Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Doug Evans <dje@google.com>
To: Joel Brobecker <brobecker@adacore.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [RFC] --with-iconv-path
Date: Fri, 06 May 2011 15:33:00 -0000	[thread overview]
Message-ID: <BANLkTikT53VF-UbvdzvxZQAe3u1hxPzrbQ@mail.gmail.com> (raw)
In-Reply-To: <20110506151212.GA19356@adacore.com>

On Fri, May 6, 2011 at 8:12 AM, Joel Brobecker <brobecker@adacore.com> wrote:
>> We have one environment where the iconv binary comes from a
>> non-standard location.  This patch lets one tell gdb where to find it.
>
> Does it mean that the libiconv libraries are at one location, and
> that the binary is at a different location? Looking at your patch,
> it looks like you might have a situation where the name of the iconv
> binary itself is different...

There is no libiconv library, there's just iconv_{open,close,} in
glibc and the iconv binary.

>> 2011-05-05  Doug Evans  <dje@google.com>
>>
>>       * configure.ac: New configure option --with-iconv-path.
>>       * configure: Regenerate.
>>       * config.in: Regenerate.
>>       * charset.c (find_charset_names): Use ICONV_PATH if defined.
>
> "path" had me confused, and made me think that you were going
> to pass the name of the directory where iconv is located. But
> the patch seems to be contradicting that.
>
> If we don't need to specify the name of the iconv directory,
> then perhaps we could just focus on the directory name itself.
> We could even make that directory name relatocatable, the same
> way we make some of the paths relocatable as well.
>
> Normally, we have sets of switches that look like this:
>
>        --with-xxx=PATH
>        --with-xxx-lib=PATH
>        --with-xxx-include=PATH
>        etc
>
> We already have --with-libiconv-prefix.  So, how about naming it
> --with-libiconv-bin ?
>
> If we think that it would be convenient to specify the iconv exe
> as well, how about --with-libiconv-exe or --with-libiconv-program?

--with-libiconv-{bin,exe} "works for me"

For reference sake, the binary lives in the expected location given
--with-libiconv-prefix (i.e., bin/iconv), but how do I effect the
necessary change without breaking anything.

If folks are happy with adding $WITH_LIBICONV_PREFIX/bin to the front
of the search path instead (so no new configure option), I can do
that.
Seems reasonable, but I guess it *could* break someone's installation somewhere.
OTOH, if people are passing --with-libiconv-prefix they may have
iconvlist() in which case the iconv binary isn't used, so we're safe.


  reply	other threads:[~2011-05-06 15:33 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-06  0:27 Doug Evans
2011-05-06 15:12 ` Joel Brobecker
2011-05-06 15:33   ` Doug Evans [this message]
2011-05-06 16:13     ` Joseph S. Myers
2011-05-06 17:57 ` Tom Tromey

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=BANLkTikT53VF-UbvdzvxZQAe3u1hxPzrbQ@mail.gmail.com \
    --to=dje@google.com \
    --cc=brobecker@adacore.com \
    --cc=gdb-patches@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox