Narwhals v1.5.5

Changes

• Revert "patch: allow lit to broadcast as left operand" (#858)
• patch: allow lit to broadcast as left operand (#854)

✨ Enhancements

• feat: implement when/then/otherwise for PyArrow (#859)
• feat: add when-then-otherwise expression (#588)
• feat: use validate_column_comparand in zip_with (#857)
• feat: make DataFrame.select stricter: non-string columns can only be selected using nw.col (#855)

🛠️ Other improvements

• chore: simplify arrow (#860)

Thank you to all our contributors for making this release possible!
@FBruzzesi, @MarcoGorelli and @aivanoved

Narwhals v1.5.4

Changes

• ci: Correct labeller regex (#852)

✨ Enhancements

• feat: Add is_pandas_like_dataframe and is_pandas_like_series (#848)
• feat: Support more indexing: boolean lists to DataFrame.filter and Series.filter, add DataFrame.row, (#847)

🛠️ Other improvements

• patch: sum_horizontal ignore nulls (#825)
• chore: nox revamp and coverage threshold change (#850)

Thank you to all our contributors for making this release possible!
@DeaMariaLeon, @FBruzzesi, @MarcoGorelli and @anopsy

Narwhals v1.5.3

Changes

✨ Enhancements

• feat: dask in api-completeness table (#741)
• feat: add DaskExpr.quantile (#835)
• feat: mean_horizontal (#843)
• feat: support __getitem__ with slices for columns (#839)

📖 Documentation

• docs: add narwhals-gif to README.md (#845)
• docs: Fix indentation for getitem (#844)

🛠️ Other improvements

• ci: changed release-drafter regex (#846)
• test: remove to_numpy and to_list in tests (#842)

Thank you to all our contributors for making this release possible!
@DeaMariaLeon, @FBruzzesi, @MarcoGorelli and @anopsy

Narwhals v1.5.2

Changes

🐞 Bug fixes

• fix: __array__ for pandas (pyarrow dtypes) was raising for old versions of PyArrow (#837)

Thank you to all our contributors for making this release possible!
@MarcoGorelli

Narwhals v1.5.1

Changes

✨ Enhancements

• feat: write_csv (#832)
• feat: support dtype and copy in DataFrame.array (#826)
• feat: allow for non-string columns in select for pandas backend (#828)

🐞 Bug fixes

• fix: improve validate columns (#829)

Thank you to all our contributors for making this release possible!
@DeaMariaLeon and @MarcoGorelli

Narwhals v1.5.0

Changes

✨ Enhancements

• feat: add dt.to_string to DaskExpr (#796)
• feat: add DaskExpr.total_minutes, total_seconds, total_milliseconds, total_microseconds, total_nanoseconds (#811)
• feat: dask expr is_unique & is_duplicated (#803)
• feat: add is_in to DaskExpr and mark is_duplicated and is_unique as not implemented (#802)

📦 Build system

• build: Add cudf and modin to deps (#816)

🛠️ Other improvements

• test: Remove some to_numpy (#820)
• build: Add cudf and modin to deps (#816)
• test: Remove some to_list() (#812)
• test: Use compare dicts in is_duplicated_test (#809)
• ci: Modified drafter's labeller regex to match titles with uppercase (#807)
• chore: use compare_dict in series tests (#801)
• test: Improve duration_attribute_tests (#798)
• chore: stable api narwhalify de-duplication (#799)
• test: Add cudf_constructor to conftest (#797)

Thank you to all our contributors for making this release possible!
@DeaMariaLeon, @FBruzzesi, @MarcoGorelli, @benrutter, @lucianosrp, @luke396 and @thomasjpfan

Narwhals v1.4.2

Changes

• [pre-commit.ci] pre-commit autoupdate (#308)

✨ Enhancements

• feat: add diff() method to Dask backend (#793)
• feat: add dask Expr.null_count (#792)

🐞 Bug fixes

• fix: cuDF compat (#794)

📖 Documentation

• docs: add Narwhals.dependencies to API reference (#791)

Thank you to all our contributors for making this release possible!
@MarcoGorelli, @anopsy, @pre-commit-ci, @pre-commit-ci[bot] and @raisadz

Narwhals v1.4.1

Changes

✨ Enhancements

• feat: add narwhals.new_series (#787)
• feat: Support Arrow PyCapsule Interface for export (#786)
• feat: improve typing for to_pandas / to_numpy / to_arrow (#783)
• feat: extend dataframe drop method (#773)

🛠️ Other improvements

• chore: import overhaul (#788)
• chore: rename _native_dataframe to _native_frame (#785)
• chore: rename internal dtype functions (#780)

Thank you to all our contributors for making this release possible!
@FBruzzesi, @MarcoGorelli and @anopsy

Narwhals v1.4.0

Changes

🚀 Performance improvements

• perf: remove some lambdas from _dask backend (#765)

✨ Enhancements

• feat: allow from_dict to not have native_namespace arg if all inputs are already narwhals Series (#760)
• feat: extend dataframe drop_nulls method (#779)
• feat: implement LazyFrame.head for Dask (plus: run tpc-h q2 query with Dask present) (#768)
• feat: add str.replace and str.replace_all (#750)
• feat: dask lazyframe remaining methods (#778)
• feat: dask sum_horizontal (#775)
• feat: dask namespace lit method (#772)
• feat: DaskSelectorNamespace (#771)
• feat: dataframe to_arrow method (#770)
• feat: series to_arrow method (#769)
• feat: dask expr is_first|last_distinct methods (#764)
• feat: dask expr __invert__, is_null, n_unique methods (#763)
• feat: dask expr len method (#762)
• feat: Series.pipe and Expr.pipe (#582)

🐞 Bug fixes

• fix: empty with_columns not working for pandas backend (#757)

📖 Documentation

• docs: add Altair to "used by" section (#776)
• docs: include "downloads per month" in readme (#774)
• docs: run tpch q1 with more libraries (#759)
• docs: use animals instead of cars in example (#758)
• docs: patch Series.cat page (#755)

🛠️ Other improvements

• chore: simplify Expr.clip (#777)
• chore: tests series_only refactor (#766)

Thank you to all our contributors for making this release possible!
@FBruzzesi, @MarcoGorelli, @lucianosrp and @muddi900

Narwhals v1.3.0

Changes

🚀 Performance improvements

• perf: keep nw.len totally lazy for dask (#749)

✨ Enhancements

• feat: Added Expr.any and Expr.all to Dask (#748)
• feat: arrow __mod__ method (#618)
• feat: allow tpc-h q1 to be run with Dask as backend (#747)
• feat: add NotImplementedError error for index changing methods Dask (#744)
• feat: Added Expr.abs to Dask (#728)
• feat: add Dask Expr.count (#731)
• feat: dask expr name namespace (#735)
• feat: add date to datetime namespace (#713)
• feat: dask expr dunder methods (#740)

Thank you to all our contributors for making this release possible!
@FBruzzesi, @MarcoGorelli, @anopsy, @lucianosrp and @mistShard