[gs-devel] Re: Upcoming Ghostscript 8.60 release

[Ray Johnston]

Otani-san,

We have assigned this issue to Toshiya-san (mpsuzuki at hiroshima-u.ac.jp)
who now works for Artifex under an agreement that allows us to give
him access to all of our test files.

Please coordinate with him on the testing of the cjkv patches. If
you need examples that show the problems, he should be able to come
up with test cases by examining the confidential files.

Thank you very much for your continuing effort on this issue.

Best Regards,
Ray Johnston
------------------------------------------------------------------------

Koji Otani wrote:
> I'm sorry for regressions.
> I want to check these, but I didn't find these test
> files. (spec_gs.pdf, regression test .../comparefiles/*).
> Please tell me where these files are or how to get these files.
> ----
> Koji Otani.
>
> From: Till Kamppeter <till.kamppeter at gmail.com>
> Subject: Re: [gs-devel] Re: Upcoming Ghostscript 8.60 release
> Date: Tue, 14 Aug 2007 22:28:27 +0100
> Message-ID: <46C21E7B.7090905 at gmail.com>
>
> till.kamppeter> Otani-san,
> till.kamppeter> 
> till.kamppeter> we have reverted your patches in the Ghostscript SVN repositories for 
> till.kamppeter> now. They are now attached to the following bug report as proposed patches:
> till.kamppeter> 
> till.kamppeter> http://bugs.ghostscript.com/show_bug.cgi?id=689304
> till.kamppeter> 
> till.kamppeter> Please check the two regressions and post new patches if you have a fix. 
> till.kamppeter> Thank you.
> till.kamppeter> 
> till.kamppeter>     Till
> till.kamppeter> 
> till.kamppeter> 
> till.kamppeter> Till Kamppeter wrote:
> till.kamppeter> > It seems that the CJK patches break rendering of non-CJK documents, see
> till.kamppeter> > 
> till.kamppeter> > http://bugs.ghostscript.com/show_bug.cgi?id=689404
> till.kamppeter> > http://bugs.ghostscript.com/show_bug.cgi?id=689405
> till.kamppeter> > 
> till.kamppeter> > The "rev 8185" mentioned in the bug report is the revision with which 
> till.kamppeter> > the CJK patches were introduced.
> till.kamppeter> > 
> till.kamppeter> > Otani-san and Toratani-san, can you check this? Thanks.
> till.kamppeter> > 
> till.kamppeter> >    Till
> till.kamppeter> > 
> till.kamppeter> > 
> till.kamppeter> > Koji Otani wrote:
> till.kamppeter> >> From: Ralph Giles <giles at ghostscript.com>
> till.kamppeter> >> Subject: Re: [gs-devel] Re: Upcoming Ghostscript 8.60 release
> till.kamppeter> >> Date: Mon, 13 Aug 2007 13:27:37 -0700
> till.kamppeter> >> Message-ID: <20070813202737.GB28970 at ghostscript.com>
> till.kamppeter> >>
> till.kamppeter> >> giles> On Mon, Aug 13, 2007 at 10:53:02AM -0700, Ray Johnston wrote:
> till.kamppeter> >> giles> giles> > Possibly some documentation about what's automatic and 
> till.kamppeter> >> what is not with
> till.kamppeter> >> giles> > the mkromfs and genint processing is needed to make it easier 
> till.kamppeter> >> to know
> till.kamppeter> >> giles> > how to add files.
> till.kamppeter> >> giles> giles> Probably a good idea. The patch didn't just called 
> till.kamppeter> >> runlibfile on the new giles> library files, without an associated 
> till.kamppeter> >> magic '%% Replace' directive. The
> till.kamppeter> >> giles> attached patch does this, and presumedly also works. I tested 
> till.kamppeter> >> tiger, but
> till.kamppeter> >> giles> not the cjkv support.
> till.kamppeter> >> giles>
> till.kamppeter> >> Thank you. I tested this patch with CJK font. It's OK.
> till.kamppeter> >>
> till.kamppeter> >> giles> Questions:
> till.kamppeter> >> giles> giles> 1. I gather from the implementation of translate_path() 
> till.kamppeter> >> in geninit.c giles> that unix-style paths like (cjkv/cjvkinit.c) are 
> till.kamppeter> >> portable postscript
> till.kamppeter> >> giles> under ghostscript. Is this correct? Might be be cleaner to move 
> till.kamppeter> >> them
> till.kamppeter> >> giles> up a level regardless?
> till.kamppeter> >> giles> giles> 2. Inclusion of the module is conditional on the NOCJKV 
> till.kamppeter> >> symbol. I giles> couldn't figure out what this is for. If it's just 
> till.kamppeter> >> for testing, can
> till.kamppeter> >> giles> it be removed? We wish to avoid unnecessary conditionals in the 
> till.kamppeter> >> code
> till.kamppeter> >> giles>
> till.kamppeter> >> This conditionals for analyzing when problems occur.
> till.kamppeter> >> If with defined NOCJKV problems still occur, we can suspect codes not in
> till.kamppeter> >> CJK patch.
> till.kamppeter> >>
> till.kamppeter> >> ---
> till.kamppeter> >> Koji Otani
> till.kamppeter> >>
> till.kamppeter> > 
> till.kamppeter> > 
>
>