Re: [Libcdio-devel] libcdio-2.0.0rc2 (was libcdio-1.1.0 - fix double free error)

[Rocky Bernstein]

On Sat, Dec 30, 2017 at 2:31 PM, Chris Clayton <address@hidden>
wrote:

>
>
> On 30/12/17 03:43, Rocky Bernstein wrote:
> > In my final testing I discovered clang-static-analyzer and made another
> pass at adding free's on error exits, and
> > removing dead code and other lint-like things.
> >
> > As a result there is yet another RC candidate https://rocky.github.io/
> libcdio-2.0.0rc4.tar.bz2
> >
> > Unless I hear of problems I'll release that in a couple of days.
> >
>
> I said I'd build libcdio-2.0.0rc (I've used rc4) and then rebuild and test
> all the packages I have on my system that
> depend on the libraries provided by libcdio. The results are pretty good.
>

Many thanks for testing and the feedback.



> The following applications all build without change and from the testing I
> have done, appear to work fine (although I
> can't claim to have tested them extensively):
>
>         libcdio-paranoia1-10.2+0.94+2   (tested via ripperX)
>         gvfs-1.34.1                     (tested by inserting an audio cd
> and seeing the icon appear on the desktop)
>         mplayer-1.3.0                   (tested by playing an audio cd)
>         kodi-17.6                       (ditto)
>         gstreamer-plugins-ugly-1.12.4   (tested by playing an audio cd
> with gst123)
>         audacious-plugins-3.9           (tested by playing an audio cd
> with audacious)
>         mpv-0.28.0                      (tested by playing an audio cd)
>
> The only package I can't build is vcdimager. I haven't used that (or the
> applications that depend on the library it
> provides) for a long time, so I can't say I'm that bothered.


I can fix vcdimager and I think I still have commit access. I'll do that
though after the 2.0.0 release when I als
start addressing pycdio.

