On Thu, May 15, 2008 at 12:54:26AM +0100, Luca Bigliardi wrote:
> On Wed, May 14, 2008 at 08:06 PM, MrGoil wrote:
>
> > May you enlighten me on this:
>
> I think the difference comes from spidermonkey version / xulrunner
> version.
>
> > JavaScript-C 1.6.1 2008-03-01
>
> > Installiert:1.8.0.15~pre080323b-0etch2
>
> > Installiert:1.8.0.15~pre080323b-0etch1
[...]
> JavaScript-C 1.7.0 2007-10-03
[...]
> > also I like to know if you are compiling the spider with -DJS_THREADSAFE
> > support.
>
> It seems defined in xulrunner sources 1.8.1.14.
sorry to say but i need to make a point here: debian is an unreliable
mess when it comes to such versioning schemes (to not mention the
recent openSSL security desaster and such solipsistic decisions taken
by package mantainers). in fact javascript spidermonkey 1.8 *does not
exists* and i cannot understand why the mantainer renamed the package
to "xulrunner" and raised the upstream version number, this makes no
sense under any circumstances.
my final decision on the issue regarding freej is to keep javascript
linked statically inside freej. sorry, but we are writing code and we
cannot rely on misleading bug-reports due to wrong packaging and
such. After running a few tests my conclusion is: if you like to run a
working binary of FreeJ then i do NOT recommend linking dynamically
javascript and DISCOURAGE all dynamic linking in debian and other
distros.
i'll leave the dynamic linking flag in configure, but please use it on
your own risk. feel free to experiment but PLEASE: *** do not include
a freej package in debian that links dynamically to xulrunner unless
you fully tested it under all circumstances ***
that's all for what i'm concerned about this debian packaging issue,
please forward it on the debian bugtracker if necessary.
ciao
- --
Jaromil, dyne.org developer,
http://jaromil.dyne.org
GPG: 779F E8B5 47C7 3A89 4112 64D0 7B64 3184 [ B534 0B5E ]
