ntg-context - mailing list for ConTeXt users
 help / color / mirror / Atom feed
From: Mojca Miklavec <mojca.miklavec.lists@gmail.com>
To: Marco Pessotto <melmothx@gmail.com>
Cc: mailing list for ConTeXt users <ntg-context@ntg.nl>,
	Peter Breitenlohner <peb@mppmu.mpg.de>,
	Taco Hoekwater <taco@elvenkind.com>,
	Hartmut Henkel <hartmut_henkel@gmx.de>,
	Karl Berry <karl@freefriends.org>
Subject: Re: Testing TeX Live 2012
Date: Fri, 18 May 2012 19:20:11 +0200	[thread overview]
Message-ID: <CALBOmsatkLYm_mBT4GNwxXGb=BPWEOgMoTrc4JuwOTr0M3FhRw@mail.gmail.com> (raw)
In-Reply-To: <874nrdctrm.fsf@universe.krase.net>

On Fri, May 18, 2012 at 6:56 PM, Marco Pessotto wrote:
> Taco Hoekwater <taco@elvenkind.com> writes:
>
>> I guessed as much. Unfortunately, as I cannot reproduce the bug, this is
>> where it ends for me. Perhaps Hartmut has a 32-bit system somewhere (CC).
>>
>> Best wishes,
>> Taco
>>
>> PS Hartmut: full thread is here:
>> http://archive.contextgarden.net/message/20120516.012657.49ebfe34.en.html
>
> Just for the sake of giving more (probably useless) information:
>
>  1. I updated the sources today and rebuild luatex (0.70.2). Same crash
>  (not that I thought it would be magically fixed by itself, but just to
>  be sure).
>
>  2. I diffed the luatex sources of texlive-2011 with the pretest one,
>  and are just a bunch of chunks. Now, I don't know anything about C, but
>  it doesn't seem a radical change.

Taco didn't commit any update at all, except for version change (and
Khaled did some math-related updates recently).

> So I thought that maybe it's poppler
>  the culprit, and I tried to rebuild with the sistem poppler. But here
>  it fails miserably during compilation.

You can try to checkout revision 26300 from TeX Live repository. That
should give you a good reference.
    http://tug.org/svn/texlive?view=revision&revision=26301

> Now, shooting in the dark: the sources of TeXlive ship poppler 0.20.
> I compiled and tested the svn trunk of luatex, and it works (no crash,
> as reported by Luigi), but it's linked against poppler 0.18, as far as I
> can see.

You are not shooting in the dark. This was my earlier guess as well.
It's crashing while dealing with pdf and basically the only relevant
change between TL 2011 & 12 is a different poppler version.

> Well, that was all I could do, please let me know if I can do anything
> else.

If you could also try to build revision 26300 with debugging symbols
(and possibly try to upload the binary somewhere, so that others could
try as well), that might help. I either have or can try to install a
virtual machine with i386-linux, but building TeX Live in virtual
machine will take forever, so having a working binary with all
debugging symbols could help. (I'm not 100% sure if the binary for mac
has been compiled before or after poppler update.)

If you like debugging, you could set a few breakpoints in the
following lines that work with poststruct:

./texk/web2c/luatexdir/lua/lpdflib.c:        pdf_set_pos(static_pdf,
static_pdf->posstruct->pos);
./texk/web2c/luatexdir/lua/lpdflib.c:        (void)
calc_pdfpos(static_pdf->pstruct, static_pdf->posstruct->pos);
./texk/web2c/luatexdir/lua/lpdflib.c:        (void)
calc_pdfpos(static_pdf->pstruct, static_pdf->posstruct->pos);
./texk/web2c/luatexdir/lua/lpdflib.c:                lua_pushnumber(L,
static_pdf->posstruct->pos.h);
./texk/web2c/luatexdir/lua/lpdflib.c:                lua_pushnumber(L,
static_pdf->posstruct->pos.v);
./texk/web2c/synctexdir/synctex-luatex.h:#define SYNCTEX_CURV
(dimen_par(page_height_code)-static_pdf->posstruct->pos.v)
./texk/web2c/synctexdir/synctex-luatex.h:#define SYNCTEX_CURH
static_pdf->posstruct->pos.h

and compare behaviour of old & new luatex. But that's just a blind guess.

Mojca
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

  reply	other threads:[~2012-05-18 17:20 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-16  1:26 Mojca Miklavec
2012-05-16 19:02 ` Marco Pessotto
2012-05-17 12:00   ` Hans Hagen
2012-05-17 12:14     ` Mojca Miklavec
2012-05-17 12:37     ` Marco Pessotto
2012-05-17 14:30       ` Mojca Miklavec
2012-05-17 15:44         ` Marco Pessotto
2012-05-17 15:55           ` luigi scarso
2012-05-17 16:25         ` Marco Pessotto
2012-05-17 17:01           ` luigi scarso
2012-05-17 17:47             ` luigi scarso
     [not found]           ` <CALBOmsY0GyDb1LDv-ENej97wQjO_cQuUP7rj3nTdvdAO-Awr5A@mail.gmail.com>
2012-05-17 17:31             ` Marco Pessotto
2012-05-18  9:23               ` Taco Hoekwater
2012-05-18 12:49                 ` Marco Pessotto
2012-05-18 14:20                   ` Taco Hoekwater
2012-05-18 16:56                     ` Marco Pessotto
2012-05-18 17:20                       ` Mojca Miklavec [this message]
2012-05-18 18:20                         ` Marco Pessotto
2012-05-21 15:23                           ` Taco Hoekwater
2012-05-21 15:54                             ` Marco Pessotto
2012-05-21 19:22                               ` Marco Pessotto
2012-05-21 19:45                                 ` Henning Hraban Ramm
2012-05-22  8:09                                 ` Taco Hoekwater
2012-05-22  8:25                                   ` Marco Pessotto
2012-05-22  8:31                                     ` Marco Pessotto
2012-05-22  9:13                                       ` Taco Hoekwater

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='CALBOmsatkLYm_mBT4GNwxXGb=BPWEOgMoTrc4JuwOTr0M3FhRw@mail.gmail.com' \
    --to=mojca.miklavec.lists@gmail.com \
    --cc=hartmut_henkel@gmx.de \
    --cc=karl@freefriends.org \
    --cc=melmothx@gmail.com \
    --cc=ntg-context@ntg.nl \
    --cc=peb@mppmu.mpg.de \
    --cc=taco@elvenkind.com \
    /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;
as well as URLs for NNTP newsgroup(s).