ENH: support non default indexes in writing to Parquet #18629

dhirschfeld · 2017-12-04T13:43:05Z

closes Support non-default indexes in to_parquet #18581
tests added / passed
passes git diff upstream/master -u -- "*.py" | flake8 --diff
whatsnew entry

jorisvandenbossche · 2017-12-04T16:19:19Z

We need to check how this interacts with the columns keyword when reading back in. Ideally selecting only a few columns to read will still preserve your index (I think?), but I guess this is currently not happening (at least with pyarrow).

codecov · 2017-12-04T17:48:26Z

Codecov Report

Merging #18629 into master will decrease coverage by 0.01%.
The diff coverage is 90.9%.

@@            Coverage Diff             @@
##           master   #18629      +/-   ##
==========================================
- Coverage   91.59%   91.58%   -0.02%     
==========================================
  Files         155      155              
  Lines       51255    51260       +5     
==========================================
- Hits        46949    46945       -4     
- Misses       4306     4315       +9

Flag	Coverage Δ
#multiple	`89.44% <90.9%> (ø)`	⬆️
#single	`40.66% <0%> (-0.12%)`	⬇️

Impacted Files	Coverage Δ
pandas/io/parquet.py	`66.26% <90.9%> (+0.88%)`	⬆️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/core/frame.py	`97.81% <0%> (-0.1%)`	⬇️
pandas/core/indexes/datetimes.py	`95.68% <0%> (+0.09%)`	⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2c903d5...a14edbe. Read the comment docs.

codecov · 2017-12-04T17:48:37Z

Codecov Report

Merging #18629 into master will decrease coverage by 0.03%.
The diff coverage is 53.19%.

@@            Coverage Diff             @@
##           master   #18629      +/-   ##
==========================================
- Coverage    91.6%   91.57%   -0.04%     
==========================================
  Files         153      153              
  Lines       51317    51363      +46     
==========================================
+ Hits        47011    47034      +23     
- Misses       4306     4329      +23

Flag	Coverage Δ
#multiple	`89.43% <53.19%> (-0.02%)`	⬇️
#single	`40.73% <17.02%> (-0.1%)`	⬇️

Impacted Files	Coverage Δ
pandas/io/parquet.py	`60.19% <53.19%> (-4.37%)`	⬇️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/plotting/_converter.py	`64.78% <0%> (-1.74%)`	⬇️
pandas/core/frame.py	`97.81% <0%> (-0.1%)`	⬇️
pandas/core/generic.py	`95.9% <0%> (ø)`	⬆️
pandas/core/indexes/interval.py	`93.8% <0%> (ø)`	⬆️
pandas/core/indexes/base.py	`96.43% <0%> (ø)`	⬆️
pandas/core/indexes/multi.py	`96.29% <0%> (+0.01%)`	⬆️
pandas/core/indexes/datetimes.py	`95.7% <0%> (+0.01%)`	⬆️
... and 8 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 37086a0...0fbdc14. Read the comment docs.

jschendel · 2017-12-04T19:57:20Z

pandas/io/parquet.py

+                             "to make the index into column(s)".format(
+                                 type(df.index)))
+
+        if not df.index.equals(RangeIndex.from_range(range(len(df)))):


Could create the RangeIndex via RangeIndex(len(df)) here to make this a little cleaner.

jreback

lgtm. some changes; pls. add a bug fix note in 0.22.0

jreback · 2017-12-05T00:14:02Z

pandas/io/parquet.py

-    if df.index.name is not None:
-        raise ValueError("parquet does not serialize index meta-data on a "
-                         "default index")
+    # *unless* we're using pyarrow >= 0.7.1 which does support multi-indexes


so it might be better (though a bit more refactoring), to move the validation of index and validation of columns into methods on a base class to the impl

e.g.

class BaseImpl(object): def validate_index(....): .. def validate_columns(....): .... class PyArrowImpl(BaseImpl): def write(...): if pyarrow < 0.7.1: self.validate_index() self.validate_columns() class FastParquetImpl(BaseImpl): .....

as it makes things cleaner.

Refactored in the latest commits...

jreback · 2017-12-05T00:15:36Z

pandas/tests/io/test_parquet.py

+            df.index = index
+            self.check_round_trip(df, engine)
+
+        # index with meta-data


could factor these 2 tests here from pyarrow impl (and also from fastparquet) into another test (that succeeds on all impls and versions)

Since the behaviour changes based on the library and the version of the library I'm not sure how else to do it while maintaining test coverage on all versions.

