third_codes/eland
1
0
Fork 0
mirror of https://github.com/elastic/eland.git synced 2025-07-11 00:02:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fixing docstring format

Browse Source
This commit is contained in:
Stephen Dodson 2019-11-26 11:10:18 +00:00
parent 86686ebb18
commit 93dadc054c
1 changed files with 51 additions and 50 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

101
eland/dataframe.py
View File

@ -92,6 +92,7 @@ class DataFrame(NDFrame):
<BLANKLINE>
[5 rows x 2 columns]
"""
def __init__(self,
client=None,
index_pattern=None,
@ -339,59 +340,59 @@ class DataFrame(NDFrame):
def info_es(self):
# noinspection PyPep8
"""
A debug summary of an eland DataFrame internals.
A debug summary of an eland DataFrame internals.
This includes the Elasticsearch search queries and query compiler task list.
This includes the Elasticsearch search queries and query compiler task list.
Returns
-------
str
A debug summary of an eland DataFrame internals.
Returns
-------
str
A debug summary of an eland DataFrame internals.
Examples
--------
>>> df = ed.DataFrame('localhost', 'flights')
>>> df = df[(df.OriginAirportID == 'AMS') & (df.FlightDelayMin > 60)]
>>> df = df[['timestamp', 'OriginAirportID', 'DestAirportID', 'FlightDelayMin']]
>>> df = df.tail()
>>> df
timestamp OriginAirportID DestAirportID FlightDelayMin
12608 2018-02-10 01:20:52 AMS CYEG 120
12720 2018-02-10 14:09:40 AMS BHM 255
12725 2018-02-10 00:53:01 AMS ATL 360
12823 2018-02-10 15:41:20 AMS NGO 120
12907 2018-02-11 20:08:25 AMS LIM 225
<BLANKLINE>
[5 rows x 4 columns]
>>> print(df.info_es())
index_pattern: flights
Index:
index_field: _id
is_source_field: False
Mappings:
capabilities: _source es_dtype pd_dtype searchable aggregatable
AvgTicketPrice True float float64 True True
Cancelled True boolean bool True True
Carrier True keyword object True True
Dest True keyword object True True
DestAirportID True keyword object True True
... ... ... ... ... ...
OriginLocation True geo_point object True True
OriginRegion True keyword object True True
OriginWeather True keyword object True True
dayOfWeek True integer int64 True True
timestamp True date datetime64[ns] True True
<BLANKLINE>
[27 rows x 5 columns]
Operations:
tasks: [('boolean_filter', {'bool': {'must': [{'term': {'OriginAirportID': 'AMS'}}, {'range': {'FlightDelayMin': {'gt': 60}}}]}}), ('field_names', ['timestamp', 'OriginAirportID', 'DestAirportID', 'FlightDelayMin']), ('tail', ('_doc', 5))]
size: 5
sort_params: _doc:desc
_source: ['timestamp', 'OriginAirportID', 'DestAirportID', 'FlightDelayMin']
body: {'query': {'bool': {'must': [{'term': {'OriginAirportID': 'AMS'}}, {'range': {'FlightDelayMin': {'gt': 60}}}]}}, 'aggs': {}}
post_processing: ['sort_index']
<BLANKLINE>
"""
Examples
--------
>>> df = ed.DataFrame('localhost', 'flights')
>>> df = df[(df.OriginAirportID == 'AMS') & (df.FlightDelayMin > 60)]
>>> df = df[['timestamp', 'OriginAirportID', 'DestAirportID', 'FlightDelayMin']]
>>> df = df.tail()
>>> df
timestamp OriginAirportID DestAirportID FlightDelayMin
12608 2018-02-10 01:20:52 AMS CYEG 120
12720 2018-02-10 14:09:40 AMS BHM 255
12725 2018-02-10 00:53:01 AMS ATL 360
12823 2018-02-10 15:41:20 AMS NGO 120
12907 2018-02-11 20:08:25 AMS LIM 225
<BLANKLINE>
[5 rows x 4 columns]
>>> print(df.info_es())
index_pattern: flights
Index:
index_field: _id
is_source_field: False
Mappings:
capabilities: _source es_dtype pd_dtype searchable aggregatable
AvgTicketPrice True float float64 True True
Cancelled True boolean bool True True
Carrier True keyword object True True
Dest True keyword object True True
DestAirportID True keyword object True True
... ... ... ... ... ...
OriginLocation True geo_point object True True
OriginRegion True keyword object True True
OriginWeather True keyword object True True
dayOfWeek True integer int64 True True
timestamp True date datetime64[ns] True True
<BLANKLINE>
[27 rows x 5 columns]
Operations:
tasks: [('boolean_filter', {'bool': {'must': [{'term': {'OriginAirportID': 'AMS'}}, {'range': {'FlightDelayMin': {'gt': 60}}}]}}), ('field_names', ['timestamp', 'OriginAirportID', 'DestAirportID', 'FlightDelayMin']), ('tail', ('_doc', 5))]
size: 5
sort_params: _doc:desc
_source: ['timestamp', 'OriginAirportID', 'DestAirportID', 'FlightDelayMin']
body: {'query': {'bool': {'must': [{'term': {'OriginAirportID': 'AMS'}}, {'range': {'FlightDelayMin': {'gt': 60}}}]}}, 'aggs': {}}
post_processing: ['sort_index']
<BLANKLINE>
"""
buf = StringIO()
super()._info_es(buf)