All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog and this project adheres to Semantic Versioning.
- Add
get_list_continuous_queries,drop_continuous_query, andcreate_continuous_querymanagement methods for continuous queries (#681 thx @lukaszdudek-silvair) - query() now accepts a bind_params argument for parameter binding (#678 thx @clslgrnc)
- Add consistency param to InfluxDBClient.write_points (#643 thx @RonRothman)
- Update test suite to add support for Python 3.7 and InfluxDB v1.6.4 and 1.7.4 (#692 thx @clslgrnc)
- Update classifiers tuple to list in setup.py (#697 thx @Hanaasagi)
- Update documentation for empty
delete_seriesconfusion - Update tests/tutorials_pandas.py to reference
lineprotocol, bug injson(#797 thx @Aeium)
- Fix 'TypeError: Already tz-aware' introduced with recent versions of Panda (#671, #676, thx @f4bsch @clslgrnc)
- Pass through the "method" kwarg to DataFrameClient queries
- Finally add a CHANGELOG.md to communicate breaking changes (#598)
- Test multiple versions of InfluxDB in travis
- Add SHARD DURATION parameter to retention policy create/alter
- Update POST/GET requests to follow verb guidelines from InfluxDB documentation
- Update test suite to support InfluxDB v1.3.9, v1.4.2, and v1.5.4
- Fix performance degradation when removing NaN values via line protocol (#592)
- Dropped support for Python3.4
- Connect to InfluxDB path running on server (#556 thx @gladhorn)
- Escape measurement names in DataFrameClient (#542 thx @tzonghao)
- Escape tags that end with a backslash (#537 thx @vaniakov)
- Add back mistakenly-dropped database parameter (#540)
- Add PyPI status to README.md
- Fix bad session mount scheme (#571 thx @vaniakov)
- Fixed issue with DataFrameClient calling to_datetime function (#593 thx @dragoshenron)
- Escape columns in DataFrameClient for line protocol (#584 thx @dmuiruri)
- Convert DataFrameClient times from int to np.int64 (#495 thx patrickhoebeke)
- Updated pandas tutorial (#547 thx @techaddicted)
- Explicitly set numpy version for tox (#563)
- Removed UDP precision restrictions on timestamp (#557 thx @mdhausman)
- Add pool size parameter to client constructor (#534 thx @vaniakov)
- Add ping method to client for checking connectivity (#409 thx @pmenglund)
- Add retry logic & exponential backoff when a connection fails (#508)
- Declare which setuptools version is required in PyPy env
- Functions for drop_measurement and get_list_measurements in InfluxDBClient (#402 thx @Vic020)
- Allow single string as data argument in write (#492 thx @baftek)
- Support chunked queries in DataFrameClient (#439 thx @gusutabopb)
- Add close method to InfluxDBClient (#465 thx @Linux-oiD)
- PEP257 linting & code compliance (#473)
- Fix broken tags filtering on a ResultSet (#511)
- Improve retry codepath for connecting to InfluxDB (#536 thx @swails)
- Clean up imports using six instead of sys.version (#536 thx @swails)
- Replace references to dataframe.ix with dataframe.iloc (#528)
- Improve performance of tag processing when converting DataFrameClient to line protocol (#503 thx @tzonghao)
- Typo in Content-Type header (#513 thx @milancermak)
- Clean up README.md formatting
- Catch TypeError when casting to float to return False with objects (#475 thx @BenHewins)
- Improve efficiency of tag appending in DataFrameClient when converting to line protocol (#486 thx @maxdolle)
- Drop requirement for all fields in SeriesHelper (#518 thx @spott)
- use_udp and udp_port are now private properties in InfluxDBClient