[Qgis-developer] automatic sip files generation

Thu Mar 30 12:55:08 PDT 2017

Hi

On 3/30/17 8:31 PM, Tim Sutton wrote:
> Hi
>
>> On 30 Mar 2017, at 5:08 PM, Matthias Kuhn <matthias at opengis.ch
>> <mailto:matthias at opengis.ch>> wrote:
>>
>> Hi Denis,
>>
>> Thanks a lot for finishing this.
>>
>> I am really excited and expect less troubles on the sip front in the
>> long-term.
>>
>> Tim, actually the generated python code now includes python DocStrings,
>> so it should be much easier to inspect an appropriate python
>> documentation directly with `help(QgsVectorLayer)` etc.
>>
>> And to automatically generate an up-to-date python documentation.
>
> This is really great! I guess the next step would be to see if we can
> use autodoc to generate python documentation on the python libs. If
> that works it will go a long way towards bump starting the Python API
> work that Wes and Francisco were going to work on under a small QGIS
> funded project. The project never got out the starting gates because
> we could not come up with an elegant build system.
Having more people improve on this system will be very welcome, there
are plenty of nice little bells and whistles to add!

>
> Might I suggest that we do our python documentation in a sphinx
> friendly way? So for example where you did:
>
> @versionadded 3.0
> in https://github.com/qgis/QGIS/pull/4318/files#diff-356d7cb762f7b9dfa4526f102450a9aaR41
>
> it would become
>
> ..versionadded:: 3.0
>
> See http://www.sphinx-doc.org/en/stable/markup/para.html
>
> Since we are in green fields with python docstrings here it would be
> nice to have everything clean and neat and pythonic out the gate. Or
> are there some restrictions in the sippy annotations that require you
> to use a certain docstring style?
These doc strings are generated from the C++ doc strings. But since they
are a pretty simple syntax, it shouldn't be hard to improve sipify.pl to
also convert a set of doxygen tags to their pythonic equivalent. We
might want to unify the doxygen style as well within this process.

Cheers
Matthias
> Regards
>
> Tim
>
>>
>> Cheers
>> Matthias
>>
>> On 03/30/2017 05:02 PM, Denis Rouzaud wrote:
>>>
>>>
>>> Le jeu. 30 mars 2017 à 16:56, Tim Sutton <tim at kartoza.com
>>> <mailto:tim at kartoza.com>
>>> <mailto:tim at kartoza.com>> a écrit :
>>>
>>>    Hi
>>>
>>>>    On 30 Mar 2017, at 3:18 PM, Denis Rouzaud
>>>> <denis.rouzaud at gmail.com <mailto:denis.rouzaud at gmail.com>
>>>>    <mailto:denis.rouzaud at gmail.com>> wrote:
>>>>
>>>>    Dear all,
>>>>
>>>>    I'd like to raise the attention on a new workflow I just merged to
>>>>    automatically generate SIP files using the header as source only.
>>>>
>>>>    A PR is pending for the documentation, but you can read it here
>>>>    https://github.com/3nids/QGIS-Documentation/blob/6e3a9fb6f98a2869aef27c01b7ab70eb0a4c7556/source/docs/developers_guide/codingstandards.rst#sip-bindings
>>>>    (waiting to be merged).
>>>>
>>>>    For the moment, only 3 files are generated (qgsattributes,
>>>>    qgsfeature and qgsvectorlayer). One can starts adding new files to
>>>>    the process by adapting their header and removing them from
>>>>    python/auto_sip.blacklist.
>>>
>>>    This is really great! A couple of questions:
>>>
>>>    1) the blacklist implies that any new .h files added to the source
>>>    tree will automatically have sip bindings generated unless you
>>>    explicitly blacklist them?
>>>
>>>
>>> As soon as you add the include in python/core/core.sip (or gui,
>>> analysis), the SIP is considered as automatically generated from header
>>> except if you blacklist it.
>>>
>>>
>>>    2) are you thinking to maintain the sip files in Git still or remove
>>>    the from revision in favour of having them as build time artifacts
>>>    only? I ask this because I often direct people to use the sip files
>>>    as a reference guide for the python API if they want to dig under
>>>    the hood a bit more.
>>>
>>>
>>> The goal is to remove the SIP files and have them created on build time.
>>> But:
>>> 1) I except a few errors in the script to arise (so better to have an
>>> easy access to the SIP files)
>>> 2) it needs to reach a significant point of ported headers.
>>>
>>> So, in a somehow close or far future, yes they'll be removed from source
>>> (hence no blacklist anymore only manual files such as
>>> core/conversions.sip).
>>>
>>>
>>> _______________________________________________
>>> Qgis-developer mailing list
>>> Qgis-developer at lists.osgeo.org <mailto:Qgis-developer at lists.osgeo.org>
>>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>>>
>> _______________________________________________
>> Qgis-developer mailing list
>> Qgis-developer at lists.osgeo.org <mailto:Qgis-developer at lists.osgeo.org>
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>
> —
>
>
>
>
>
>
>
>
>
> *Tim Sutton*
>
> *Co-founder:* Kartoza
> *Project chair:* QGIS.org <http://QGIS.org>
>
> Visit http://kartoza.com <http://kartoza.com/> to find out about open
> source:
>
> Desktop GIS programming services
> Geospatial web development
> GIS Training
> Consulting Services
>
> *Skype*: timlinux 
> *IRC:* timlinux on #qgis at freenode.net <http://freenode.net>
>
> Kartoza is a merger between Linfiniti and Afrispatial
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-developer/attachments/20170330/f8029c94/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: KartozaNewLogoThumbnail.jpg
Type: image/jpeg
Size: 6122 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/qgis-developer/attachments/20170330/f8029c94/attachment-0001.jpg>