Feel free to push a fix if you'd like as I'm knocking off for the evening shortly...

jreback

minor comments. if you can refactor the tests a bit ok (if not no big deal). ping on green.

jreback · 2017-12-05T11:34:50Z

pandas/io/parquet.py

+        self._validate_index(df)
+
+    def write(self, df, path, compression, **kwargs):
+        raise NotImplementedError()


import and use pandas.core.common.AbstractMethodError here instead

jreback · 2017-12-05T11:36:15Z

doc/source/whatsnew/v0.22.0.txt

@@ -77,6 +77,7 @@ Other Enhancements
 - :func:`Series.fillna` now accepts a Series or a dict as a ``value`` for a categorical dtype (:issue:`17033`)
 - :func:`pandas.read_clipboard` updated to use qtpy, falling back to PyQt5 and then PyQt4, adding compatibility with Python3 and multiple python-qt bindings (:issue:`17722`)
 - Improved wording of ``ValueError`` raised in :func:`read_csv` when the ``usecols`` argument cannot match all columns. (:issue:`17301`)
+- Enabled the use of non-default indexes in ``to_parquet`` with pyarrow>=0.7.0 (:issue:`18581`)


use :func:`DataFrame.to_parquet` here, use double-backticks around pyarrow>=0.7.0

jorisvandenbossche

fastparquet also supports writing the index, so no need to still raise the errors in that case.

Also, to repeat my comment from before: we need to decide how this interacts with columns keyword before merging this.

dhirschfeld · 2017-12-05T12:42:07Z

fastparquet also supports writing the index, so no need to still raise the errors in that case.

It seems it might just be easier all-round to just bump the pyarrow/fastparquet deps to the min required to support non-default indices - i.e. >=0.7.0 for pyarrow and ??? for fastparquet

jreback · 2017-12-05T13:10:16Z

would be ok with bumping pyarrow dep
i think fp is ok as is (i bumped it when doing this originally)

pep8speaks · 2017-12-06T12:04:50Z

Hello @dhirschfeld! Thanks for updating the PR.

Cheers ! There are no PEP8 issues in this Pull Request. 🍻

Comment last updated on December 11, 2017 at 17:00 Hours UTC

jreback

doc changes, ping on green.

jreback · 2017-12-06T13:43:57Z

ci/requirements-3.5.sh

@@ -8,4 +8,4 @@ echo "install 35"
 conda remove -n pandas python-dateutil --force
 pip install python-dateutil



can you check if we have pyarrow installed in ci/* anywhere else (and if so update the pinned versions)

I checked and changed any pinned versions to the new minimum. If it wasn't already pinned I left it as is since a normal install will pull in later versions that the minimum.

jreback · 2017-12-06T13:46:30Z

doc/source/whatsnew/v0.22.0.txt

@@ -163,7 +163,7 @@ Other Enhancements
 - :func:`pandas.read_clipboard` updated to use qtpy, falling back to PyQt5 and then PyQt4, adding compatibility with Python3 and multiple python-qt bindings (:issue:`17722`)
 - Improved wording of ``ValueError`` raised in :func:`read_csv` when the ``usecols`` argument cannot match all columns. (:issue:`17301`)
 - :func:`DataFrame.corrwith` now silently drops non-numeric columns when passed a Series. Before, an exception was raised (:issue:`18570`).
-
+- Enabled the use of non-default indexes in :func:`DataFrame.to_parquet` where the underlying engine supports it (:issue:`18581`)


also there is another section on changed deps, pls add another entry which says pyarrow is now min 0.7.0 for parquet (also may need to update in io.rst)

jorisvandenbossche · 2017-12-06T13:50:39Z

@jreback not merge on green, see my request for changes on which I got no reply yet

BTW, I think we can do this for the 0.21.1 milestone. The other enhancements to to/read_parquet also went into that one (seeing it as fixes on very new functionality)

dhirschfeld · 2017-12-07T09:59:33Z

@jorisvandenbossche - as of fastparquet=0.1.3 it doesn't appear to support multi-indexes

...at least reading them anyway. Interestingly, it seems to write them fine and the file written by fastparquet can then be read in perfectly by pyarrow.

