[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [aspell-devel] Makefile for Aspell / MinGW-gcc3.2
From: |
Kevin Atkinson |
Subject: |
Re: [aspell-devel] Makefile for Aspell / MinGW-gcc3.2 |
Date: |
Mon, 7 Oct 2002 14:33:58 -0400 (EDT) |
On Mon, 7 Oct 2002, Thorsten Maerz wrote:
> Hi all,
>
> I wrote a plain makefile to work around the libtool issues when compiling
> Aspell using MinGW GCC, and it seems to create working executables
> (i.e. "aspell check foo.txt" works perfectly).
>
> In case of interest, the makefile is contained in aspell-mingw.diff (take
> a look at the Readme.txt for further instructions). You can get it from
> http://ezytools.sourceforge.net/aspell/
>
> Problem 1: WIN32_RELOCATABLE
> I didnt include the /bin dir detection patch (posted from Ruurd Reimtsa),
> as it seemed to work only when calling "aspell config", but failed when
> using "aspell check foo.txt". In that case, config.cpp/Config::Config()
> is called at least twice (found out by inserting a printf), and the
> retrieve_bool() function bails out the second time. It turned out, that
> calling have("set-prefix") returns "false" then.
> As I am *very* new to aspell (regarding usage and sourcecode), I havent
> been able to find out more on this. Does anyone have an idea to get this
> working as expected?
The WIN32_RELOCATABLE has not been testes since Aspell 0.33. A lot has
changed then. I will look into this if I get the time. But no promises.
> Problem 2: Calling new_aspell_speller()
> I am trying to link aspell to a Gtk-1.3 based application (Sylpheed-Claws).
> I can compile it using MSVC as well as with MinGW, but both versions
> show the same strange behaviour: The list of available dictionaries can
> successfully be received, but the call to new_aspell_speller() crashes.
> While the MSVC version just dies instead of falling back to the debugger
> (goodbye backtrace...), the gcc compiled version falls back to gdb and
> shows that it stopped at msvcrt_a_iname().
> I am currently unsure if this is caused by an incorrect compilation of
> Aspell (that's why I said "seems to work" some lines above) or by the
> calling application (e.g. a missing or wrong passed parameter).
> Has anyone encountered (solved?) similar problems?
Please compile with '-g' and without optimizations in order to get a
meaningful backtrace. Then post it it here (to aspell-devel) and maybe
I can figure out what the problem is.
BTW: Will cygwin not work for you? I believe people have been able to get
Aspell working under cygwin.
--
http://kevin.atkinson.dhs.org