Re: RFC: gtk-doc and gobject introspection

On Tue, Aug 9, 2011 at 12:28, Stefan Kost <ensonic hora-obscura de> wrote:
On 08/08/11 17:10, Tomeu Vizoso wrote:
> On Sun, Aug 7, 2011 at 14:27, Stefan Kost <ensonic hora-obscura de> wrote:
> For Python, we are extending the API that is provided through g-i via
> "overrides": Python modules that add API on top. So part of the
> sources for the Python API are outside the C library module (right now
> inside pyobject, one day should be split out).
One thing we ned to agree on the community is that do we want all our
api-docs look alike or do we want python docs look like other python
docs etc. In the later case I don't see much things that gtk-doc can
help here with.

I think that, at least, they should be all branded the same way.
(logo, font, header, footer, color scheme, etc.)

Then I would tend to think that, given one language, the developer would like to have all documentation organized the same way (so generated by the same tool)

What do you think?
> I was expecting that g-ir-parser would be able to put in the .gir
> files all the information that gtk-doc requires, being able to drop
> the overlapping code in gtk-doc that parses the sources and thus
> having gtk-doc using .gir files as input.
If nobody minds that gtk-doc will need to depend on a perl-xml parser,
that could be done in the long term. But as I tried to docuemtn, there
is still information that is not in the gir files (like the doc
structure). This can be addressed though.

> Is that not realistic by some reason?
I just hope that this would not complicate builds, as then gtk-doc would
depend on gir files and not all projects that use gtkdoc might want to
use gobject introspection.


> Regards,
> Tomeu

desktop-devel-list mailing list
desktop-devel-list gnome org

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