Not sure what to do about that but since fastparquet throws a sensible error I'd be inclined to just leave it and a future update will likely add the missing functionality.

from numpy.random import randn
import pandas as pd

dates = pd.date_range('01-Jan-2018', '01-Dec-2018', freq='MS')
df = pd.DataFrame(randn(2*len(dates), 3), columns=list('ABC'))
index = pd.MultiIndex.from_product([['Level1', 'Level2'], dates])
df.index = index

df.to_parquet('tmp_pa.pq', engine='pyarrow')
df.to_parquet('tmp_fp.pq', engine='fastparquet', compression='UNCOMPRESSED')

df1 = pd.io.parquet.read_parquet('tmp_pa.pq', engine='pyarrow')
df1.equals(df)

dhirschfeld · 2017-12-07T10:18:05Z

@jorisvandenbossche - you're right about the columns argument not preserving the index:

The workaround, at the cost of efficiency, is simple enough:

pd.io.parquet.read_parquet('tmp_pa.pq', engine='pyarrow')[['A','B']]

...but you have to know beforehand that the file was saved with a multi-index - i.e. it's a silent failure mode which isn't ideal.

If I name the index levels you can actually pull them out by name:

df = pd.DataFrame(randn(2*len(dates), 3), columns=list('ABC'))
index = pd.MultiIndex.from_product([['Level1', 'Level2'], dates], names=['level','date'])
df.index = index

pd.io.parquet.read_parquet('tmp_pa.pq', engine='pyarrow', columns=['A','B','level','date']).head()

...so, ideally if pandas can tell if the file has a multi-index the index names could then always be prepended to the columns argument.

dhirschfeld · 2017-12-07T10:19:35Z

@jorisvandenbossche - were there other changes you wanted that I've missed?

jorisvandenbossche · 2017-12-07T10:31:58Z

Not sure what to do about that but since fastparquet throws a sensible error I'd be inclined to just leave it and a future update will likely add the missing functionality.

It's a bit unfortunate it only comes up when reading it back in, so either we are ok with that, or either we need to to raise on writing MIs with fastparquet.

If I name the index levels you can actually pull them out by name:

That will not always work. Eg if they have no names, and pyarrow master has also changed this and the index levels will never have the level names but a __index_level_n_ pattern, so the workaround you show will not work anymore with the new release.

The thing we can do is the following:

In [85]: import pyarrow.parquet as pq

In [86]: f = pq.ParquetFile("__test_mi.parquet")

In [87]: f.metadata.metadata[b'pandas']
Out[87]: b'{"columns": [{"name": "A", "pandas_type": "float64", "numpy_type": "float64", "metadata": null}, {"name": "B", "pandas_type": "float64", "numpy_type": "float64", "metadata": null}, {"name": "C", "pandas_type": "float64", "numpy_type": "float64", "metadata": null}, {"name": "__index_level_0__", "pandas_type": "unicode", "numpy_type": "object", "metadata": null}, {"name": "__index_level_1__", "pandas_type": "datetime", "numpy_type": "datetime64[ns]", "metadata": null}], "index_columns": ["__index_level_0__", "__index_level_1__"], "pandas_version": "0.22.0.dev0+288.g6b6cfb8"}'

In [88]: import json

In [89]: pandas_metadata = json.loads(f.metadata.metadata[b'pandas'].decode('utf8'))

In [90]: pandas_metadata['index_columns']
Out[90]: ['__index_level_0__', '__index_level_1__']

and then add those to the columns list.
And something similar for fastparquet:

In [98]: import fastparquet

In [99]: f = fastparquet.ParquetFile("__test_mi.parquet")

In [100]: json.loads(f.key_value_metadata['pandas'])['index_columns']
Out[100]: ['__index_level_0__', '__index_level_1__']

dhirschfeld · 2017-12-07T10:33:15Z

Heh, beat me to it:

import pyarrow.parquet as pq
filemetadata = pq.read_metadata('tmp_fp.pq')
metadata = eval(filemetadata.metadata.get(b'pandas'), None, dict(null=None))
metadata['index_columns']

...eval is pretty nasty!

jorisvandenbossche · 2017-12-07T10:33:53Z

I also think that if there is a default index (range-like index without name, the case that was allowed now), we should not write it to the parquet file, because that would just be waste of time and space.

