Merge branch 'main' into shuowei-anywidget-nested-strcut-array

Author: shuoweil

.librarian/state.yaml

@@ -1,7 +1,7 @@
 image: us-central1-docker.pkg.dev/cloud-sdk-librarian-prod/images-prod/python-librarian-generator@sha256:c8612d3fffb3f6a32353b2d1abd16b61e87811866f7ec9d65b59b02eb452a620
 libraries:
   - id: bigframes
-    version: 2.31.0
+    version: 2.32.0
     last_generated_commit: ""
     apis: []
     source_roots:

CHANGELOG.md

@@ -4,6 +4,34 @@
 
 [1]: https://pypi.org/project/bigframes/#history
 
+## [2.32.0](https://github.com/googleapis/google-cloud-python/compare/bigframes-v2.31.0...bigframes-v2.32.0) (2026-01-05)
+
+
+### Documentation
+
+* generate sitemap.xml for better search indexing (#2351) ([7d2990f1c48c6d74e2af6bee3af87f90189a3d9b](https://github.com/googleapis/google-cloud-python/commit/7d2990f1c48c6d74e2af6bee3af87f90189a3d9b))
+* update supported pandas APIs documentation links (#2330) ([ea71936ce240b2becf21b552d4e41e8ef4418e2d](https://github.com/googleapis/google-cloud-python/commit/ea71936ce240b2becf21b552d4e41e8ef4418e2d))
+* Add time series analysis notebook (#2328) ([369f1c0aff29d197b577ec79e401b107985fe969](https://github.com/googleapis/google-cloud-python/commit/369f1c0aff29d197b577ec79e401b107985fe969))
+
+
+### Features
+
+* Enable multi-column sorting in anywidget mode (#2360) ([1feb956e4762e30276e5b380c0633e6ed7881357](https://github.com/googleapis/google-cloud-python/commit/1feb956e4762e30276e5b380c0633e6ed7881357))
+* display series in anywidget mode (#2346) ([7395d418550058c516ad878e13567256f4300a37](https://github.com/googleapis/google-cloud-python/commit/7395d418550058c516ad878e13567256f4300a37))
+* Refactor TableWidget and to_pandas_batches (#2250) ([b8f09015a7c8e6987dc124e6df925d4f6951b1da](https://github.com/googleapis/google-cloud-python/commit/b8f09015a7c8e6987dc124e6df925d4f6951b1da))
+* Auto-plan complex reduction expressions (#2298) ([4d5de14ccdd05b1ac8f50c3fe71c35ab9e5150c1](https://github.com/googleapis/google-cloud-python/commit/4d5de14ccdd05b1ac8f50c3fe71c35ab9e5150c1))
+* Display custom single index column in anywidget mode (#2311) ([f27196260743883ed8131d5fd33a335e311177e4](https://github.com/googleapis/google-cloud-python/commit/f27196260743883ed8131d5fd33a335e311177e4))
+* add fit_predict method to ml unsupervised models (#2320) ([59df7f70a12ef702224ad61e597bd775208dac45](https://github.com/googleapis/google-cloud-python/commit/59df7f70a12ef702224ad61e597bd775208dac45))
+
+
+### Bug Fixes
+
+* vendor sqlglot bigquery dialect and remove package dependency (#2354) ([b321d72d5eb005b6e9295541a002540f05f72209](https://github.com/googleapis/google-cloud-python/commit/b321d72d5eb005b6e9295541a002540f05f72209))
+* bigframes.ml fit with eval data in partial mode avoids join on null index (#2355) ([7171d21b8c8d5a2d61081f41fa1109b5c9c4bc5f](https://github.com/googleapis/google-cloud-python/commit/7171d21b8c8d5a2d61081f41fa1109b5c9c4bc5f))
+* Improve strictness of nan vs None usage (#2326) ([481d938fb0b840e17047bc4b57e61af15b976e54](https://github.com/googleapis/google-cloud-python/commit/481d938fb0b840e17047bc4b57e61af15b976e54))
+* Correct DataFrame widget rendering in Colab (#2319) ([7f1d3df3839ec58f52e48df088057fc0df967da9](https://github.com/googleapis/google-cloud-python/commit/7f1d3df3839ec58f52e48df088057fc0df967da9))
+* Fix pd.timedelta handling in polars comipler with polars 1.36 (#2325) ([252644826289d9db7a8548884de880b3a4fccafd](https://github.com/googleapis/google-cloud-python/commit/252644826289d9db7a8548884de880b3a4fccafd))
+
 ## [2.31.0](https://github.com/googleapis/google-cloud-python/compare/bigframes-v2.30.0...bigframes-v2.31.0) (2025-12-10)
 
 

bigframes/core/compile/sqlglot/aggregations/windows.py

@@ -174,7 +174,7 @@ def _compile_group_by_key(key: ex.Expression) -> sge.Expression:
     if key.dtype == dtypes.FLOAT_DTYPE:
         expr = sge.Cast(this=expr, to="STRING")
     elif key.dtype == dtypes.GEO_DTYPE:
-        expr = sge.Cast(this=expr, to="BYTES")
+        expr = sge.func("ST_ASBINARY", expr)
     elif key.dtype == dtypes.JSON_DTYPE:
         expr = sge.func("TO_JSON_STRING", expr)
     return expr

bigframes/core/compile/sqlglot/expressions/blob_ops.py

@@ -29,9 +29,9 @@ def _(expr: TypedExpr) -> sge.Expression:
     return sge.func("OBJ.FETCH_METADATA", expr.expr)
 
 
-@register_unary_op(ops.ObjGetAccessUrl)
-def _(expr: TypedExpr) -> sge.Expression:
-    return sge.func("OBJ.GET_ACCESS_URL", expr.expr)
+@register_unary_op(ops.ObjGetAccessUrl, pass_op=True)
+def _(expr: TypedExpr, op: ops.ObjGetAccessUrl) -> sge.Expression:
+    return sge.func("OBJ.GET_ACCESS_URL", expr.expr, sge.convert(op.mode))
 
 
 @register_binary_op(ops.obj_make_ref_op)

bigframes/core/compile/sqlglot/expressions/string_ops.py

@@ -48,12 +48,14 @@ def _(expr: TypedExpr, op: ops.StrExtractOp) -> sge.Expression:
     # Cannot use BigQuery's REGEXP_EXTRACT function, which only allows one
     # capturing group.
     pat_expr = sge.convert(op.pat)
-    if op.n != 0:
-        pat_expr = sge.func("CONCAT", sge.convert(".*?"), pat_expr, sge.convert(".*"))
-    else:
+    if op.n == 0:
         pat_expr = sge.func("CONCAT", sge.convert(".*?("), pat_expr, sge.convert(").*"))
+        n = 1
+    else:
+        pat_expr = sge.func("CONCAT", sge.convert(".*?"), pat_expr, sge.convert(".*"))
+        n = op.n
 
-    rex_replace = sge.func("REGEXP_REPLACE", expr.expr, pat_expr, sge.convert(r"\1"))
+    rex_replace = sge.func("REGEXP_REPLACE", expr.expr, pat_expr, sge.convert(f"\\{n}"))
     rex_contains = sge.func("REGEXP_CONTAINS", expr.expr, sge.convert(op.pat))
     return sge.If(this=rex_contains, true=rex_replace, false=sge.null())
 

bigframes/version.py

@@ -12,8 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "2.31.0"
+__version__ = "2.32.0"
 
 # {x-release-please-start-date}
-__release_date__ = "2025-12-10"
+__release_date__ = "2026-01-05"
 # {x-release-please-end}

biome.json

@@ -0,0 +1,16 @@
+{
+  "formatter": {
+    "indentStyle": "space",
+    "indentWidth": 2
+  },
+  "javascript": {
+    "formatter": {
+      "quoteStyle": "single"
+    }
+  },
+  "css": {
+    "formatter": {
+      "quoteStyle": "single"
+    }
+  }
+}

tests/unit/core/compile/sqlglot/aggregations/test_windows.py

@@ -114,7 +114,7 @@ def test_apply_window_if_present_grouping_no_ordering(self):
         )
         self.assertEqual(
             result.sql(dialect="bigquery"),
-            "value OVER (PARTITION BY `col1`, CAST(`col2` AS STRING), TO_JSON_STRING(`col3`), CAST(`col4` AS BYTES))",
+            "value OVER (PARTITION BY `col1`, CAST(`col2` AS STRING), TO_JSON_STRING(`col3`), ST_ASBINARY(`col4`))",
         )
 
     def test_apply_window_if_present_range_bounded(self):

tests/unit/core/compile/sqlglot/expressions/snapshots/test_blob_ops/test_obj_get_access_url/out.sql

@@ -11,7 +11,7 @@ WITH `bfcte_0` AS (
 ), `bfcte_2` AS (
   SELECT
     *,
-    OBJ.GET_ACCESS_URL(`bfcol_4`) AS `bfcol_7`
+    OBJ.GET_ACCESS_URL(`bfcol_4`, 'R') AS `bfcol_7`
   FROM `bfcte_1`
 ), `bfcte_3` AS (
   SELECT

tests/unit/core/compile/sqlglot/expressions/snapshots/test_string_ops/test_str_extract/out.sql

@@ -5,13 +5,19 @@ WITH `bfcte_0` AS (
 ), `bfcte_1` AS (
   SELECT
     *,
+    IF(
+      REGEXP_CONTAINS(`string_col`, '([a-z]*)'),
+      REGEXP_REPLACE(`string_col`, CONCAT('.*?(', '([a-z]*)', ').*'), '\\1'),
+      NULL
+    ) AS `bfcol_1`,
     IF(
       REGEXP_CONTAINS(`string_col`, '([a-z]*)'),
       REGEXP_REPLACE(`string_col`, CONCAT('.*?', '([a-z]*)', '.*'), '\\1'),
       NULL
-    ) AS `bfcol_1`
+    ) AS `bfcol_2`
   FROM `bfcte_0`
 )
 SELECT
-  `bfcol_1` AS `string_col`
+  `bfcol_1` AS `zero`,
+  `bfcol_2` AS `one`
 FROM `bfcte_1`