In case anyone is driven to fix it up, the errors I get are:
>
> info.c: In function '_init_segments':
> info.c:109:13: error: too many arguments to function 'iso9660_fs_readdir'
>    entlist = iso9660_fs_readdir(p_obj->img, "SEGMENT", true);
>              ^~~~~~~~~~~~~~~~~~
> In file included from info_private.h:35:0,
>                  from info.c:28:
> /usr/include/cdio/iso9660.h:1060:14: note: declared here
>  CdioList_t * iso9660_fs_readdir (CdIo_t *p_cdio, const char psz_path[]);
>               ^~~~~~~~~~~~~~~~~~
>
> info.c:144:3: error: too few arguments to function '_cdio_list_free'
>    _cdio_list_free (entlist, true);
>    ^~~~~~~~~~~~~~~
> In file included from info_private.h:34:0,
>                  from info.c:28:
> /usr/include/cdio/ds.h:46:6: note: declared here
>  void _cdio_list_free (CdioList_t *p_list, int free_data, CdioDataFree_t
> free_fn);
>       ^~~~~~~~~~~~~~~
>
> info.c: In function 'vcdinfo_visit_lot':
> info.c:1750:5: error: too few arguments to function '_cdio_list_free'
>      _cdio_list_free(p_obj->offset_x_list, true);
>      ^~~~~~~~~~~~~~~
> In file included from info_private.h:34:0,
>                  from info.c:28:
> /usr/include/cdio/ds.h:46:6: note: declared here
>  void _cdio_list_free (CdioList_t *p_list, int free_data, CdioDataFree_t
> free_fn);
>       ^~~~~~~~~~~~~~~
>
> info.c:1753:5: error: too few arguments to function '_cdio_list_free'
>      _cdio_list_free(p_obj->offset_list, true);
>      ^~~~~~~~~~~~~~~
> In file included from info_private.h:34:0,
>                  from info.c:28:
> /usr/include/cdio/ds.h:46:6: note: declared here
>  void _cdio_list_free (CdioList_t *p_list, int free_data, CdioDataFree_t
> free_fn);
>       ^~~~~~~~~~~~~~~
>
> info.c: In function 'vcdinfo_close':
> info.c:2084:7: error: too few arguments to function '_cdio_list_free'
>        _cdio_list_free(p_obj->offset_list, true);
>        ^~~~~~~~~~~~~~~
> In file included from info_private.h:34:0,
>                  from info.c:28:
> /usr/include/cdio/ds.h:46:6: note: declared here
>  void _cdio_list_free (CdioList_t *p_list, int free_data, CdioDataFree_t
> free_fn);
>       ^~~~~~~~~~~~~~~
>
> info.c:2086:7: error: too few arguments to function '_cdio_list_free'
>        _cdio_list_free(p_obj->offset_x_list, true);
>        ^~~~~~~~~~~~~~~
> In file included from info_private.h:34:0,
>                  from info.c:28:
> /usr/include/cdio/ds.h:46:6: note: declared here
>  void _cdio_list_free (CdioList_t *p_list, int free_data, CdioDataFree_t
> free_fn);
>       ^~~~~~~~~~~~~~~
>
> At top level:
> info.c:62:19: warning: '_rcsid' defined but not used
> [-Wunused-const-variable=]
>  static const char _rcsid[] = "$Id: info.c,v 1.38 2005/08/03 08:15:37
> rocky Exp $";
>                    ^~~~~~
> I guess these are the result of the API changes that Rocky mentioned in
> his recent posting.
>
>
> > On Sun, Dec 24, 2017 at 8:31 AM, Rocky Bernstein <address@hidden <mailto:
> address@hidden>> wrote:
> >
> >     I should have also added that the Perl bindings, Device::Cdio  (on
> gtihub it is
> >     https://github.com/rocky/Perl-Devel-Cdio <https://github.com/rocky/
> Perl-Devel-Cdio> ) have now been updated for the
> >     2.0.0 API . I have changed its version number to 2.0.0 to match for
> now.  After the libcdio 2.0.0 release the Perl
> >     package  will be cut and put on CPAN.
> >
> >     I worked on that because there were reports of that getting updated
> recently on Debian.
> >     Since I had to work on the CD-TEXT code, the Perl bindings now more
> fully support support that aspect. The mmc
> >     bindings are still lacking.
> >
> >     After that the I'll work on the Python bindings and probably fill in
> at least the same features as Perl. (n the past
> >     I have heard of projects that have used that.
> >
> >     After that I'll finally work on Ruby. I am not aware of anyone
> recently using that.
> >
> >
> >     On Sun, Dec 24, 2017 at 8:19 AM, Rocky Bernstein <address@hidden
> <mailto:address@hidden>> wrote:
> >
> >
> >
> >         On Sun, Dec 24, 2017 at 5:23 AM, Chris Clayton <
> address@hidden <mailto:address@hidden>> wrote:
> >
> >
> >
> >
> >             I've built RC2 but noticed that the library SO numbers
> haven't been bumped - you highlighted the need for
> >             the bump in
> >             your post to the list at http://lists.gnu.org/archive/
> html/libcdio-devel/2017-12/msg00008.html
> >             <http://lists.gnu.org/archive/html/libcdio-devel/2017-12/
> msg00008.html>. Running "autoreconf -fi"
> >             in the source directory updates the autotools files to
> produce libraries with the correct SO numbers.
> >
> >             You'll be pleased to hear that the build completed
> successfully. I've also rebuilt libcdio-paranoia (version
> >             10.2+0.94+2), audacious and audacious-plugins (both at
> version 3.9). Audacious plays audio cds successfully.
> >
> >             From libcdio itself, cd-info reads an audio cd and, via
> libcddb I assume, shows the correct track
> >             information from the
> >             cdtext.
> >
> >             I'll try to build and test a few more applications that use
> libcdio (gvfs, mplayer, kodi...) after the
> >             holiday and let
> >             you know the outcome.
> >
> >
> >
> >
> >         Thanks again for checking. I have rebuilt the tarball with the
> autoreconf and it is now called RC3 and is
> >         available in https://rocky.github.io/libcdio-2.0.0rc3.tar.bz2 <
> https://rocky.github.io/libcdio-2.0.0rc3.tar.bz2>
> >
> >         I too will be checking on various OS's, but I will wait to hear
> back from you before the final release. Thanks.
> >         <https://rocky.github.io/libcdio-2.0.0rc2.tar.bz2>
> >
> >
> >
>