jorisvandenbossche · 2017-12-07T10:34:24Z

Yep, the json.loads is the better alternative :-)

jreback · 2017-12-07T11:22:03Z

BTW, I think we can do this for the 0.21.1 milestone. The other enhancements to to/read_parquet also went into that one (seeing it as fixes on very new functionality)

since we are bumping the version I don't think this is a good idea. marked for 0.22

I am ok with this fix, @jorisvandenbossche

dhirschfeld · 2017-12-07T11:32:22Z

For pyarrow the columns fix is as simple as setting use_pandas_metadata=True:

    def read(self, path, columns=None, **kwargs):
        path, _, _ = get_filepath_or_buffer(path)
        parquet_file = self.api.parquet.ParquetFile(path)
        kwargs['use_pandas_metadata'] = True
        return parquet_file.read(columns=columns, **kwargs).to_pandas()

For fastparquet I would use:

    def read(self, path, columns=None, **kwargs):
        path, _, _ = get_filepath_or_buffer(path)
        parquet_file = self.api.ParquetFile(path)
        metadata = json.loads(parquet_file.key_value_metadata['pandas'])
        if columns is not None:
            columns = set(chain(columns, metadata['index_columns']))
        return parquet_file.to_pandas(columns=columns, **kwargs)

...but since it doesnt' support multi-indexes anyway it's a moot point.

jreback · 2017-12-10T16:19:43Z

pandas/tests/io/test_parquet.py

@@ -6,6 +6,7 @@
 from warnings import catch_warnings

 import numpy as np
+from numpy.random import randn


we don't do the import like this anywhere else, appreciate changing it

jreback · 2017-12-10T16:20:29Z

pandas/tests/io/test_parquet.py

+        df.columns = pd.MultiIndex.from_tuples([('a', 1), ('a', 2), ('b', 1)]),
+        self.check_error_on_write(df, engine, ValueError)
+
+    def test_multiindex_with_columns(self, engine):


you can just pass pa instead of engine to skip fp

jorisvandenbossche

@dhirschfeld Thanks for the update. I think there are still some left-overs to clean-up from the undo removing support for older versions, but I can also put some time in it this evening if needed.

jorisvandenbossche · 2017-12-10T16:15:45Z

pandas/io/parquet.py

+        self.validate_dataframe(df)
+        if self._pyarrow_lt_070:
+            self._validate_write_lt_070(
+                df, path, compression, coerce_timestamps, **kwargs


Is there a reason you pass path, compression, coerce_timestamps here?

Just a left over from a refactoring. Should be removed as they're no longer needed...

jorisvandenbossche · 2017-12-10T16:16:04Z

pandas/io/parquet.py

-        if self._pyarrow_lt_060:
-            table = self.api.Table.from_pandas(df, timestamps_to_ms=True)
-            self.api.parquet.write_table(
-                table, path, compression=compression, **kwargs)


Do we need to keep this for older pyarrow versions ?

jorisvandenbossche · 2017-12-10T16:17:17Z

pandas/io/parquet.py

+        if not isinstance(df.index, Int64Index):
+            msg = (
+                "parquet does not support serializing {} for the index;"
+                "you can .reset_index() to make the index into column(s)"


maybe we should add here that they can also install latest pyarrow or fastparquet version (same for the others)

jorisvandenbossche · 2017-12-10T16:18:25Z

pandas/io/parquet.py

-class FastParquetImpl(object):
+        parquet_file = self.api.parquet.ParquetFile(path)
+        if self._pyarrow_lt_070:
+            parquet_file.path = path


I would just pass path to the _read_lt_070 function?

jorisvandenbossche · 2017-12-10T23:45:20Z

@dhirschfeld sorry for taking over your PR, but I want to get this merged for 0.21.1, and didn't know if you would have time today/tomorrow.
I restored some of the testing code for testing the older versions, and fixed some bugs in the code to get those working. And addressed the comments of myself and @jreback.

But feel free to review my changes

dhirschfeld · 2017-12-11T01:17:00Z

@jorisvandenbossche - that's fine, I'd only be able to look into it tonight so am happy for you to take the lead! I'm hoping to make use of it so am keen to get it in too...

dhirschfeld · 2017-12-11T12:52:04Z

LGTM!

jorisvandenbossche · 2017-12-11T13:27:10Z

And all CI is passing!

@jreback more comments?
(the compat for older versions is maybe not the cleanest, but it will be directly removed on master anyway .. (I think we want to do that?))

jreback

tiny style comments, otherwise lgtm.

jreback · 2017-12-11T14:43:50Z

pandas/io/parquet.py

+    def validate_dataframe(df):
+        if not isinstance(df, DataFrame):
+            raise ValueError("to_parquet only supports IO with DataFrames")
+        # must have value column names (strings only)


can you add a blank line before comments, easier to read

jorisvandenbossche · 2017-12-11T18:57:39Z

@dhirschfeld Thanks a lot for starting the PR!

) fastparquet automatically names an index 'index' if it doesn't already have a name (cherry picked from commit 8d7e876)

dhirschfeld · 2017-12-12T01:07:15Z

Thanks @jorisvandenbossche for getting this over the line! I'm excited to make use of this new functionality! :)

fastparquet automatically names an index 'index' if it doesn't already have a name (cherry picked from commit 8d7e876)

Version 0.22.0 * tag 'v0.22.0': (777 commits) RLS: v0.22.0 DOC: Fix min_count docstring (pandas-dev#19005) DOC: More 0.22.0 updates (pandas-dev#19002) TST: Remove pow test in expressions COMPAT: Avoid td.skip decorator DOC: 0.22.0 release docs (pandas-dev#18983) DOC: Include 0.22.0 whatsnew Breaking changes for sum / prod of empty / all-NA (pandas-dev#18921) ENH: Added a min_count keyword to stat funcs (pandas-dev#18876) RLS: v0.21.1 DOC: Add date to whatsnew (pandas-dev#18740) DOC: Include 0.21.1 whatsnew DOC: Update relase notes (pandas-dev#18739) CFG: Ignore W503 DOC: fix options table (pandas-dev#18730) ENH: support non default indexes in writing to Parquet (pandas-dev#18629) BUG: Fix to_latex with longtable (pandas-dev#17959) (pandas-dev#17960) Parquet: Add error message for no engine found (pandas-dev#18717) BUG: Categorical data fails to load from hdf when all columns are NaN (pandas-dev#18652) DOC: clean-up whatsnew file for 0.21.1 (pandas-dev#18690) ...

* releases: (777 commits) RLS: v0.22.0 DOC: Fix min_count docstring (pandas-dev#19005) DOC: More 0.22.0 updates (pandas-dev#19002) TST: Remove pow test in expressions COMPAT: Avoid td.skip decorator DOC: 0.22.0 release docs (pandas-dev#18983) DOC: Include 0.22.0 whatsnew Breaking changes for sum / prod of empty / all-NA (pandas-dev#18921) ENH: Added a min_count keyword to stat funcs (pandas-dev#18876) RLS: v0.21.1 DOC: Add date to whatsnew (pandas-dev#18740) DOC: Include 0.21.1 whatsnew DOC: Update relase notes (pandas-dev#18739) CFG: Ignore W503 DOC: fix options table (pandas-dev#18730) ENH: support non default indexes in writing to Parquet (pandas-dev#18629) BUG: Fix to_latex with longtable (pandas-dev#17959) (pandas-dev#17960) Parquet: Add error message for no engine found (pandas-dev#18717) BUG: Categorical data fails to load from hdf when all columns are NaN (pandas-dev#18652) DOC: clean-up whatsnew file for 0.21.1 (pandas-dev#18690) ...

jorisvandenbossche added IO Parquet parquet, feather Enhancement labels Dec 4, 2017

jschendel reviewed Dec 4, 2017

View reviewed changes

jreback requested changes Dec 5, 2017

View reviewed changes

jorisvandenbossche requested changes Dec 5, 2017

View reviewed changes

dhirschfeld force-pushed the parquet-non-default-indexes branch from 1345847 to 44b0fe8 Compare December 6, 2017 12:04

dhirschfeld force-pushed the parquet-non-default-indexes branch 2 times, most recently from 0f7007a to 7b25fbe Compare December 6, 2017 12:12

jorisvandenbossche changed the title ~~Parquet non default indexes~~ ENH: support non default indexes in writing to Parquet Dec 6, 2017

jreback requested changes Dec 6, 2017

View reviewed changes

jorisvandenbossche mentioned this pull request Dec 6, 2017

DOC: Update parquet metadata format description around index levels #18201

Merged

jreback added this to the 0.22.0 milestone Dec 7, 2017

jreback approved these changes Dec 7, 2017

View reviewed changes

undo version changes in ci and install docs

cd26110

jreback reviewed Dec 10, 2017

View reviewed changes

jorisvandenbossche reviewed Dec 10, 2017

View reviewed changes

jorisvandenbossche added 3 commits December 10, 2017 18:41

undo testing changes for old pyarrow versions

1acb00c

fix test for older versions of pyarrow

0f7b526

use read_pandas for pyarrow < 0.7

67dffde

further clean-up

79de86b

jorisvandenbossche force-pushed the parquet-non-default-indexes branch from 5ed8e2d to 79de86b Compare December 10, 2017 23:46

jreback added this to the 0.21.1 milestone Dec 11, 2017

jreback added the Needs Backport label Dec 11, 2017

jreback approved these changes Dec 11, 2017

View reviewed changes

style update

0fbdc14

jorisvandenbossche approved these changes Dec 11, 2017

View reviewed changes

TomAugspurger mentioned this pull request Dec 11, 2017

RLS 0.21.1 #18244

Closed

58 tasks

jorisvandenbossche merged commit 8d7e876 into pandas-dev:master Dec 11, 2017

dhirschfeld deleted the parquet-non-default-indexes branch December 12, 2017 01:06

TomAugspurger pushed a commit that referenced this pull request Dec 12, 2017

ENH: support non default indexes in writing to Parquet (#18629)

1062903

fastparquet automatically names an index 'index' if it doesn't already have a name (cherry picked from commit 8d7e876)

TomAugspurger removed the Needs Backport label Dec 12, 2017

jorisvandenbossche mentioned this pull request Dec 12, 2017

COMPAT: drop pyarrow < 0.7 and fastparquet < 0.1 for Parquet IO #18742

Closed

hweecat mentioned this pull request Sep 13, 2020

ENH: Support MultiIndex columns in parquet (#34777) #36305

Merged

5 tasks

		@@ -8,4 +8,4 @@ echo "install 35"
		conda remove -n pandas python-dateutil --force
		pip install python-dateutil

ENH: support non default indexes in writing to Parquet #18629

ENH: support non default indexes in writing to Parquet #18629

Conversation

dhirschfeld commented Dec 4, 2017 • edited Loading

jorisvandenbossche commented Dec 4, 2017

codecov bot commented Dec 4, 2017 • edited Loading

Codecov Report

codecov bot commented Dec 4, 2017 • edited Loading

Codecov Report

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jorisvandenbossche left a comment

Choose a reason for hiding this comment

dhirschfeld commented Dec 5, 2017

jreback commented Dec 5, 2017

pep8speaks commented Dec 6, 2017 • edited Loading

Comment last updated on December 11, 2017 at 17:00 Hours UTC

jreback left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jorisvandenbossche commented Dec 6, 2017

dhirschfeld commented Dec 7, 2017 • edited Loading

dhirschfeld commented Dec 7, 2017

dhirschfeld commented Dec 7, 2017

jorisvandenbossche commented Dec 7, 2017 • edited Loading

dhirschfeld commented Dec 7, 2017

jorisvandenbossche commented Dec 7, 2017

jorisvandenbossche commented Dec 7, 2017

jreback commented Dec 7, 2017

dhirschfeld commented Dec 7, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jorisvandenbossche left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jorisvandenbossche commented Dec 10, 2017

dhirschfeld commented Dec 11, 2017

dhirschfeld commented Dec 11, 2017

jorisvandenbossche commented Dec 11, 2017

jreback left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jorisvandenbossche commented Dec 11, 2017

dhirschfeld commented Dec 12, 2017

dhirschfeld commented Dec 4, 2017 •

edited

Loading

codecov bot commented Dec 4, 2017 •

edited

Loading

codecov bot commented Dec 4, 2017 •

edited

Loading

pep8speaks commented Dec 6, 2017 •

edited

Loading

dhirschfeld commented Dec 7, 2017 •

edited

Loading

jorisvandenbossche commented Dec 7, 2017 •

edited

Loading