Re: gEDA-dev: Making a windows binary available

[Peter Clifton <pcjc2-at-NOSPAMcam_dot_ac_dot_uk>]

On Tue, 2008-11-25 at 21:05 -0500, Ales Hvezda wrote:
> Hi PeterC,
> 
> [snip]
> >1. Don't bother - don't make it widely available, and hope the fact it 
> >   announces 1.5.0 as a version (for my git HEAD + ... + 
> >   cairo_experiment branch code) won't cause too much grief with a 
> >   restricted distribution.
> >
> >2. Leave the version string as is (I'm not sure if we have 
> >   infrastructure to add a git SHA1 to the version?), and patch the
> >   code which prints in the log, and the about box, to make clear what 
> >   the release really is.
> 
> 	I do not really care much for the above two options as then the
> poor user doesn't know what version they are running and you won't either.

Ok, I'll do something other than the above when I make a build properly
available.

(You saw the 'CC list of the few people who may have a copy of my test
installer. That is clearly labeled as "geda-1.5.0+git
+cairo_(ALPHA).exe", although the gEDA software inside bears the 1.5.0
version straight from the git repository).

> >3. Try to add something to the version string: 
> >   1.5.0.20080706~79384bb59399cb084c6d6ccebe9725080d363c99
> 
> 	Yikes!
> 
> >
> >   (Perhaps truncating the SHA1 at some point)
> >
> >   or:
> >
> >   1.5.0.20080706~pcjc2.cairo_experiment.release1
> 
> That might be okay.  You can achieve a custom branding/version of gEDA/gaf
> by going into libgeda/include/defines.h and modifying (line 46):
> 
> #define PREPEND_VERSION_STRING ""
>
> Please define a string that includes at least: "pcjc2". 

I don't think it is particularly nice to prepend something to the
version string. (Append would be nicer), but I guess since distro's
requiring an incremental sorting of version numbers will never get this,
it doesn't really matter.

Personally, I'd prefer 1.5.0.20080706~pcjc2-cairo... or something.

I'll try to dig up what I did for a CUED customised release. I think I
edited the code to support appending a string.

If I statically link libgeda (which I believe Cesar's scripts do by
default), that skirts round the possibility of clashing libgeda SO
versions as well.

> Also, I would really prefer it the w32 releases be hosted on server
> that you control or, as you suggested, given out on-demand.

I'm not certain what you mean by that. I don't control _any_ servers, so
to the letter, that requirement is difficult.

I have my CUED webspace (www2.eng.cam.ac.uk/~pcjc2/...), but I don't
have control such that I could get any download stats, for example.

Are you just saying that you don't want to host W32 installers on our
piece of sourceforge.net or geda.seul.org?

If so, is that because less officially hosted binaries can go away, or
be disclaimed as unsupported easier, or is there some "Linux only"
politics at SEUL. (e.g. Given the acronym "Simple End User Linux").

Regards,

-- 
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)



_______________________________________________
geda-dev mailing list
geda-dev@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev