sha,message,author_date,committer_date,raw_author,raw_committer,repo,author,committer
93a4039f6c6eb765f5b2dc1ba286b263a931dac6,"Use correct dtype for RGB image alpha channel (#1893)

* Fix alpha channel logic for integer RGB images

* Use named argument for concat axis",2018-02-12T22:12:12Z,2018-02-12T22:12:12Z,3ce4a2c03644349db2459c99f8b3abfb25e90fed,b0f4ce5150c3c23b1766db99ba73f575ddfc0e71,13221727,12229877,10050469
65e5f05938dc40c6e169377f8c0b6e7774d96866,"Read small integers as float32, not float64 (#1840)

* Read small integers as float32, not float64

AKA the ""I just wasted 4.6 TB of memory"" patch.

* Tests now include a clean flake8 run",2018-01-23T20:15:28Z,2018-01-23T20:15:28Z,3ce4a2c03644349db2459c99f8b3abfb25e90fed,5f199557d0f8f69fbea5e027a407146e2669a812,13221727,12229877,
6aa225f5dae9cc997e232c11a63072923c8c0238,"Normalisation for RGB imshow (#1819)

* Normalisation for RGB imshow

* Add test for error checking",2018-01-19T05:01:06Z,2018-01-19T05:01:06Z,3ce4a2c03644349db2459c99f8b3abfb25e90fed,5f199557d0f8f69fbea5e027a407146e2669a812,13221727,12229877,
0d69bf9dbf281f0f0f48ac2fadda61a82533aac3,"Make `flake8 xarray` pass (#1824)

* autoflake

autoflake --in-place xarray/*.py xarray/*/*.py

* autopep8

autopep8 --in-place --max-line-length 120 --global-config setup.cfg xarray/*.py xarray/*/*.py

* manual checking

* flake8 entire codebase

* fixup test

* Xarray is now flake8-clean with dask settings

* Fix all flake8 issues under default settings",2018-01-14T20:49:20Z,2018-01-14T20:49:20Z,3ce4a2c03644349db2459c99f8b3abfb25e90fed,f10b21bed2846b879806f87039b77245b18e7671,13221727,12229877,1217238
289f95a25c08796532807c669bbb5e12a79270c2,"Support RGB[A] arrays in plot.imshow() (#1796)

* Allow RGB plots from DataArray.plot.imshow

* Allow RGB[A] dim for imshow to be in any order

Includes new `rgb` keyword to tell imshow about that dimension, and much
error handling in inference.

* Use true RGB color for Rasterio gallery page

* Add whats-new entry",2018-01-11T03:14:35Z,2018-01-11T03:14:35Z,3ce4a2c03644349db2459c99f8b3abfb25e90fed,f10b21bed2846b879806f87039b77245b18e7671,13221727,12229877,1217238
cb161a1fbc9771d30ff4a7fc7e7b51a14122ab42,"Plot nans (#1782)

* Fix formatting

* Cross-reference matplotlib docs for plot methods

* Handle all-NaN data in plot.imshow

* Handle array dim with size one in plt.imshow

* Add whats-new entry",2017-12-15T17:31:38Z,2017-12-15T17:31:38Z,3ce4a2c03644349db2459c99f8b3abfb25e90fed,f10b21bed2846b879806f87039b77245b18e7671,13221727,12229877,1217238
d08efaf902cae5e5f28afff7d6f8182e35a53f46,"Shorter repr for attributes (#1322)

* Shorter repr for attributes

NetCDF files often have tens of attributes, including multi-paragraph
summaries or the full modification history of the file.  It's great to
have this available in the .attrs, but we can truncate it substantially
in the repr!  Hopefully this will stop people writing `data.attrs = {}`
and discarding metadata in interactive workflows for the sake of cleaner
output.

* Add test for attribute repr",2017-04-03T00:47:44Z,2017-04-03T00:47:44Z,3ce4a2c03644349db2459c99f8b3abfb25e90fed,f10b21bed2846b879806f87039b77245b18e7671,13221727,12229877,1217238