Hi Joris, Really thanks for pointing out where the doc sources are! I will start the PR and share it with you so that we can work on it together. You know, I can do the ORC reader & writer with options and you can do the dataset integration that you did.
Best, Ian On Thursday, November 25, 2021, Joris Van den Bossche < jorisvandenboss...@gmail.com> wrote: > Hi Ian, > > Yes, more documentation regarding ORC would be very welcome! I think > your list of missing docs is correct: > > - It's briefly mentioned in the Python API docs > (https://arrow.apache.org/docs/python/api/formats.html#orc-files), but > incomplete > - The C++ reference docs list the OrcFileFormat for the dataset API > (https://arrow.apache.org/docs/cpp/api/dataset.html#_ > CPPv4N5arrow7dataset13OrcFileFormatE), > but not the direct ORC interface (like is done for Parquet at > https://arrow.apache.org/docs/cpp/api/formats.html, for which the > source lives at > https://github.com/apache/arrow/blob/master/docs/source/ > cpp/api/formats.rst) > - There is indeed no user guide. The Parquet python doc page lives at > https://github.com/apache/arrow/blob/master/docs/source/python/parquet.rst > > Best, > Joris > > On Wed, 24 Nov 2021 at 04:55, Ian Joiner <iajoiner...@gmail.com> wrote: > > > > Hi, > > > > Today I found that pretty much none of our ORC-related work (e.g. ORC > > writer in C++ & Python, Arrow Dataset with ORC) has ever been documented. > > This is something we have to fix or users won’t even be aware that ORC > > support exists, let alone how to use it. > > > > From my understanding it seems that we miss the following docs: > > 1. C++ and Python API reference (partially missing) > > 2. User Guide (entirely absent) > > > > As the person who created and self-assigned > > https://issues.apache.org/jira/browse/ARROW-13231 I’d like to spend the > > next a couple of days fixing it. Could you guys please point me towards > > what actually needs to be revised? In particular where is the source of > > https://arrow.apache.org/docs/python/parquet.html ? > > Really thanks! > > > > Ian >
