Hi Grégory, On Fri, Jan 20, 2017 at 8:59 AM, Even Rouault <even.roua...@spatialys.com> wrote:
> On vendredi 20 janvier 2017 05:45:28 CET Grégory Bataille wrote: > > > Hi everyone, > > > > > > I just subscribed to this list. Due to current interest, driven by my > > > current job, I'd like to participate on gdal2tiles. > > > However I'm not sure how to go about it and how active the development > > > still is. > > > I, for example, encountered an issue, that after investigation led me > there > > > https://trac.osgeo.org/gdal/ticket/6057. It's a ticket with a proposed > > > solution and code patch, but that has seen no activity in 18 months. > > > > > > Would you mind advising on the best way to contribute? Is the > > > development/review process, on gdal2tiles still active and is it worth > > > investing time on it? > > > > Hi Grégory, > > > > To be honest, I haven't processed gdal2tiles tickets for years as an > experiment to see how much time it would take for an existing or new > contributor to step up ;-) > > I'd suggest you to submit pull requests on the github mirror so that the > Travis-CI tests are run on your patches. And I'd strongly encourage you to > write new tests in ../autotest/pyscripts/test_gdal2tiles.py for each > issue fixed. > > > > Even > If you want an example of how to speed things up with Python's multiprocessing module, check out https://github.com/mapbox/rio-mbtiles/blob/master/mbtiles/scripts/cli.py. I've found in rio-mbtiles that the small amount of inter-process communication overhead is worth it and multiprocessing is the only good way to achieve CPU intensive concurrency with Python. Yours, -- Sean Gillies
_______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev