294605 - Use UTF-8 (Unicode) charset encoding for current templates

Status: RESOLVED FIXED

(Bugzilla :: Bugzilla-General, defect)

Description

[Marc Schumann [:Wurblzap]]

To prepare the way for bug 126266, and to make life easier for localizers,
templates should be encoded in UTF-8. (Most templates use 7-bit ASCII encoding
and therefore automatically UTF-8, too.)

Some templates have started using ISO-8859-1, though; this bug converts them to
UTF-8.

Comment 1

[Marc Schumann [:Wurblzap]]

Attachment: Patch

Comment 2

[Marc Schumann [:Wurblzap]]

It turns out UTF BOM header bytes may be displayed interspersed into pages if
templates are UTF-8 encoded and the page's charset encoding is something else
(ISO-8859-1, for example).

Morphing bug to want plain ASCII.

Comment 3

[Marc Schumann [:Wurblzap]]

Attachment: ASCII patch

Not too nice to mangle somebody's name :(

Comment 4

[Frédéric Buclin]

Comment on attachment 183995 [details] [diff] [review]
ASCII patch

r=LpSolit assuming I am the only one in this situation. But I would be very
happy if we could find a way to keep my first name with 'é' in it. :(

Comment 5

[Dave Miller [:justdave]]

(In reply to comment #2)
> It turns out UTF BOM header bytes may be displayed interspersed into pages if
> templates are UTF-8 encoded and the page's charset encoding is something else
> (ISO-8859-1, for example).

Uh, say what?  UTF-8 doesn't need byte order marks (That's a UCS2/UTF-16 thing),
and we shouldn't be using them.  The page charset should always ben UTF-8 as
well once bug 126266 lands.

> Morphing bug to want plain ASCII.

Please don't.  Or convince me why (the arguments presented thusfar are not
enough to convince me).

Comment 6

[Marc Schumann [:Wurblzap]]

(In reply to comment #2)
> It turns out UTF BOM header bytes may be displayed interspersed into pages if
> templates are UTF-8 encoded and the page's charset encoding is something else
> (ISO-8859-1, for example).

Well, when trying to set up a test case to show the effect I saw, I couldn't,
but all pages displayed just fine. I asked Frédéric to confirm, and he didn't
see any problems either. I think when testing this yesterday I must accidentally
have saved some template file with BOM header bytes.

So, back to the original assumption it is, therefore to the original goal, and
thus to the original patch.

Comment 7

[Frédéric Buclin]

Comment on attachment 183888 [details] [diff] [review]
Patch

seems to work...

Comment 8

[Frédéric Buclin]

In order to prevent such things in the future, runtests.pl should warn if
non-UTF8 characters exist in templates. Is there already a bug about that? If
not, should we open one?


Checking in template/en/default/admin/milestones/confirm-delete.html.tmpl;
/cvsroot/mozilla/webtools/bugzilla/template/en/default/admin/milestones/confirm-delete.html.tmpl,v
 <--  confirm-delete.html.tmpl
new revision: 1.3; previous revision: 1.2
done
Checking in template/en/default/admin/milestones/deleted.html.tmpl;
/cvsroot/mozilla/webtools/bugzilla/template/en/default/admin/milestones/deleted.html.tmpl,v
 <--  deleted.html.tmpl
new revision: 1.3; previous revision: 1.2
done
Checking in template/en/default/admin/versions/confirm-delete.html.tmpl;
/cvsroot/mozilla/webtools/bugzilla/template/en/default/admin/versions/confirm-delete.html.tmpl,v
 <-- confirm-delete.html.tmpl
new revision: 1.3; previous revision: 1.2
done
Checking in template/en/default/global/user-error.html.tmpl;
/cvsroot/mozilla/webtools/bugzilla/template/en/default/global/user-error.html.tmpl,v
 <--  user-error.html.tmpl
new revision: 1.109; previous revision: 1.108
done