[Bug 31668] break C API to drop deprecated stuff

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Wed Feb 1 12:59:15 CET 2012


https://bugs.freedesktop.org/show_bug.cgi?id=31668

--- Comment #13 from Simon McVittie <simon.mcvittie at collabora.co.uk> 2012-02-01 03:59:15 PST ---
(In reply to comment #11)
> http://cgit.freedesktop.org/~jonny/telepathy-glib/log/?h=remove-deprecations
> 
> Here is some more stuff that's been deprecated to remove.

This branch looks ideal, please merge.

(In reply to comment #12)
> > * tp_message_to_text should still return whether there's non-text content
> [...] returns TRUE if there *is* text content (i.e. @text is set),
> and the latter returns NULL and sets @non_text_content to TRUE if there *isn't*
> text content... I think the former is nicer.

There's a subtle distinction between what I'm asking for and what you seem to
be suggesting: the point of the NON_TEXT_CONTENT flag was not "there is no text
here", it was "by condensing this Message into plain-text, you are losing
something".

Consider this composite message:

    text/plain: "hey look at this funny picture lol"
    image/jpeg: ...

This would "text-ize" as:

    ("hey look at this funny picture lol", NON_TEXT_CONTENT)

which a text-only UI could display as something like

    hey look at this funny picture lol
    [Non-text content not displayed, sorry!]

Meanwhile, this message (which could be produced automatically by a
sufficiently clever CM receiving only the HTML version):

    text/html: "look at this <a href="http://...">website</a>", alternative for
AAAA
    text/plain: "look at this website <http://...>", alternative for AAAA

would "text-ize" as:

    ("look at this website <http://...>", 0)

which a text-only UI would display without the apology.

I'm not sure how useful this would really be in practice, though...

> > * Commented-out Call stuff would be better #ifdef'd?
> 
> #ifdef'd on what?

I don't know, #ifdef FIXME_REINSTATE_WHEN_CALL_IS_STABLE or something. Or you
could leave them commented-out (or #if 0) and have an additional comment that's
something distinctive we can grep for, or references a new bug that says
"reinstate Call-related things in tp-glib when it's stable", or something.

I do agree with you that it's a good idea not to follow the usual "don't
comment out code" rule in this case.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA Contact for the bug.
You are the assignee for the bug.



More information about the telepathy-bugs mailing list