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
26657336,MDExOlB1bGxSZXF1ZXN0MjY2NTczMzY=,302,closed,0,Variables no longer conflict if they are broadcast equal and rather are promoted to use common dimensions,1217238,"Fixes #243.

The idea here is that variables should not conflict if they are equal after being broadcast against each other; rather variables should be promoted to the common dimensions. This should resolve a number of annoyances causes by mixing scalar and non-scalar variables.

This PR includes fixes for `concat`, `Dataset.merge` (and thus `Dataset.update` and `Dataset.__setitem__`) and `Dataset`/`DataArray` arithmetic (via `Coordinates.merge`).
",2014-12-29T19:19:42Z,2014-12-29T19:53:14Z,2014-12-29T19:52:57Z,2014-12-29T19:52:57Z,71d8f38d6a179860afcf9dc2243bb424e6f9a0c2,,799013,0,d876417ad7ad44e7d0d18ffcaa3df7e281c23db2,fb9a5dbfbc9e0585f8ee14af42d1fa41d86ef9d7,MEMBER,,13221727,https://github.com/pydata/xarray/pull/302,
26786572,MDExOlB1bGxSZXF1ZXN0MjY3ODY1NzI=,304,closed,0,Switch to use nan-skipping aggregation functions by default and add .median() method,1217238,"TODO:
- ~~update documentation~~ (I'll do this later)
- ~~update minimum required numpy version to 1.9? (so we can use np.nanmedian)~~ (added an informative error message for median)

