pull_requests: 77031572
This data as json
| id | node_id | number | state | locked | title | user | body | created_at | updated_at | closed_at | merged_at | merge_commit_sha | assignee | milestone | draft | head | base | author_association | auto_merge | repo | url | merged_by |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 77031572 | MDExOlB1bGxSZXF1ZXN0NzcwMzE1NzI= | 895 | closed | 0 | Tweaks for opening datasets | 885575 | I tweaked the open_dataset() and open_mfdataset() functions for better performance with the PyNIO engine. 1. I use a lot of HDF4 files with what I guess I'll call "malformed names," where the file name does not end with ".hdf". PyNIO is able to figure out the format of the file, but it prints an annoying warning. As far as I can tell, the only way to shut off the warning is to tell it the format of the file, which I've added a `format` option for. 2. I added an option called `only_variables` which specifies which variables to load from the dataset in the event you don't want to load all variables. Say, for example, I have a dataset with 47 variables in it, but I only need 3 of them. If the data are not cached, then only loading the 3 I need cuts the I/O time in half. If they _are_ cached, then loading only the 3 takes 20% of the time to load the full dataset. The `only_variables` option behaves pretty similarly to `drop_variables`. The default is to load all variables. | 2016-07-11T22:08:05Z | 2020-11-08T20:18:01Z | 2020-11-08T20:18:01Z | 6045f00d3078bb21137b6b1a7c05112881db9d83 | 0 | 18ed2e1876c108cbca4e7ef327c58366fc4104d5 | 7bf9df9d75c40bcbf2dd28c47204529a76561a3f | NONE | 13221727 | https://github.com/pydata/xarray/pull/895 |
Links from other tables
- 3 rows from pull_requests_id in labels_pull_requests