nanogui: fonts in nanox


Previous by date: 7 Dec 1999 13:15:49 -0000 Re: X11 compatibility and pseudo-keyboards., Greg Haerr
Next by date: 7 Dec 1999 13:15:49 -0000 Re: X11 compatibility and pseudo-keyboards., Chris Ross
Previous in thread: 7 Dec 1999 13:15:49 -0000 Re: fonts in nanox, Greg Haerr
Next in thread: 7 Dec 1999 13:15:49 -0000 Re: fonts in nanox, Darran D. Rimron

Subject: RE: fonts in nanox
From: Chris Ross ####@####.####
Date: 7 Dec 1999 13:15:49 -0000
Message-Id: <Pine.LNX.4.10.9912071306050.24428-100000@vortex.ukshells.co.uk>


======================================================
| Chris Ross ( Boris` )         ####@####.#### |
|                          http://www.darkrock.co.uk |
`----------------------------------------------------' 

On Mon, 6 Dec 1999, Greg Haerr wrote:

> On Monday, December 06, 1999 3:15 AM, Chris Ross  wrote:
> : i have been playing about with nanox some more, and i have
> : been thinking it would be nice to have some sort fo dynamic 
> : fontness. this is how i see we do it:
> : 
> : the client requests a font
> :  - whether that bea wild stab in the dark or from
> :    a list obtained from the server....
> : 
> : the server looks in it's font cache and say hrmm do i 
> : have this loaded?
> :  - yes, update last accessed time
> :  - no, load font into memory
> : 
> Yes, this is relatively easy to do, but a couple things need to happen:
> 
> 	o most important, we need a standard font naming scheme.  Yes,
> this could be kluged into just passing the filename (without path) for the
> time
> being, but then there's no standard way to name fonts.
> 
> 	o we add another couple entry points in the screen driver interface,
> LoadFont()
> and UnloadFont.  These routines just read a disk file in the internal format
> into 
> an alloc'd area of the screen driver, and return.
> 
> 	Currently, the font is passed as a number (0-MAXFONT) that is
> defined in device.h.  The additional fonts would be added above these
> compiled-
> in numbers.
> 

i think we should use a system similar to x.,,...? would this be okay?


> 	Microwindows currently doesn't depend on any operating system
> support
> for file i/o.  So this option should likely remain a compile-time support
> option, for 
> operating systems without file i/o or filesystems.
> 

true - but we could have a devfileio.c which has hooks for fs io
so os's that support itcan, and those that can't, dont. 
for those that can't we could compile in certain limited files.

> 	o you might want to make sure that we have converters for all your
> favorite fonts.  I used Ben Pfaff's converter for bdf files, and wrote one
> for 
> any windows truetype of bitmapped font.  These converters create in-core
> font structures in C to be linked in.
> 

i'm going to take a look at the xfs andwrite a curt down version
to link into the nanox server and microwindows - i will add support
for bitmapped fonts etc thats allready there...

> 
> 
> 
> : i then think it would be possible to unload fonts  that
> : haven' been used for a while.
> 
> 	Yep. I propose that the application that calls GrLoadFont also
> call GrUnloadFont.
> 

 we could make it so that when the app
requests a font the server loads it automatically, that way
the app need not know how to load and unload fonts....

> : 
> : if this hasn't been planned to be done anyone, i'm happy
> : to start work on it. 
> 
> Go ahead and take a hack at it.  If you need help, let me know.
> 
> Greg
> 
> 

will do.


Previous by date: 7 Dec 1999 13:15:49 -0000 Re: X11 compatibility and pseudo-keyboards., Greg Haerr
Next by date: 7 Dec 1999 13:15:49 -0000 Re: X11 compatibility and pseudo-keyboards., Chris Ross
Previous in thread: 7 Dec 1999 13:15:49 -0000 Re: fonts in nanox, Greg Haerr
Next in thread: 7 Dec 1999 13:15:49 -0000 Re: fonts in nanox, Darran D. Rimron


Powered by ezmlm-browse 0.20.