fixes #209
xref #130
",2015-01-03T20:19:26Z,2015-01-04T16:05:30Z,2015-01-04T16:05:28Z,2015-01-04T16:05:28Z,1c61b5007ed78b9d99a4ab06e724c9244acc46d7,,799013,0,bf299aa6f94833172784652173e91bd7b0ea7814,9ead00b42399b4dbba93b9559bdee6548a5436b8,MEMBER,,13221727,https://github.com/pydata/xarray/pull/304,
26973957,MDExOlB1bGxSZXF1ZXN0MjY5NzM5NTc=,306,closed,0,Fix coercion of numeric strings to objects,1217238,"Fixes #305
",2015-01-07T17:45:23Z,2015-02-23T06:09:10Z,2015-01-07T18:14:31Z,2015-01-07T18:14:31Z,c5805324eafa2804ce3066acb4cbc0577c6548b7,,799013,0,329c7f193344d19bf2938fc37acca5fad6396755,2ba9c321aee791a98f63b2dd17decc8ffb665bea,MEMBER,,13221727,https://github.com/pydata/xarray/pull/306,
27010288,MDExOlB1bGxSZXF1ZXN0MjcwMTAyODg=,307,closed,0,Skip NA in groupby groups,1217238,"This makes xray consistent with the behavior of pandas.
",2015-01-08T06:40:17Z,2015-01-08T06:51:12Z,2015-01-08T06:51:10Z,2015-01-08T06:51:10Z,c7a943335b0b1ac1305ef4e2a85ad915e9d8017c,,799013,0,b8bfc9b1c3d098037a4071fb1bd12d70aaf527b1,546d24049357fc04e995d3ff70344698acf81c1e,MEMBER,,13221727,https://github.com/pydata/xarray/pull/307,
27171205,MDExOlB1bGxSZXF1ZXN0MjcxNzEyMDU=,309,closed,0,Fix typos in docs,2058401,,2015-01-12T01:00:09Z,2015-01-12T01:44:11Z,2015-01-12T01:43:43Z,2015-01-12T01:43:43Z,0cba0f12b8b8c805d26cb242d134485648d5d021,,799013,0,3d216be4eb9e7510b10bc00c18c23dea43026edb,19df8d202b1d8054019e7e42365c67cdde6ff448,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/309,
27392995,MDExOlB1bGxSZXF1ZXN0MjczOTI5OTU=,310,closed,0,More robust CF datetime unit parsing,514053,"This makes it possible to read datasets that don't follow CF datetime conventions perfectly, such as the following example which (surprisingly) comes from NCEP/NCAR (you'd think they would follow CF!)

```
ds = xray.open_dataset('http://thredds.ucar.edu/thredds/dodsC/grib/NCEP/GEFS/Global_1p0deg_Ensemble/members/GEFS_Global_1p0deg_Ensemble_20150114_1200.grib2/GC')
print ds['time'].encoding['units']

u'Hour since 2015-01-14T12:00:00Z'
```
",2015-01-14T23:19:07Z,2015-01-14T23:36:34Z,2015-01-14T23:35:27Z,2015-01-14T23:35:27Z,96f2c394961b37f6e1238539bb254259c543b8ff,1217238,799013,0,d5115cb1947b0679cc9998665a71f5d85e260623,4a4be4ace8f42dbc7c4ab016ab58a46812b37ad1,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/310,
27728932,MDExOlB1bGxSZXF1ZXN0Mjc3Mjg5MzI=,311,closed,0,Bug fix for DataArray.to_dataframe with coords with different dimensions,1217238,,2015-01-21T01:40:06Z,2015-01-21T01:44:29Z,2015-01-21T01:44:28Z,2015-01-21T01:44:28Z,42ee237d34f3054181eebab41bf5f06b3a16882b,,799013,0,afb658fad24a43edcb42ce76a5551c905c1a6baf,575a7117179ce9e0c6c1b928445aeaead5e808d1,MEMBER,,13221727,https://github.com/pydata/xarray/pull/311,
28612773,MDExOlB1bGxSZXF1ZXN0Mjg2MTI3NzM=,312,closed,0,BUG: Fix slicing with negative step size,1217238,"As identified here:
https://github.com/perrette/dimarray/commit/ad4ab4d049f49881b28120d276337b2cab5e4061
",2015-02-04T04:32:07Z,2015-02-04T04:34:46Z,2015-02-04T04:34:39Z,2015-02-04T04:34:39Z,288b83ebba811c8f87868859369ba693624d44a3,,799013,0,6a62642298b1bce93ec444f92cd54ffc37ea193e,299404da3f3be5b562baf350fb178cf9d0b99c63,MEMBER,,13221727,https://github.com/pydata/xarray/pull/312,
28617914,MDExOlB1bGxSZXF1ZXN0Mjg2MTc5MTQ=,313,closed,0,Fix decoding missing coordinates,1217238,"Fixes #308
",2015-02-04T07:19:01Z,2015-02-04T07:21:03Z,2015-02-04T07:21:01Z,2015-02-04T07:21:01Z,61f33f06eaf53b6aabbf509b664414a546ace672,,799013,0,ed4492a39d42752e708c726e2b578a7a449da807,2efda08952b80b43dfcac07dc2b5ae6fff468561,MEMBER,,13221727,https://github.com/pydata/xarray/pull/313,
28783997,MDExOlB1bGxSZXF1ZXN0Mjg3ODM5OTc=,315,closed,0,Bug fix for multidimensional reindex edge case,1217238,,2015-02-06T04:09:09Z,2015-02-06T04:10:23Z,2015-02-06T04:10:21Z,2015-02-06T04:10:21Z,d3aaa5558481a17f2b7631a07aace14d875a05c8,,799013,0,48a004de46fe72e92145109c08b08512dab1c34c,2d954c45b03c3571c2cbf907543bad78e2cd8fbd,MEMBER,,13221727,https://github.com/pydata/xarray/pull/315,
29021638,MDExOlB1bGxSZXF1ZXN0MjkwMjE2Mzg=,317,closed,0,Fall back to netCDF4 if pandas can’t parse a date,141709,"Addresses #316
",2015-02-10T16:31:34Z,2016-04-01T14:23:10Z,2015-02-10T18:37:32Z,2015-02-10T18:37:32Z,6d270b42920ecb46a1f296ceb5705d58edfb175f,,799013,0,b68b03c3b7a2e666f76d96fabd659dd78c5bd693,8f7fc86e014ca38af880476e1467d0b8236ef2e8,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/317,
29033210,MDExOlB1bGxSZXF1ZXN0MjkwMzMyMTA=,318,closed,0,Fix DataArray.loc indexing with Ellipsis: da.loc[...],1217238,,2015-02-10T18:46:37Z,2015-02-10T18:59:32Z,2015-02-10T18:59:31Z,2015-02-10T18:59:31Z,f24994982b98c546657178d536978bbd46c91d43,,799013,0,2a49692917c0368143e145038bb1d7b2ff3b34e8,14de7b991cc136ab932a2fd9ef4ce0a1ef77eea4,MEMBER,,13221727,https://github.com/pydata/xarray/pull/318,
29250720,MDExOlB1bGxSZXF1ZXN0MjkyNTA3MjA=,321,closed,0,Automatic label-based alignment for math and Dataset constructor,1217238,"Fixes #186.

