[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: RFS: libharu (updated)

On Fri, Jun 18, 2010 at 10:27 PM, Johan Van de Wauw
<johan.vandewauw@gmail.com> wrote:

> 1) the source package contains bindings for python, pascal, freebasic,
> ruby and C#. Since my knowledge of those languages is too small to
> support these bindings, I chose not to package them (yet).

Fair enough.

> 2) Somewhat related: the source package contains a number of sample
> programs. I was unsure what I should do with them. Package them as
> documentation in the -dev package?  Or a seperate documentation
> package? Packaging them as a binary package does not seem useful.

Put them in libharu-dev.examples so they'll be installed in

> 3) I used git-buildpackage to create this package. I hope someone can
> create a repository on alioth (even if it is not sponsored  I hope my
> packaging efforts can be stored on a more reliable place than my pc).

Probably you want collab-maint, check the first paragraph here:


>> Looking at the upstream download directory, I'd strongly recommend
>> they switch to a consistent naming scheme.
> I agree, but at least their current build system (make dist) wiil give
> names consistent with the watch file and the last release.

Ok, great.

>> The last release was a long time ago and the git repository has recent
>> commits, you might want to ask them when the next release will be.
> I did. However I received no answer, and my messages to their
> mailinglist didn't even pass moderation. I have chosen to package the
> current release, but I backported some patches to this build. The
> major new functionality that would be part of the release is imho not
> very relevant for most users (I have never seen pdf files with u3d
> annotations). If they decide to release anyway, most of the packaging
> can be reused.


>> After modifying the packaging to add -Wall to CFLAGS, one gcc warning:
>> hpdf_u3d.c: In function 'HPDF_U3D_LoadU3D':
>> hpdf_u3d.c:136: warning: 'type' may be used uninitialized in this function
> I checked the code, and 'type' is always initialized when it is used.

Hmmm, ok.



Reply to: