fix: fixed serialization of DataFrame with empty (NaN) values, fixed escaping whitespaces, fixed order of tags

[rolincova]

Closes #118

• Fixed serialization of DataFrame with empty (NaN) values
• Fixed serialization of \n, \r and \t
• Fixed order of tags

• CHANGELOG.md updated
• Rebased/mergeable
• Tests pass
• Sign CLA (if not already signed)

[codecov-commenter]

Codecov Report

Merging #123 into master will increase coverage by 0.21%.
The diff coverage is 96.87%.

@@            Coverage Diff             @@
##           master     #123      +/-   ##
==========================================
+ Coverage   88.71%   88.93%   +0.21%     
==========================================
  Files          22       23       +1     
  Lines        1826     1852      +26     
==========================================
+ Hits         1620     1647      +27     
+ Misses        206      205       -1     

Impacted Files	Coverage Δ
...fluxdb_client/client/write/dataframe_serializer.py	96.66% <96.66%> (ø)
influxdb_client/client/write_api.py	97.53% <100.00%> (+0.91%)	⬆️

---

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 f74d183...cd01183. Read the comment docs.

[bednar]

Please also fix escaping of white spaces:

    def test_tag_escaping_key_and_value(self):
        from influxdb_client.extras import pd, np

        now = pd.Timestamp('2020-04-05 00:00+00:00')

        data_frame = pd.DataFrame(data=[["new\nline", "carriage\nreturn", "t\tab", np.int64(100.5)], ],
                                  index=[now + timedelta(hours=1), ],
                                  columns=["new\nline", "carriage\rreturn", "t\tab", "l\ne\rv\tel"])

        points = data_frame_to_list_of_points(data_frame=data_frame,
                                              point_settings=PointSettings(),
                                              data_frame_measurement_name='h\n2\ro\t_data',
                                              data_frame_tag_columns={"new\nline", "carriage\rreturn", "t\tab"})

        self.assertEqual(1, len(points))
        self.assertEqual("h\\n2\\ro\\t_data,carriage\\rreturn=carriage\\nreturn,new\\nline=new\\nline,t\\tab=t\\tab l\\ne\\rv\\tel=2i",
                         points[0])

[7yl4r]

Looks like the 3.7 check might just need to be rerun; it's a 500 error from the influx server.

[bednar]

Hi @7yl4r,

we want to also fix a problem with serialization of whitespaces and order of tags (https://v2.docs.influxdata.com/v2.0/write-data/best-practices/optimize-writes/#sort-tags-by-key)

I hope that we are be able to merge it soon.

Regards

[bednar]

Please also fix order of tags:

def test_tags_order(self):
    from influxdb_client.extras import pd, np

    now = pd.Timestamp('2020-04-05 00:00+00:00')

    data_frame = pd.DataFrame(data=[["c", "a", "b", np.int64(2)], ],
                              index=[now + timedelta(hours=1), ],
                              columns=["c", "a", "b", "level"])

    points = data_frame_to_list_of_points(data_frame=data_frame,
                                          point_settings=PointSettings(),
                                          data_frame_measurement_name='h2o',
                                          data_frame_tag_columns={"c", "a", "b"})

    self.assertEqual(1, len(points))
    self.assertEqual("h2o,a=a,b=b,c=c level=2i 1586048400000000000", points[0])

https://v2.docs.influxdata.com/v2.0/write-data/best-practices/optimize-writes/#sort-tags-by-key

[bednar]

LGTM