This will be a major breaking change for v0.4. For example, we can now do things like this:

```
In [5]: x = xray.DataArray(range(5), dims='x')

In [6]: x
Out[6]:
<xray.DataArray (x: 5)>
array([0, 1, 2, 3, 4])
Coordinates:
  * x        (x) int64 0 1 2 3 4

In [7]: x[:4] + x[1:]
Out[7]:
<xray.DataArray (x: 3)>
array([2, 4, 6])
Coordinates:
  * x        (x) int64 1 2 3
```
",2015-02-13T09:31:43Z,2015-03-03T06:24:02Z,2015-02-13T22:19:29Z,2015-02-13T22:19:29Z,b3bb2ebeeb9466d9ed2f7c41865b6c6105a10e01,,799013,0,6ccf629511dce9ef3fb6074fe2133d3277b506da,2eb0d96fc9884e8d6355524d5b8eb972f3bdf0b3,MEMBER,,13221727,https://github.com/pydata/xarray/pull/321,
29500687,MDExOlB1bGxSZXF1ZXN0Mjk1MDA2ODc=,322,closed,0,Support reindexing with an optional fill method,1217238,"e.g., pad, backfill or nearest
",2015-02-18T04:32:47Z,2015-02-18T04:42:00Z,2015-02-18T04:41:59Z,2015-02-18T04:41:59Z,824bdcccba8226e3dd50dcf38fe26e3771751fd0,,799013,0,f57a6c825113e0f6480afa88cee012ba17da5f2d,9fbd15d66b54e8ed7cf5e21a1202c8d69777e031,MEMBER,,13221727,https://github.com/pydata/xarray/pull/322,
29595225,MDExOlB1bGxSZXF1ZXN0Mjk1OTUyMjU=,325,closed,0,Rename Dataset.vars -> data_vars and remove deprecated aliases,1217238,"Fixes #293
",2015-02-19T09:01:45Z,2015-02-19T19:31:17Z,2015-02-19T19:31:11Z,2015-02-19T19:31:11Z,5aba479e0294f72ef734f223d3d810ff90d73ff4,,799013,0,fcdbc7901e453740f4d96f4ee6bd905e91444e54,0cc5e23c170ab1bce471b1f3d2e44bea681c32f9,MEMBER,,13221727,https://github.com/pydata/xarray/pull/325,
29671328,MDExOlB1bGxSZXF1ZXN0Mjk2NzEzMjg=,327,closed,0,Cleanly apply generic ndarrays to DataArray.groupby,2002703,"This is special cased for np.ndarrays: applying to DataArrays
is not only inefficient but would also be wrong if the applied
function wanted to change metadata.

Fixes #326 
",2015-02-20T03:47:15Z,2015-02-20T04:41:10Z,2015-02-20T04:41:08Z,2015-02-20T04:41:08Z,13cdb1eb4ac2f7b8ad88be705005dead93ef8356,,799013,0,57ea69c8988baf3d12ac3775e762ff45bed7ee71,3e3c84203c3f20515bb8d21ec72e9dcc0e2a568b,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/327,
29759819,MDExOlB1bGxSZXF1ZXN0Mjk3NTk4MTk=,329,closed,0,Dataset.apply works if func returns like-shaped ndarrays,1217238,"This extends the recent change by @IamJeffG (#327).
",2015-02-21T20:54:00Z,2015-02-23T00:35:25Z,2015-02-23T00:35:23Z,2015-02-23T00:35:23Z,0fcefec07391a9b5fd8d4559fe7c01d624456c59,,799013,0,602a2101d2d832d94e81b5aed3e2bbbe2e24d3eb,3309f62d7a80bfb8c8aa843a7fbbad546087435d,MEMBER,,13221727,https://github.com/pydata/xarray/pull/329,
29783867,MDExOlB1bGxSZXF1ZXN0Mjk3ODM4Njc=,330,closed,0,Improved error handling for datetime decoding errors,1217238,"Fixes #323

