RE: Tiny font workaound... maybe?

Adrian,

I'm sorry, I have not had a chance to try your changes, I don't have CVS
access at work through our firewall. I haven't forgotten about it, I just
need to get to it.

BTW, did you or AL have any comments/suggestions on the modified code I sent
out?

-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of
Adrian E. Feiguin
Sent: Thursday, February 24, 2005 8:55 PM
To: [email protected]
Cc: Al Hooton; gtkextra
Subject: Re: [gtkextra] Tiny font workaound... maybe?


I understand Al's position, but I want to explore all possible solutions
before giving up. I just committed to CVS a new code for gtkplotgdk.
Very simple idea, instead of a stand-alone object, I make it deppend on
a widget parent that will provide the pango context. I'm guessing that
if fonts appear fine in the gtkwidgets, they'll do as well on the canvas
with the new approach. Can you please verify this? Thanks a lot!

<ADRIAN>

[email protected] wrote:

>Al,
>I completely agree with you. This fix should be temporary and that is why
I'm keeping the pango code in the source code. I wasn't even thinking of
making this part of a CVS branch but if you guys think that it should then
we can try it.
>
>A brief explanation of events so far:
>
>-Introduced the preprocessor flag _USE_PANGO_BASED_CODE into the
gtkextra/Makefile
>
>-The plot routines and the GtkPSfont definition was updated incorporating
the functionality from gtkextra1.
>
>-I have built two sets of gtkextra2 libraries, one with
_USE_PANGO_BASED_CODE defined and one without.
>
>-Testing has been limited to the test* programs included in gtkextra2.
>
>The results of the tests are encouraging, only a few test cases reported
errors (from GTK). I have to say that I have not attempted to fix them since
the errors appear on both the PANGO enabled code and the PANGO disabled
code:
>
>testcandles
>error reported:
>** (lt-testcandles:10501): CRITICAL **: gtk_plot_candle_draw_legend:
assertion `GTK_WIDGET_REALIZED(data->plot)' failed
>-If I'm correct the widget is realized when gtk_widget_show() is called.
Maybe an operation that expected the widget realized was called before the
call to widget_show()?
>
>testboxes
>error reported:
>** (lt-testboxes:10520): CRITICAL **: gtk_plot_box_draw_legend: assertion
`GTK_WIDGET_REALIZED(data->plot)' failed
>-Same as above.
>
>testcharsel and testgtkfont
>error reported:
>-I added _USE_PANGO_BASED_CODE in one line inside gtkfontcombo.c to apeace
the compiler. The testcharsel and testgtkfont programs however are still
affected by the tiny font bug even when not defining _USE_PANGO_BASED_CODE.
I am not surprised since I have not done any work in any of those widgets,
maybe you guys can point me into what needs to be changed there and I can
look at it.
>
>Now, I imagine you guys want to try this out right? :)
>
>Attached should be the tar of my gtkextra modded files. Untar the archive
then before doing anything else I would suggest looking at the code to see
if you agree with the stuff I changed. Simply do "grep _USE_PANGO_BASED_CODE
*" in the gtkextra source directory and that will show you the places where
I have modified the code (including the Makefile). To build you will most
likely have to re-run the configure script so you might want to note the
change in the gtkextra/Makefile and add it again to the resulting Makefile.
I am a newbe at using autoconf so I know the way I placed the
_USE_PANGO_BASED_CODE flag in the Makefile is not ideal but maybe you guys
can fix it ;).
>
>Hope this works, please let me know if there are any issues/suggestions/etc
on improving this "patch". Reading the bug report pretty much confirms that
the problem is not in gtkextra but in Gkt/Pango/Gnome so I'm not sure how we
can fix it.
>
>Suerte!
>-Marco
>
>
>
>
>>On Tue, 2005-02-22 at 19:14 -0500, MQ wrote:
>>
>>
>>>Al,
>>>when you have noticed the tiny font issue, has it been with gtkextra
widgets
>>>or other widgets and gtk based applications also?
>>>
>>>
>>	It's everything using gtk.  I noticed it when I did my first install of
>>Mandrake 10.1 Official.  Lots of text was out of whack, in many
>>gtk-based apps.  That's when I started poking around and found this:
>>
>>http://bugzilla.gnome.org/show_bug.cgi?id=104341
>>
>>	This is the issue we're seeing.  It has nothing to do with GtkExtra
>>directly, but affects it.  My guess (only a guess) is that the "old"
>>approach of using the GdkFont stuff (as gtkextra-1 does) is not affected
>>because that whole set of libraries doesn't pick up the gconf dpi
>>setting.  Pango does, however, so gtkextra-2 is exposed to the problem
>>because of the port to Pango.  However, this is just a guess, I have not
>>had the time to dig down far enough to prove this is the case.
>>
>>	I've been thinking more about this through the day, and shifting my
>>position some, so here's my completely unrequested and free-of-charge
>>opinion...  While it might be nice to hack gtkextra-2 to get around this
>>problem for a while, ultimately it shouldn't be in the gtkextra-2 code.
>>The pressure seems to be building on the Gnome/GTK teams to change GTK
>>so it uses the X server dpi instead of keeping it's own (at least, based
>>on the bug thread referenced above, see the last comment from the
>>Mandrake Cooker group).  The main gtkextra-2 codebase should wait for
>>the root cause bug to be fixed, before anything is done to gtkextra-2
>>itself (at least in the main CVS branch).
>>
>>	There is also a *lot* of testing that would have to be done on
>>gtkextra-2 to know that the changes hadn't introduced other issues,
>>although I'm sure you're doing a good job with the changes, Marco.  At
>>this point, I would have to vote that these kinds of changes don't show
>>up in the main gtkextra-2 codebase.  Adrian, maybe there should be a
>>branch in CVS, if you want to incorporate Marco's work for others to
>>use, that wouldn't affect the gtkextra-2 mainline for the foreseeable
>>future?  Hopefully, before too much of this has to go on, an updated gtk
>>would be released that fixes the underlying problem.  Hey, I can
>>hope....
>>
>>-Al
>>
>>
>>--
>>
>>
>>Al Hooton
>>al-at-hootons-dot-org
>>
>>========
>>Maintainer of the linux-based open-source project Parapin.
>>http://www.sf.net/projects/parapin/
>>========
>>
>>
>>
>>
>>-------------------------------------------------------
>>SF email is sponsored by - The IT Product Guide
>>Read honest & candid reviews on hundreds of IT Products from real users.
>>Discover which products truly live up to the hype. Start reading now.
>>http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
>>_______________________________________________
>>Scigraphica-gtkextra mailing list
>>[email protected]
>>https://lists.sourceforge.net/lists/listinfo/scigraphica-gtkextra
>>
>>



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Scigraphica-gtkextra mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scigraphica-gtkextra



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click