0.11

0.11

We are happy to announce the first MonetDBe-Python release for 2022! The most important changes are:

• Dropped support for Python < 3.7
• Manylinux2014 wheels for Apple Silicon M1 (aarch64)
• Manylinux2014 wheels for Linux arm64 (aarch64)
• Built with the latest MonetDB Jul2021 release
• Speed improvements for numpy result fetching and inserting for string, DateTime, bool and unsigned types.

Details

• Make an Apple M1 (Apple Silicon arm) wheel #107
• Fix test suite for Windows by @gijzelaerr in #140
• Add regression test for #136 bigint result by @gijzelaerr in #147
• add regression test for issue #118 by @gijzelaerr in #146
• Fix all CI builds by @gijzelaerr in #150
• Issue151 monetdbe bind by @aris-koning in #153
• Add support for decimal types. by @aris-koning in #155
• add python 3.10 to test targets by @gijzelaerr in #157
• Add regression test for #158 by @gijzelaerr in #159
• Update windows builds by @gijzelaerr in #160
• #139 use append for more types by @gijzelaerr in #161
• Cleanup old oct2020 references and remove backward compat code by @gijzelaerr in #164

Full Changelog: 0.10...0.11

0.10

This release is based on the Jul2021 branch of MonetDB.
The only other big change is that qmark style queries are now parsed on the MonetDB engine side, not Python, which should speed up execution.

Other changes:

• Test suite fails for windows with Oct2020 & Jul2021 branch #133
• Queries on INT and BIGINT returning incorrect results in Python? bug #136
• Building sphinx docs fails: ModuleNotFoundError: No module named 'monetdbe._lowlevel' #131
• actual pip install missing from installation instructions in doc #125
• Improve automated building of binary wheels #124
• monetdbe bind/execute can't handle all cases (yet) #129
• Issue #54 use prepare bind for query format #110
• Set up CI to push binary wheels for Windows for every release build automation windows #53
• assume 128-bit enabled? enhancement #106

0.9

New features

• windows binary wheel #17
• Simpe csv reader/writer #97
• Add support for fast inserting of numpy columns and/or dataframes #42

Bug fixes

• MonetDB docker images are outdated build automation #101
• Fix various Windows issues windows #100
• Get MonetDBlite test suite working (tests/test_lite/test_pylite00.py) #43
• Have all SQLite regression tests pass #32
• Tests failing on Windows windows #82
• monetdbe_column_timestamp.is_null() gives sigabrt bug #92
• MonetDBe on windows broken windows #96
• use monetdb_get_table and get_column to fetch metadata #35
• SQLException:monetdbe.monetdbe_append:Cannot append 6 into column 't' when running CFFI test suite bug #93
• MonetDB release build is linked to debug DLLs (Windows) bug windows #95
• REAL data type issue with Pandas bug #83
• Connect call optional arguments enhancement #66
• Early bailout upon accessing a server database bug #64
• Oct2020 monetdb branch build fails #85
• Github actions for OSX 10.15 fails bug #88

0.8.4

Only change:

• Fix OSX binary wheel #76

Minor release, speed improvements

changes:

• Avoid debugging and assertions in wheels #74

bug fix release

the only change:

• libcrypto not properly picked up in binary wheel #72

bug-fix release

This is a bug-fix release and solves these issues:

• Current monetdbe wheel and test container give error: Crypt backend hash not found bug #68
• Improper error message bug #63
• fetchdf() and fetchnumpy() do not work for all data types bug #62
• lowlevel.query() sometimes receives empty connection #60
• Crashing on MONETDB url (monetdb://localhost:5000/) bug #49

First public release

The first public release of MonetDBe-Python! MonetDBe-Python is a Python frontend for MonetDBe, a blazingly fast column-store database engine. Note that this is a pre-release, and things might still break here and there.

Features included in this release are:

• An easy to install non-service database system! That means you are not required to run a database service in the background.
• A cleaned up and modernized version of the notorious MonetDB column-store as the backend
• SQLite compatible syntax and usage
• .fetchnumpy() and .fetchdf() methods for super-fast fetching of result dataset
• Linux and OSX binary wheels supplied

What to expect in the future:

• Fast insertions of NumPy arrays and Pandas dataframes
• Windows binary wheels
• Other compatibility and performance improvements

Please let us know what you think! If you experience any issues please let us know in the issue tracker here in Github.

to install, just run:

$ pip install monetdbe

Support operating systems are OS X 10.13+ and Linux. Support Python versions are 3.6 and newer. Note that you need pip 19.3 or newer to install these binary wheels. You can upgrade your pip with pip (pip install --upgrade pip)

0.6

Various bugfixes, and we now also have binary wheels for OSX.

https://github.com/MonetDBSolutions/MonetDBe-Python/issues?q=is%3Aclosed+milestone%3A0.6

0.5

See an overview of fixed issues on the issue tracker:

https://github.com/MonetDBSolutions/MonetDBe-Python/milestone/4?closed=1