Releases: coroa/pandas-indexing
Version 0.6.1
What's Changed
Full Changelog: v0.6.0...v0.6.1
Version 0.6.0
What's Changed
- feat(core): formatlevel and extractlevel get optional patterns by @coroa in #61
- feat(iamc): Add Resolver to new iamc sub-package by @coroa in #60
Full Changelog: v0.5.2...v0.6.0
Version 0.5.2
What's Changed
- enh(units): Load pint and pint-pandas only on demand by @coroa in #57
- feat(selectors): Allow mixing selectors with Series and callables by @coroa in #59
Full Changelog: v0.5.1...v0.5.2
Version 0.5.1
What's Changed
- fix(selectors): Make isin selector return a Series by @coroa in #55
- fix(utils): Fix shell_pattern_to_regex to work with brackets by @coroa in #56
Full Changelog: v0.5.0...v0.5.1
Version 0.5.0
What's Changed
Breaking
Deprecates the drop argument of extractlevel
in favour of keep and changes the default to keep=False
, ie. if keep is not specified, the split level will be removed.
Enhancements
Full Changelog: v0.4.2...v0.5.0
Version 0.4.2
What's Changed
Full Changelog: v0.4.1...v0.4.2
Version 0.4.1
What's Changed
Full Changelog: v0.4.0...v0.4.1
Version 0.4.0
Version 0.3.1
What's Changed
Full Changelog: v0.3...v0.3.1
Version 0.3
What's Changed
Breaking
- Align Series and DataFrame in
assignlevel
in #38
Previously, assignlevel
ignored the index of Series
or DataFrame
passed in, which caused subtle assignment bugs
over at the concordia
repository (since semijoin
sometimes updated the ordering even with how="right"
).
From this version on, assignlevel
will use a right-semijoin
to align any Series
or DataFrame
before adding it to the
index. Since the semijoin
uses the new fail_on_reorder
method introduced in v0.2.10
, this might fail in very particular situations. Then the DataFrame
/Series
to which the new levels are to be assigned needs to be sort_index
ed before.
Enhancements
Full Changelog: v0.2.10...v0.3