We now get an error message with a lovely traceback when opening a dataset with invalid time units. For example:

```
Traceback (most recent call last):
  File ""/Users/shoyer/dev/xray/xray/test/test_conventions.py"", line 340, in test_invalid_units_raises_eagerly
    decode_cf(ds)
  File ""/Users/shoyer/dev/xray/xray/conventions.py"", line 775, in decode_cf
    decode_coords)
  File ""/Users/shoyer/dev/xray/xray/conventions.py"", line 716, in decode_cf_variables
    decode_times=decode_times)
  File ""/Users/shoyer/dev/xray/xray/conventions.py"", line 676, in decode_cf_variable
    data = DecodedCFDatetimeArray(data, units, calendar)
  File ""/Users/shoyer/dev/xray/xray/conventions.py"", line 340, in __init__
    raise ValueError(msg)
ValueError: unable to decode time units 'foobar since 123' with the default calendar. Try opening your dataset with decode_times=False. Full traceback:
Traceback (most recent call last):
  File ""/Users/shoyer/dev/xray/xray/conventions.py"", line 331, in __init__
    decode_cf_datetime(example_value, units, calendar)
  File ""/Users/shoyer/dev/xray/xray/conventions.py"", line 130, in decode_cf_datetime
    delta = _netcdf_to_numpy_timeunit(delta)
  File ""/Users/shoyer/dev/xray/xray/conventions.py"", line 72, in _netcdf_to_numpy_timeunit
    return {'seconds': 's', 'minutes': 'm', 'hours': 'h', 'days': 'D'}[units]
KeyError: 'foobars'
```

Also includes a fix for a datetime decoding issue reported on the mailing list:
https://groups.google.com/forum/#!topic/xray-dev/Sscsw5dQAqQ
",2015-02-23T03:08:40Z,2015-02-25T22:00:36Z,2015-02-23T03:11:03Z,2015-02-23T03:11:03Z,bb0ea831d39f861ab41767681078f90f73e50664,,799013,0,ad346cea7831c6f53215ef60c28d7c17d1860e1d,e82835013595795c50656bc66b586b751227c204,MEMBER,,13221727,https://github.com/pydata/xarray/pull/330,
29787573,MDExOlB1bGxSZXF1ZXN0Mjk3ODc1NzM=,331,closed,0,Documentation updates anticipating v0.4,1217238,,2015-02-23T05:51:42Z,2015-02-23T06:18:39Z,2015-02-23T06:18:35Z,2015-02-23T06:18:35Z,18fd604faf636f7836886029eac3323c73f37552,,799013,0,a8bec91dd56f7c29753eefe56a060ee2b5f70065,102ade4ba851f5d6e0e40b6e2777bccacbc9b490,MEMBER,,13221727,https://github.com/pydata/xarray/pull/331,
29837685,MDExOlB1bGxSZXF1ZXN0Mjk4Mzc2ODU=,332,closed,0,Update time.season to use text labels like 'DJF',1217238,"Previously, I used numbers 1 through 4 for the sake of consistency with
pandas, but such labels really were impossible to keep track of.
",2015-02-23T19:31:58Z,2015-02-23T19:43:41Z,2015-02-23T19:43:39Z,2015-02-23T19:43:39Z,a1f8e398dc4a6132c0ceeed5049afaf33944b162,,799013,0,4f53c31e681fb2835886bc8a250ba0498c257d4d,ea907af0ca5b8ee7c63b77581705827cd229ba55,MEMBER,,13221727,https://github.com/pydata/xarray/pull/332,
29863225,MDExOlB1bGxSZXF1ZXN0Mjk4NjMyMjU=,333,closed,0,Unify netCDF4 and scipy backends in the public API,1217238,"Fixes #273
and half of #272

To serialize a dataset to a string/bytes, simply use `ds.to_netcdf()`. This behavior copies `DataFrame.to_csv()` from pandas. The legacy `dump` and `dumps` methods are deprecated.

My main concern is that the default ""format"" option is depends on what dependencies the user has installed or if they are saving a file. That seems non-ideal, but may perhaps be the most pragmatic choice given the limitations of the netCDF4 format.

This change also adds:
- Support for writing datasets to a particular NETCDF4 group
- Support for opening netCDF3 files from disk even without netCDF4-python if scipy is installed.

