifdef'ed only so far) and get's rid of extra-jar.mn, which was just for localized screenshots. I made configure now error on --enable-help-viewer for browser. I didn't touch the toolkit part of help viewer, that's bug 425541.
Assignee: nobody → l10n
Status: NEW → ASSIGNED
Attachment #329794 - Flags: review?(ted.mielczarek)
Comment on attachment 329794 [details] [diff] [review] removing help files, clean up jar.mn, error on --enable-help-viewer for browser requesting review from gavin on the browser parts, ted for the configure.in part.
Attachment #329794 - Flags: review?(gavin.sharp)
Comment on attachment 329794 [details] [diff] [review] removing help files, clean up jar.mn, error on --enable-help-viewer for browser Do we really need to error out in configure? Is the viewer code completely useless without content?
Attachment #329794 - Flags: review?(gavin.sharp) → review+
The only use cases I would have would be an extension that wants it and manually opens it -- of course we don't ship the help viewer in fx 3.0, so they can't. The other one would be some custom built with offline help, but that doesn't work with just that option either. I'm on the side of 'error', unless there's sufficient noise in the extensions ecosystem to get offline help for them, and we'd add that back. mconnor, beltzner?
OK, error sounds fine.
Error seems entirely correct to me, FWIW.
Attachment #329794 - Flags: review?(ted.mielczarek) → review+
Comment on attachment 329794 [details] [diff] [review] removing help files, clean up jar.mn, error on --enable-help-viewer for browser landed on mozilla-central, http://hg.mozilla.org/mozilla-central/index.cgi/rev/9871082391f5. Requesting approval to land this on the 1.9.0.x branch, too. We really shouldn't bother new loclizers with content that's never shown, in particular if it's this much content.
Attachment #329794 - Flags: approval220.127.116.11?
Please take bug 448020 as well if you take this on the branch.
Axel, any reason we can't just leave this be for 1.9.0.x given that 1.9.1 is so close?
We still have people looking at cvs to start new localizations, and it'd be nice to not expose that huge amount of data to them without any real value. We might move over to 3.1 soon, but I'm not sure exactly when, it does depend a bit on how good we're doing with taking new locales on 3.0.x, too.
Comment on attachment 329794 [details] [diff] [review] removing help files, clean up jar.mn, error on --enable-help-viewer for browser Approved for 18.104.22.168. Please land in CVS. a=ss
Attachment #329794 - Flags: approval22.214.171.124? → approval126.96.36.199+
Could whoever lands this in CVS land bug 448020 at the same time?
We've had a new localization team (in this case, it was the team in Thailand) who did a significant amount of l10n work on these files recently without knowing that it was ultimately a waste of their time. The sooner we can get these off the server, the better. If we can act more quickly on bugs like this in the future, it would be appreciated. 2 months is a bit long imho.
They can use their work on SUMO, but needs to be adapted to Firefox 3. They should talk with djst, coordinator of SUMO.
Landed in CVS (including bug 448020), I'll land l10n follow ups shortly, unless .l10n disagrees. Leaving open to leave that on my lap.
Is this fixed, or is there anything left to remove?
Yeah, we can resolve this one.
Status: ASSIGNED → RESOLVED
Last Resolved: 10 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.