This data as json
| sha | message | author_date | committer_date | raw_author | raw_committer | repo | author | committer |
|---|---|---|---|---|---|---|---|---|
| 093518207f4a3210ec692308da2b181a646115d6 | Parallel open_mfdataset (#1983) * proof of concept implementation for parallel open using dask.bag * parallel open option in open_mfdataset * use dask delayed, further tests to check that dask arrays make it through * parallel=None by default and automagicalize to True when dask distributed is used * docs and minor improvement to delayed/compute * 500 file max * skip many-files-test on windows * cleanup parallel open a bit * refactor parallel open a bit, fix tests and fixtures * default parallel open when dask is installed + style fixes * fix parallel kwarg * pynio from conda-forge/dev for pynio test suite * manually skip tests * parallel defaults to false | 2018-04-20T12:04:22Z | 2018-04-20T12:04:22Z | 5f199557d0f8f69fbea5e027a407146e2669a812 | cd792325681cbad9f663f2879d8b69f1edbb678f | 13221727 | 19864447 |