How do I get kdialog to be used by thunderbird?

Duncan 1i5t5.duncan at cox.net
Sun Nov 15 09:46:10 GMT 2009


Steven W. Orr posted on Sat, 14 Nov 2009 17:19:17 -0500 as excerpted:

> I have looked everywhere. I'd like to be able to save files by using
> kdialog in thunderbird and firefox instead of using the gtk file picker.
> 
> Inside thunderbird I have the ability to set
> ui.allow_platform_file_picker
> 
> to true but if I do it uses the ugly builtin picker. False gets me to
> the gtk file picker. Is there a way to speciffy this in kde somehow? Or
> is this a thunderbird/firefox problem?
> 
> I'm running kde-4.3.2.

Here's the deal.  Applications on Linux/*ix today normally (normally, 
because there are some exceptions that build directly against X and 
provide their own widgets, but most newer ones use a toolkit, since it's 
easier) compile against a toolkit.  The toolkit provides widgets such as 
the file picker, etc.  On Linux, the Mozilla family of apps chose to 
compile against the gtk+ toolkit.  Thus, they'd normally use it.  
However, in the Mozilla case, they're also cross-platform and use a 
Chrome-based (this chrome was before the Google Chrome browser, it's what 
the Mozilla XUL based UI is called) UI by default, cross-platform.  Thus, 
the choice is between the built-in XUL/Chrome based file-picker, and the 
"platform" toolpicker -- in the case of *ix (Linux and the BSDs, also 
including OSX if it's built using the *ix/X/GTK based framework instead 
of the Apple framework), GTK+.

Back in the KDE 3 era, there was an experimental Qt-based (well, kde 
directly or qt, I'm not sure) Gecko based browser hacked up at one of the 
hack-fests, as a proof of concept, and there's a lot of KDE users that 
would have liked to have seen it go somewhere, but nobody seemed to take 
it and run with it and that single proof of concept it remained.

Meanwhile, the new popular kid on the block is webkit, which of course 
was based on KDE's own KHTML.  Apple Safari was the one that took and ran 
with it, but they have continued to work with the community to improve 
it, and then Google chose it over Mozilla Gecko to base Chrome on, and 
there are various other smaller browsers and other misc HTML/web based 
users.  Qt-4.4 or 4.5 (IDR which) added a webkit-based HTML rendering 
widget too, and plasma already uses that for the various HTML/web based 
plasmoids.  However, as of qt-4.5 the qt-webkit widget is still 
apparently somewhat immature and buggy -- plasma is apparently one of the 
first big users and they found some bugs and potentially really useful 
but still missing bits of the API.  (qt-4.6 is supposed to be rather 
better, and more kde could convert with it, but that's not going to be 
required until kde-4.4, in February, I think, and maybe kde-4.5, I'm not 
sure.)  It's likely that at some point in the future konqueror will 
switch over from the KHTML base KDE still maintains, which as I mentioned 
was where webkit started, to the webkit base, very likely the qt-webkit 
base.  But qt-webkit still has some growing to do first.  Meanwhile, 
konqueror continues to use khtml, and it simply doesn't have enough 
market share on its own to get websites testing with it and developing 
for it, so mozilla products still tend to be necessary for many KDE 
users, even if they are based on gtk instead of the qt kde is based on.

Meanwhile... Mozilla isn't dead just yet!  In fact, their market share is 
still growing, AFAIK, tho perhaps Google's Chrome and thus webkit took 
some of that (but IE's still the big one, still losing the most ground 
last I knew, which is a good thing, as it's unhealthy for /any/ browser, 
even a freedomware one, to have the market share it did or people tend to 
write to it instead of to standards).  Plus, on a very practical level, 
once folks get into the Mozilla addons, they often find rather quickly 
that they have a hard time doing without them, and the fact of the matter 
is, even if webkit begins to dominate the market, as long as the various 
webkit based browser makers can't agree on a common extension standard to 
match Mozilla's, it'll be a VERY long time before anyone can come even 
close to challenging the shear variety and usefulness of all those 
Mozilla extensions!  Thus, Firefox and Mozilla in general is likely to 
remain a dominant force for some time, if only due to their extensions.

One thing's for sure, the next couple years could be /very/ interesting 
on the browser side of things!  If MSIE keeps losing share, and Webkit 
and Gecko can split it between them, we could have a very healthy three-
way race ahead!  There has actually even been some talk of Mozilla 
switching to the webkit rendering engine as well, altho at this point, 
that's probably at minimum, getting the cart before the horse, as they 
say.  It remains to be seen, but but it's not going to happen in the 
immediate future, and actually, there's a LOT of folks that would prefer 
all three of the majors end up with 25-35% of the market each, with maybe 
10% split between others (not to leave Opera out of the discussion, for 
instance, tho I'm certainly not alone in wishing they'd go open source 
and not considering them ATM because they're not), with that considered 
perhaps the most healthy outcome possible, even if it's unlikely to stay 
that way long term because the market just doesn't tend to keep that nice 
a split over time.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

___________________________________________________
This message is from the kde mailing list.
Account management:  https://mail.kde.org/mailman/listinfo/kde.
Archives: http://lists.kde.org/.
More info: http://www.kde.org/faq.html.




More information about the kde mailing list