CC @akleeman 
",2015-02-24T01:20:01Z,2015-02-25T06:21:03Z,2015-02-25T06:21:01Z,2015-02-25T06:21:01Z,a26cbb42e41a0470394ff986093d1beffe18baf5,,799013,0,b624603c93a77cf024f7dfae893b55154f9741d7,400317e9afbbfafacb3aea4ffd70e8790c936ee6,MEMBER,,13221727,https://github.com/pydata/xarray/pull/333,
29864970,MDExOlB1bGxSZXF1ZXN0Mjk4NjQ5NzA=,334,closed,0,Fix bug associated with reading / writing of mixed endian data.,514053,"The right solution to this is to figure out how to successfully round trip endian-ness, but that seems to be a deeper issue inside netCDF4 (https://github.com/Unidata/netcdf4-python/issues/346)

Instead we force all data to little endian before netCDF4 write.
",2015-02-24T01:57:43Z,2015-02-26T04:45:18Z,2015-02-26T04:45:18Z,,8634487c2196fc84708be8e49ce59213c7623dfc,,799013,0,b2bec2f7a6e02b3994a8da47aa4845810baaf136,400317e9afbbfafacb3aea4ffd70e8790c936ee6,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/334,
29962836,MDExOlB1bGxSZXF1ZXN0Mjk5NjI4MzY=,335,closed,0,Add broadcast_equals method to Dataset and DataArray,1217238,,2015-02-25T05:51:46Z,2015-02-26T04:35:52Z,2015-02-26T04:35:49Z,2015-02-26T04:35:49Z,d97c28ee4962f59e7fe36d9c96a034cc67f1f484,,799013,0,a066dc0346c4f9d40000f62f6ff1718725574152,400317e9afbbfafacb3aea4ffd70e8790c936ee6,MEMBER,,13221727,https://github.com/pydata/xarray/pull/335,
29964211,MDExOlB1bGxSZXF1ZXN0Mjk5NjQyMTE=,336,closed,0,Add Dataset.drop and DataArray.drop,1217238,"These are convenient shortcuts for removing variables or index labels from an xray object.
",2015-02-25T06:35:18Z,2015-02-25T22:01:49Z,2015-02-25T22:01:49Z,2015-02-25T22:01:49Z,78e563c864cd40cdc7f500f3ebb911eca58b2cf6,,799013,0,903ad7088b53524e4120b577b37536bea8745114,b6baa768876891a9ee91bf7f4698b470f11dab25,MEMBER,,13221727,https://github.com/pydata/xarray/pull/336,
30056409,MDExOlB1bGxSZXF1ZXN0MzAwNTY0MDk=,337,closed,0,Cleanup (mostly documentation),1217238,,2015-02-26T07:40:01Z,2015-02-27T22:22:47Z,2015-02-26T07:43:37Z,2015-02-26T07:43:37Z,75ce04bf46d4a11d7920fa3b2a425c7d23ed7523,,799013,0,ca14eb754bd04565866e9abbf31ccf03c62388c2,7ea9de85698ebe275868fc364886a5010d90f107,MEMBER,,13221727,https://github.com/pydata/xarray/pull/337,
30057152,MDExOlB1bGxSZXF1ZXN0MzAwNTcxNTI=,338,closed,0,Truncate long attributes when printing datasets,1217238,"Only the first 500 characters are now shown, e.g.,

```
In [2]: xray.Dataset(attrs={'foo': 'bar' * 1000})
Out[2]:
<xray.Dataset>
Dimensions:  ()
Coordinates:
    *empty*
Data variables:
    *empty*
Attributes:
    foo: barbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarb
arbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarba
rbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbar
barbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarb
arbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarba
rbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbar
barbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarb
arbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarbarba...
```
",2015-02-26T07:57:40Z,2015-02-26T08:06:17Z,2015-02-26T08:06:05Z,2015-02-26T08:06:05Z,4088a173e371f0047f9683d3c5473683a1c0b5cc,,799013,0,5b8dbef12748fb1aecb00aa02982afcbd3bd3456,3eb015bb7a6d9dd1c51d5fd2e16c87afcee7c0a0,MEMBER,,13221727,https://github.com/pydata/xarray/pull/338,
30249875,MDExOlB1bGxSZXF1ZXN0MzAyNDk4NzU=,346,closed,0,Fix bug where Coordinates could turn Variable objects in Dataset constructor,1217238,"This manifested itself in some variables not being written to netCDF files, because they were determined to be trivial indexes (hence that logic was also updated to be slightly less questionable).
",2015-03-01T22:08:36Z,2015-03-01T23:57:58Z,2015-03-01T23:57:55Z,2015-03-01T23:57:55Z,239596f712c8adca92e72d19811122ae037b8790,,799013,0,7c25a75ae5c076d11cc08efd18171bd7c6a9d641,ba560e16eb45ab23ddd5fd16e3b7f5dfdc3c8181,MEMBER,,13221727,https://github.com/pydata/xarray/pull/346,
30254961,MDExOlB1bGxSZXF1ZXN0MzAyNTQ5NjE=,348,closed,0,Fix Dataset aggregate boolean,1217238,"Fixes #342
",2015-03-02T02:26:27Z,2015-03-02T18:14:12Z,2015-03-02T18:14:11Z,2015-03-02T18:14:11Z,183fdd9d0168c7f65eb46dd27d8d89edb1fc928f,,799013,0,947ed235386b5e7bca56ad87b846c832b900e9d4,3546b1ac11b93ff735c1690cb22ae1db31617201,MEMBER,,13221727,https://github.com/pydata/xarray/pull/348,
30310527,MDExOlB1bGxSZXF1ZXN0MzAzMTA1Mjc=,350,closed,0,Fix Dataset repr with netcdf4 datetime objects,1217238,"Fixes #347
",2015-03-02T19:06:28Z,2015-03-02T19:25:01Z,2015-03-02T19:25:00Z,2015-03-02T19:25:00Z,5c49411822b50fa5f7bb7d3e568e3d9ee914fb45,,799013,0,a732934aac9d01d48aaedbcbb8314f290bcc198a,40476f6582473b8aaba3d6928fdfdd2c33a97f49,MEMBER,,13221727,https://github.com/pydata/xarray/pull/350,
30319637,MDExOlB1bGxSZXF1ZXN0MzAzMTk2Mzc=,351,closed,0,Switch the name of datetime components from 'time.month' to 'month',1217238,"Fixes #345

