[tin-dev] [tin 2.4.3] snapshots - please test

[Corinna Vinschen]

On Jul  3 17:13, Urs Janßen wrote:
> In <mailman.171.1530627914.592.tin-dev at tin.org>, Corinna Vinschen wrote:
> > Just built on Cygwin.  Looks good.  Apart from a few "unused variable"
> 
> Thanks, for the feedback, those are "normal" and depend on
> --enable-nntp or --enable-nntp-only.

I guessed so.

> > and "unorm_normalize_62 is deprecated" warnings,
> 
> this is on the TODO list
> 
> | o  add check for ICU unorm2.h/unorm2_normalize(); prefer it over
> |    ICU unorm.h/unorm_normalize()
> 
> volunteers? ,-)

Hmm, the call has changed significantly.  Not that easy...

> > there's one which is ...interesting:
> > 
> > pcre/pcre_study.c:419:35: warning: ‘<<’ in boolean context, did you mean ‘<’ ? [-Wint-in-bool-context]
> >              if ((tcode[c/8] && (1 << (c&7))) != 0)
> >                                 ~~~^~~~~~~~~
> > 
> > The && looks wrong.  In the equivalent piece of code in upstream
> > pcre-8.42 this reads:
> > 
> >              if ((map[c/8] & (1 << (c&7))) != 0)
> 
> the included pcre source is very outdated (> 10 years old now) and should
> either be updated or dropped (at least any system wide installed version
> should be preferred).

Dropping the included pcre would be preferrable, otherwise you get
outdated code by default.  It's even dangerous as fallback.  Better to
bail out from configure if the system has no libpcre installed.

> 
> | o  clean up included pcre stuff and update to >= pcre-8.42
> |
> | o  and pcre2 support (new API)
> |    <http://www.pcre.org/current/doc/html/>
> 
> But I'll fix that in the code. Thanks for the hint.
>  
> > For the Cygwin release I will switch to --with-pcre for the time being.
> 
> This is highly recommended.

Thanks,
Corinna

-- 
Corinna Vinschen
Cygwin Maintainer
Red Hat
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://lists.tin.org/pipermail/tin-dev/attachments/20180703/b6dfc361/attachment.sig>