Full Changelog: v2.29.0...v2.29.1

Bug Fixes

• Bound sqlglot version below 28 (#2275)

2.29.0 (2025-11-10)

Features

• Add bigframes.bigquery.st_regionstats to join raster data from Earth Engine (#2228) (10ec52f)
• Add DataFrame.resample and Series.resample (#2213) (c9ca02c)
• SQL Cell no longer escapes formatted string values (#2245) (d2d38f9)
• Support left_index and right_index for merge (#2220) (da9ba26)

Bug Fixes

• Correctly iterate over null struct values in ManagedArrowTable (#2209) (12e04d5)
• Simplify UnsupportedTypeError message (#2212) (6c9a18d)
• Support results with STRUCT and ARRAY columns containing JSON subfields in to_pandas_batches() (#2216) (3d8b17f)

Documentation

• Switch API reference docs to pydata theme (#2237) (9b86dcf)
• Update notebook for JSON subfields support in to_pandas_batches() (#2138) (5663d2a)

2.28.0 (2025-11-03)

Features

• Add bigframes.bigquery.st_simplify (#2210) (ecee2bc)
• Add Series.dt.day_name (#2218) (5e006e4)
• Polars engine supports std, var (#2215) (ef5e83a)
• Support INFORMATION_SCHEMA views in read_gbq (#1895) (d97cafc)
• Support some python standard lib callables in apply/combine (#2187) (86a2756)

Bug Fixes

• Correct connection normalization in blob system tests (#2222) (a0e1e50)
• Improve error handling in blob operations (#2194) (d410046)
• Resolve AttributeError in TableWidget and improve initialization (#1937) (4c4c9b1)

Documentation

• Update bq_dataframes_llm_output_schema.ipynb (#2004) (316ba9f)

2.27.0 (2025-10-24)

Features

• Add abs to dataframe (#2186) (c331dfe)
• Add df.groupby().corr()/cov() support (#2190) (ccd7c07)
• Add str accessor to index (#2179) (cd87ce0)
• Add support for np.isnan and np.isfinite ufuncs (#2188) (68723bc)
• Include local data bytes in the dry run report when available (#2185) (ee2c40c)
• Support len() on Groupby objects (#2183) (4191821)
• Support pa.json_(pa.string()) in struct/list if available (#2180) (5ec3cc0)

Documentation

• Update AI operators deprecation notice (#2182) (2c50310)

2.26.0 (2025-10-17)

⚠ BREAKING CHANGES

• turn Series.struct.dtypes into a property to match pandas (#2169)

Features

• Add df.sort_index(axis=1) (#2173) (ebf95e3)
• Enhanced multimodal error handling with verbose mode for blob image functions (#2024) (f9e28fe)
• Implement cos, sin, and log operations for polars compiler (#2170) (5613e44)
• Make all and any compatible with integer columns on Polars session (#2154) (6353d6e)

Bug Fixes

• blob.display() shows <NA> for null rows (#2158) (ddb4df0)
• Turn Series.struct.dtypes into a property to match pandas (#2169) (62f7e9f)

Documentation

• Clarify that only NULL values are handled by fillna/isna, not NaN (#2176) (8f27e73)
• Remove import bigframes.pandas as bpd boilerplate from many samples (#2147) (1a01ab9)

2.25.0 (2025-10-13)

Features

• Add barh, pie plot types (#2146) (5cc3c5b)
• Add Index.eq for consts, aligned objects (#2141) (8514200)
• Add output_schema parameter to ai.generate() (#2139) (ef0b0b7)
• Create session-scoped cut, DataFrame, MultiIndex, Index, Series, to_datetime, and to_timedelta methods (#2157) (5e1e809)
• Replace ML.GENERATE_TEXT with AI.GENERATE for audio transcription (#2151) (a410d0a)
• Support string literal inputs for AI functions (#2152) (7600001)

Bug Fixes

• Address typo in error message (#2142) (cdf2dd5)
• Avoid possible circular imports in global session (#2115) (095c0b8)
• Fix too many cluster columns requested by caching (#2155) (35c1c33)
• Show progress even in job optional queries (#2119) (1f48d3a)
• Yield row count from read session if otherwise unknown (#2148) (8997d4d)

Documentation

• Add a brief intro notebook for bbq AI functions (#2150) (1f434fb)
• Fix ai function related docs (#2149) (93a0749)
• Remove progress bar from getting started template (#2143) (d13abad)

2.24.0 (2025-10-07)

Features

• Add ai.classify() to bigframes.bigquery package (#2137) (56e5033)
• Add ai.generate() to bigframes.bigquery module (#2128) (3810452)
• Add ai.if_() and ai.score() to bigframes.bigquery package (#2132) (32502f4)

Bug Fixes

• Fix internal type errors with temporal accessors (#2125) (c390da1)
• Fix row count local execution bug (#2133) (ece0762)
• Join on, how args are now positional (#2140) (b711815)
• Only show JSON dtype warning when accessing dtypes directly (#2136) (eca22ee)
• Remove noisy AmbiguousWindowWarning from partial ordering mode (#2129) (4607f86)

Performance Improvements

• Scale read stream workers to cpu count (#2135) (67e46cd)

2.23.0 (2025-09-29)

Features

• Add ai.generate_double to bigframes.bigquery package (#2111) (6b8154c)

Bug Fixes

• Prevent invalid syntax for no-op .replace ops (#2112) (c311876)

Documentation

• Add timedelta notebook sample (#2124) (d1a9888)

2.22.0 (2025-09-25)

Features

• Add GroupBy.__iter__ (#1394) (c56a78c)
• Add ai.generate_int to bigframes.bigquery package (#2109) (af6b862)
• Add Groupby.describe() (#2088) (328a765)
• Implement Index.to_list() (#2106) (60056ca)
• Implement inplace parameter for DataFrame.drop (#2105) (3487f13)
• Support callable for series map method (#2100) (ac25618)
• Support df.info() with null index (#2094) (fb81eea)

Bug Fixes

• Avoid ibis fillna warning in compiler (#2113) (7ef667b)
• Negative start and stop parameter values in Series.str.slice() (#2104) (f57a348)
• Throw type error for incomparable join keys (#2098) (9dc9695)
• Transformers with non-standard column names throw errors (#2089) (a2daa3f)

2.21.0 (2025-09-17)

Features

• Add bigframes.bigquery.to_json (#2078) (0fc795a)
• Support average='binary' in precision_score() (#2080) (920f381)
• Support pandas series in ai.generate_bool (#2086) (a3de53f)

Bug Fixes

• Allow bigframes.options.bigquery.credentials to be None (#2092) (78f4001)