This lets you write things like:

```
counts = time.groupby('time.month').count()
counts.sel(month=2)
```

instead of the previously valid

```
counts.sel(**{'time.month': 2})
```

which is much more awkward. Note that this breaks existing code which relied on the old usage.

CC @jhamman 
",2015-03-02T20:55:24Z,2015-03-02T23:20:09Z,2015-03-02T23:20:07Z,2015-03-02T23:20:07Z,157993f82a8d284203d13f400412659e4166f803,,799013,0,918016b68bbd49ff7ccce081c48408c305940e40,2bbc41bf4bc459155abf59c7a9907cfdc79b1835,MEMBER,,13221727,https://github.com/pydata/xarray/pull/351,
30345380,MDExOlB1bGxSZXF1ZXN0MzAzNDUzODA=,355,closed,0,Partial fix for netCDF4 datetime issues,1217238,"xref #340
",2015-03-03T04:11:51Z,2015-03-03T05:02:54Z,2015-03-03T05:02:52Z,2015-03-03T05:02:52Z,511f79b1910e3f6bd1582c799ec1383d56d3f798,,799013,0,32ba52b39e09e23e5fe9acc94e49427985cdea87,31b6365fe9f53c76eb31ec8be8a9e78006c84c35,MEMBER,,13221727,https://github.com/pydata/xarray/pull/355,
30348750,MDExOlB1bGxSZXF1ZXN0MzAzNDg3NTA=,356,closed,0,Documentation updates,1217238,"Fixes #343

(among other small changes)
",2015-03-03T06:01:03Z,2015-03-03T06:02:57Z,2015-03-03T06:02:56Z,2015-03-03T06:02:56Z,bfd9aef74d2a9b43e810db631a3e459157640720,,799013,0,5dd967c2ec4e978b3081134873922cef7ac0c811,37cb1d2d1e8968f5011e44d3e2bcfbc7ad29e055,MEMBER,,13221727,https://github.com/pydata/xarray/pull/356,