[prev in list] [next in list] [prev in thread] [next in thread] 

List:       koffice
Subject:    Re: Font problem, possible solution
From:       Jeffrey Boser <verin () lvcm ! com>
Date:       2000-11-03 14:29:45
[Download RAW message or body]

So Qt is the one generating the bad postscript code?  That doesn't sound
right.  I didn't think qt had postscript generation in it.  But even if
QT did, then that basicly means kde is screwed until another version of
Qt comes out, which may take forever to solve this.

So basicly, we need kword to do it, and guessing is NOT a solution at
all feasable.  We can't hand-correct this situation, every bad font
name meaning we have to print to file, edit the file by hand, etc.

What we need is to use is a resource to provide the correct names, exactly
like sketch does.  And we can either use sketch's already-in-place
mechanism (and thus interoperate too), or make kword's own.  And thats
not very viable.

....jeff

On Fri, 3 Nov 2000, Donovan Rebbechi wrote:
> IMO, this kind of matching really needs to be handled in Qt ( which
> already tries to match names, but uses a sort of guesswork to do it ), not
> Koffice.
> 
> On Fri, 3 Nov 2000, Jeffrey Boser wrote:
> > I know its hard to match up fonts, between X font names and postscript
> > names for example, but there is already one program out there that has
> > a way of handling this, Sketch.
> > 
> > Sketch uses 'std.sfd' files in all font directories that need matching,
> > which contains info such as these two names.  So all sketch needs is to
> > have a list of font directories to scan, and can get the info from
> > these files, and there is no problem printing anymore.  Here's an example
> > line from one of these files:
> > 
> > ZurichBT-RomanCondensed,Zurich,Condensed,-bitstream-Zurich-normal-r-condensed,iso8859-1,0298a___
> > 
> > Basicly, thats the postscript name, the X name, the style name, XLFD's
> > info, and the file name (without prefixes) in case .afm files are needed.
> > 
> > If you contact sketch's author, you can probably get permission to use
> > this info, and add an environment variable for the font paths for kword
> > to use, and the problem should go away.  Sketch even includes a script
> > that generates these files, along with postscripts Fontmap and X's files
> > as well.

[prev in list] [next in list] [prev in thread] [next in thread]