I am +1 for this switch, but I'm definitely biased by working at Google. My thoughts:
tut definitely gets the job done, but I found it a bit awkward too. But I think the updates and the additional features of GoogleTest are probably worth it. I especially like the distinction between ASSERT.* that immediately stops a test when it finds and issue and EXPECT that lets the test move forward until the end of that test so that I can see all of the results of EXPECTS. The formerly separate "mock" capabilities are really handy too. I wrote a bunch of C++ GDAL tests using GoogleTests in the 2013-207 time frame (Even used a bit of that work as starters to the tests in PROJ, but he went way beyond what I had). I have a lot of the tests in github, but they are written against very old versions of GDAL. And, the GoogleTest ability to work with newer versions of C++ has gotten a lot stronger. C++17 wasn't even available when I wrote a lot of the test code. We have been upgrading GDAL and improving the tests, so if there is interest, I can try to do some updates to the repo. If any of the autotest2 code is used, the license should be switched from Apache 2.0 to the MIT style license mentioned here <https://github.com/OSGeo/gdal/blob/a394f9cb299b2c3c2159098483d1fece3a464fda/LICENSE.TXT#L15> . https://github.com/schwehr/gdal-autotest2/tree/master/cpp -Kurt On Wed, Nov 16, 2022 at 10:42 AM Even Rouault <even.roua...@spatialys.com> wrote: > Hi, > > As this is RFC season. I've prepared RFC88: Use GoogleTest framework for > C/C++ unit tests > > Text at https://github.com/OSGeo/gdal/pull/6720 > > Summary: > > The document proposes and describes conversion of the existing C/C++ > autotest suite to use the `GoogleTest > framework <https://github.com/google/googletest>`__. > > GoogleTest is a popular and maintained framework for C/C++ test > writing, that is a better replacement for the `TUT framework > <https://github.com/mrzechonek/tut-framework>`__ that we use currently. > > Even > > -- > http://www.spatialys.com > My software is free, but my time generally not. > > _______________________________________________ > gdal-dev mailing list > gdal-dev@lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/gdal-dev >
_______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev
