sqlglot 30.17.0__tar.gz → 30.18.0__tar.gz

{sqlglot-30.17.0 → sqlglot-30.18.0}/CHANGELOG.md

@@ -1,6 +1,92 @@
 Changelog
 =========
 
+## [v30.17.0] - 2026-08-12
+### :boom: BREAKING CHANGES
+- due to [`5c12472`](https://github.com/tobymao/sqlglot/commit/5c12472a8016dff6eb8d9aab9c9a029adab2d514) - stop optimizing OR-conditioned joins and anti joins *(PR [#8118](https://github.com/tobymao/sqlglot/pull/8118) by [@georgesittas](https://github.com/georgesittas))*:
+
+  stop optimizing OR-conditioned joins and anti joins (#8118)
+
+- due to [`07c2241`](https://github.com/tobymao/sqlglot/commit/07c22415977c21b342d578bd01b0ea7352a8a726) - support `and`, `or` function syntax *(PR [#8065](https://github.com/tobymao/sqlglot/pull/8065) by [@fivetran-amrutabhimsenayachit](https://github.com/fivetran-amrutabhimsenayachit))*:
+
+  support `and`, `or` function syntax (#8065)
+
+- due to [`3d74d62`](https://github.com/tobymao/sqlglot/commit/3d74d62c323ad3097c690092e7a4eb86aabb2f2d) - prevent pruning columns referenced by GROUP BY / HAVING / QUALIFY *(PR [#8121](https://github.com/tobymao/sqlglot/pull/8121) by [@fivetran-kwoodbeck](https://github.com/fivetran-kwoodbeck))*:
+
+  prevent pruning columns referenced by GROUP BY / HAVING / QUALIFY (#8121)
+
+- due to [`7c56116`](https://github.com/tobymao/sqlglot/commit/7c561165fad44b1af9008dcdf1d636323a28ef58) - annoate lag for mysql *(PR [#8133](https://github.com/tobymao/sqlglot/pull/8133) by [@PiyaDaswadkar](https://github.com/PiyaDaswadkar))*:
+
+  annoate lag for mysql (#8133)
+
+- due to [`ac7b6d9`](https://github.com/tobymao/sqlglot/commit/ac7b6d922b8ea2c318efe767603dd276f7a1c71f) - do not prune projections under INTERSECT/EXCEPT ALL or GROUP BY ALL *(commit by [@georgesittas](https://github.com/georgesittas))*:
+
+  do not prune projections under INTERSECT/EXCEPT ALL or GROUP BY ALL
+
+- due to [`e8836e8`](https://github.com/tobymao/sqlglot/commit/e8836e8fb1a0c59ce40bbac9832565aa8898d3a5) - annotate width_bucket for postgres *(PR [#8135](https://github.com/tobymao/sqlglot/pull/8135) by [@PiyaDaswadkar](https://github.com/PiyaDaswadkar))*:
+
+  annotate width_bucket for postgres (#8135)
+
+- due to [`e7685a7`](https://github.com/tobymao/sqlglot/commit/e7685a7480fa698e77ee0a209bf29c93c825738a) - represent shadowed clause column refs as qualified Columns *(PR [#8128](https://github.com/tobymao/sqlglot/pull/8128) by [@georgesittas](https://github.com/georgesittas))*:
+
+  represent shadowed clause column refs as qualified Columns (#8128)
+
+- due to [`8a29419`](https://github.com/tobymao/sqlglot/commit/8a29419f98c88939080db8e4fe1df34225d9135a) - stop pruning implicit GROUP BY ALL keys during projection pushdown *(PR [#8129](https://github.com/tobymao/sqlglot/pull/8129) by [@fivetran-kwoodbeck](https://github.com/fivetran-kwoodbeck))*:
+
+  stop pruning implicit GROUP BY ALL keys during projection pushdown (#8129)
+
+- due to [`8ba0435`](https://github.com/tobymao/sqlglot/commit/8ba04354d37cb60031d3b14249743620dc7b2732) - escape `gen` delimited names (strings, etc) *(PR [#8134](https://github.com/tobymao/sqlglot/pull/8134) by [@georgesittas](https://github.com/georgesittas))*:
+
+  escape `gen` delimited names (strings, etc) (#8134)
+
+- due to [`9312212`](https://github.com/tobymao/sqlglot/commit/9312212f2f3bda9b8bca191679e22c386c11edf9) - properly support `UNION BY NAME` in qualify, type inference *(PR [#8136](https://github.com/tobymao/sqlglot/pull/8136) by [@georgesittas](https://github.com/georgesittas))*:
+
+  properly support `UNION BY NAME` in qualify, type inference (#8136)
+
+- due to [`3a76dfd`](https://github.com/tobymao/sqlglot/commit/3a76dfdbb0c4409e81fd30795c0d978fd827d23e) - annotate encode for postgresql *(PR [#8137](https://github.com/tobymao/sqlglot/pull/8137) by [@PiyaDaswadkar](https://github.com/PiyaDaswadkar))*:
+
+  annotate encode for postgresql (#8137)
+
+- due to [`b7386b7`](https://github.com/tobymao/sqlglot/commit/b7386b756675567f89a098b7a5dbe598264ec2d5) - scope DML / DDL query fragments properly *(PR [#8140](https://github.com/tobymao/sqlglot/pull/8140) by [@georgesittas](https://github.com/georgesittas))*:
+
+  scope DML / DDL query fragments properly (#8140)
+
+
+### :sparkles: New Features
+- [`31f3751`](https://github.com/tobymao/sqlglot/commit/31f375116dbd504615b0ebf50e4be917bc4e9305) - **trino**: parse WHILE/DO/END WHILE routine statements [CLAUDE] *(PR [#8122](https://github.com/tobymao/sqlglot/pull/8122) by [@rusackas](https://github.com/rusackas))*
+- [`07c2241`](https://github.com/tobymao/sqlglot/commit/07c22415977c21b342d578bd01b0ea7352a8a726) - **databricks**: support `and`, `or` function syntax *(PR [#8065](https://github.com/tobymao/sqlglot/pull/8065) by [@fivetran-amrutabhimsenayachit](https://github.com/fivetran-amrutabhimsenayachit))*
+- [`9815786`](https://github.com/tobymao/sqlglot/commit/9815786785604e8d580a229b31dc90b94a58b3ad) - **trino**: parse LOOP/REPEAT/ITERATE/LEAVE routine statements [CLAUDE] *(PR [#8130](https://github.com/tobymao/sqlglot/pull/8130) by [@rusackas](https://github.com/rusackas))*
+- [`7c56116`](https://github.com/tobymao/sqlglot/commit/7c561165fad44b1af9008dcdf1d636323a28ef58) - **optimizer**: annoate lag for mysql *(PR [#8133](https://github.com/tobymao/sqlglot/pull/8133) by [@PiyaDaswadkar](https://github.com/PiyaDaswadkar))*
+- [`e8836e8`](https://github.com/tobymao/sqlglot/commit/e8836e8fb1a0c59ce40bbac9832565aa8898d3a5) - **optimizer**: annotate width_bucket for postgres *(PR [#8135](https://github.com/tobymao/sqlglot/pull/8135) by [@PiyaDaswadkar](https://github.com/PiyaDaswadkar))*
+- [`3a76dfd`](https://github.com/tobymao/sqlglot/commit/3a76dfdbb0c4409e81fd30795c0d978fd827d23e) - **optimizer**: annotate encode for postgresql *(PR [#8137](https://github.com/tobymao/sqlglot/pull/8137) by [@PiyaDaswadkar](https://github.com/PiyaDaswadkar))*
+- [`98f3d66`](https://github.com/tobymao/sqlglot/commit/98f3d663c9f110216a5495e8d5b779d9d4e8cc61) - **executor**: implement ILIKE [CLAUDE] *(PR [#8144](https://github.com/tobymao/sqlglot/pull/8144) by [@jbylund](https://github.com/jbylund))*
+- [`9a8129b`](https://github.com/tobymao/sqlglot/commit/9a8129b6f2667673f24713f4b49162ebae1f699d) - **executor**: implement LENGTH [CLAUDE] *(PR [#8145](https://github.com/tobymao/sqlglot/pull/8145) by [@jbylund](https://github.com/jbylund))*
+
+### :bug: Bug Fixes
+- [`36bc768`](https://github.com/tobymao/sqlglot/commit/36bc768fc122337e9c79eaaee900610aef4b7d42) - **prql**: `_select_all` accepts the Optional its callers pass *(commit by [@tobymao](https://github.com/tobymao))*
+- [`5c12472`](https://github.com/tobymao/sqlglot/commit/5c12472a8016dff6eb8d9aab9c9a029adab2d514) - **optimizer**: stop optimizing OR-conditioned joins and anti joins *(PR [#8118](https://github.com/tobymao/sqlglot/pull/8118) by [@georgesittas](https://github.com/georgesittas))*
+  - :arrow_lower_right: *fixes issue [#8101](https://github.com/tobymao/sqlglot/issues/8101) opened by [@dev-willbird1936](https://github.com/dev-willbird1936)*
+- [`84447c0`](https://github.com/tobymao/sqlglot/commit/84447c08e98b4006c0a87209dfb23708c2f4f177) - **sqlite**: drop a dead `is not None` guard on a sentinel-typed Token *(commit by [@tobymao](https://github.com/tobymao))*
+- [`353d0ff`](https://github.com/tobymao/sqlglot/commit/353d0ffca96acbd5407e73b832cadf5c1fedc9fe) - **sqlite**: emit STORED for generated columns instead of PERSISTED *(PR [#8123](https://github.com/tobymao/sqlglot/pull/8123) by [@dgvj-work](https://github.com/dgvj-work))*
+- [`3d74d62`](https://github.com/tobymao/sqlglot/commit/3d74d62c323ad3097c690092e7a4eb86aabb2f2d) - **optimizer**: prevent pruning columns referenced by GROUP BY / HAVING / QUALIFY *(PR [#8121](https://github.com/tobymao/sqlglot/pull/8121) by [@fivetran-kwoodbeck](https://github.com/fivetran-kwoodbeck))*
+- [`ac7b6d9`](https://github.com/tobymao/sqlglot/commit/ac7b6d922b8ea2c318efe767603dd276f7a1c71f) - **optimizer**: do not prune projections under INTERSECT/EXCEPT ALL or GROUP BY ALL *(commit by [@georgesittas](https://github.com/georgesittas))*
+- [`8a29419`](https://github.com/tobymao/sqlglot/commit/8a29419f98c88939080db8e4fe1df34225d9135a) - **optimizer**: stop pruning implicit GROUP BY ALL keys during projection pushdown *(PR [#8129](https://github.com/tobymao/sqlglot/pull/8129) by [@fivetran-kwoodbeck](https://github.com/fivetran-kwoodbeck))*
+- [`8ba0435`](https://github.com/tobymao/sqlglot/commit/8ba04354d37cb60031d3b14249743620dc7b2732) - **optimizer**: escape `gen` delimited names (strings, etc) *(PR [#8134](https://github.com/tobymao/sqlglot/pull/8134) by [@georgesittas](https://github.com/georgesittas))*
+- [`9312212`](https://github.com/tobymao/sqlglot/commit/9312212f2f3bda9b8bca191679e22c386c11edf9) - **optimizer**: properly support `UNION BY NAME` in qualify, type inference *(PR [#8136](https://github.com/tobymao/sqlglot/pull/8136) by [@georgesittas](https://github.com/georgesittas))*
+- [`1dae76b`](https://github.com/tobymao/sqlglot/commit/1dae76b5833a47c9e2bc0a72a93d94d829377a0c) - **executor**: only truncate typed division when both operands are integers [CLAUDE] *(PR [#8138](https://github.com/tobymao/sqlglot/pull/8138) by [@jbylund](https://github.com/jbylund))*
+  - :arrow_lower_right: *fixes issue [#8132](https://github.com/tobymao/sqlglot/issues/8132) opened by [@jbylund](https://github.com/jbylund)*
+- [`7cab581`](https://github.com/tobymao/sqlglot/commit/7cab581ffea8df170402381d942b6ac0781d2125) - **executor**: honor the negate flag on LIKE and ILIKE [CLAUDE] *(PR [#8139](https://github.com/tobymao/sqlglot/pull/8139) by [@jbylund](https://github.com/jbylund))*
+- [`44b73a0`](https://github.com/tobymao/sqlglot/commit/44b73a002328aea0fe0f655921ce43e43084e2d3) - **executor**: fullmatch like semantics *(commit by [@georgesittas](https://github.com/georgesittas))*
+- [`8619582`](https://github.com/tobymao/sqlglot/commit/86195827b827a34d324d6e21e0076d20031fb8a6) - **optimizer**: preserve identifier meta when restoring JSON dot part case *(commit by [@georgesittas](https://github.com/georgesittas))*
+- [`b7386b7`](https://github.com/tobymao/sqlglot/commit/b7386b756675567f89a098b7a5dbe598264ec2d5) - **optimizer**: scope DML / DDL query fragments properly *(PR [#8140](https://github.com/tobymao/sqlglot/pull/8140) by [@georgesittas](https://github.com/georgesittas))*
+
+### :recycle: Refactors
+- [`e7685a7`](https://github.com/tobymao/sqlglot/commit/e7685a7480fa698e77ee0a209bf29c93c825738a) - **optimizer**: represent shadowed clause column refs as qualified Columns *(PR [#8128](https://github.com/tobymao/sqlglot/pull/8128) by [@georgesittas](https://github.com/georgesittas))*
+
+### :wrench: Chores
+- [`1183e88`](https://github.com/tobymao/sqlglot/commit/1183e88c9661f98220c1a3719b5570e2efd645d0) - **mypy**: Bump version to v.2.3.0.post1 *(PR [#8126](https://github.com/tobymao/sqlglot/pull/8126) by [@VaggelisD](https://github.com/VaggelisD))*
+
+
 ## [v30.16.0] - 2026-08-10
 ### :boom: BREAKING CHANGES
 - due to [`32d44c5`](https://github.com/tobymao/sqlglot/commit/32d44c50d0138762c812a9326bb394eebd862464) - annotate bit_and for mysql *(PR [#8038](https://github.com/tobymao/sqlglot/pull/8038) by [@PiyaDaswadkar](https://github.com/PiyaDaswadkar))*:
@@ -15397,3 +15483,4 @@ pip install "sqlglot[c]"   # compiled — faster, but no subclassing
 [v30.13.0]: https://github.com/tobymao/sqlglot/compare/v30.12.0...v30.13.0
 [v30.14.0]: https://github.com/tobymao/sqlglot/compare/v30.13.0...v30.14.0
 [v30.16.0]: https://github.com/tobymao/sqlglot/compare/v30.15.0...v30.16.0
+[v30.17.0]: https://github.com/tobymao/sqlglot/compare/v30.16.0...v30.17.0

{sqlglot-30.17.0/sqlglot.egg-info → sqlglot-30.18.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sqlglot
-Version: 30.17.0
+Version: 30.18.0
 Summary: An easily customizable SQL parser and transpiler
 Author-email: Toby Mao <toby.mao@gmail.com>
 License-Expression: MIT
@@ -19,7 +19,7 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Provides-Extra: dev
 Requires-Dist: duckdb>=0.6; extra == "dev"
-Requires-Dist: sqlglot-mypy>=2.3.0.post1; python_version >= "3.10" and extra == "dev"
+Requires-Dist: sqlglot-mypy>=2.3.0.post2; python_version >= "3.10" and extra == "dev"
 Requires-Dist: mypy; python_version < "3.10" and extra == "dev"
 Requires-Dist: setuptools_scm; extra == "dev"
 Requires-Dist: pandas; extra == "dev"
@@ -34,10 +34,10 @@ Requires-Dist: types-pytz; extra == "dev"
 Requires-Dist: typing_extensions; extra == "dev"
 Requires-Dist: pyperf; extra == "dev"
 Provides-Extra: c
-Requires-Dist: sqlglotc==30.17.0; python_version >= "3.10" and extra == "c"
+Requires-Dist: sqlglotc==30.18.0; python_version >= "3.10" and extra == "c"
 Provides-Extra: rs
 Requires-Dist: sqlglotrs==0.13.0; extra == "rs"
-Requires-Dist: sqlglotc==30.17.0; python_version >= "3.10" and extra == "rs"
+Requires-Dist: sqlglotc==30.18.0; python_version >= "3.10" and extra == "rs"
 Dynamic: license-file
 Dynamic: provides-extra
 

{sqlglot-30.17.0 → sqlglot-30.18.0}/setup.py

@@ -9,7 +9,7 @@ setup(
             "duckdb>=0.6",
             # sqlglot-mypy 1.20+ is the build dep for sqlglotc and only ships
             # for py3.10+; on py3.9 just use upstream mypy for type checking.
-            "sqlglot-mypy >= 2.3.0.post1; python_version >= '3.10'",
+            "sqlglot-mypy >= 2.3.0.post2; python_version >= '3.10'",
             "mypy; python_version < '3.10'",
             "setuptools_scm",
             "pandas",

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '30.17.0'
-__version_tuple__ = version_tuple = (30, 17, 0)
+__version__ = version = '30.18.0'
+__version_tuple__ = version_tuple = (30, 18, 0)
 
-__commit_id__ = commit_id = 'g9a8129b6f'
+__commit_id__ = commit_id = 'g2e86ded7d'

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/__init__.py

@@ -122,6 +122,8 @@ def __getattr__(name):
     if module_name:
         with _import_lock:
             module = importlib.import_module(f"sqlglot.dialects.{module_name}")
-        return getattr(module, name)
+        attr = getattr(module, name)
+        globals()[name] = attr
+        return attr
 
     raise AttributeError(f"module {__name__} has no attribute {name}")

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/bigquery.py

@@ -10,6 +10,7 @@ from sqlglot._typing import E
 from sqlglot.parsers.bigquery import BigQueryParser
 from sqlglot.generators.bigquery import BigQueryGenerator
 from sqlglot.dialects.dialect import (
+    ASCII_LOWER,
     Dialect,
     NormalizationStrategy,
 )
@@ -50,6 +51,7 @@ class BigQuery(Dialect):
 
     # https://cloud.google.com/bigquery/docs/reference/standard-sql/lexical#case_sensitivity
     NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE
+    ASCII_ONLY_NORMALIZATION = True
 
     # bigquery udfs are case sensitive
     NORMALIZE_FUNCTIONS = False
@@ -152,7 +154,7 @@ class BigQuery(Dialect):
                 or expression.meta_get("is_table")
             )
             if not case_sensitive:
-                expression.set("this", expression.this.lower())
+                expression.set("this", expression.this.translate(ASCII_LOWER))
 
             return t.cast(E, expression)
 

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/dialect.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import importlib
 import logging
+import string
 import typing as t
 import sys
 from collections.abc import Iterable, MutableSequence
@@ -11,7 +12,7 @@ from builtins import type as Type
 
 from sqlglot import exp
 from sqlglot.dialects import DIALECT_MODULE_NAMES
-from sqlglot.errors import ParseError
+from sqlglot.errors import ParseError, TokenError
 from sqlglot.generator import Generator, unsupported_args
 from sqlglot.expressions import apply_index_offset
 from sqlglot.helper import (
@@ -77,6 +78,9 @@ UNESCAPED_SEQUENCES = {
 
 PLUGIN_GROUP_NAME = "sqlglot.dialects"
 
+ASCII_LOWER = str.maketrans(string.ascii_uppercase, string.ascii_lowercase)
+ASCII_UPPER = str.maketrans(string.ascii_lowercase, string.ascii_uppercase)
+
 
 class Dialects(str, Enum):
     """Dialects supported by SQLGLot."""
@@ -375,6 +379,10 @@ class Dialect(metaclass=_Dialect):
     NORMALIZATION_STRATEGY = NormalizationStrategy.LOWERCASE
     """Specifies the strategy according to which identifiers should be normalized."""
 
+    ASCII_ONLY_NORMALIZATION = False
+    """Whether identifiers are only normalized with respect to ASCII characters, e.g. `Ä` and
+    `ä` are different identifiers in DuckDB, but the same identifier in Spark."""
+
     IDENTIFIERS_CAN_START_WITH_DIGIT = False
     """Whether an unquoted identifier can start with a digit."""
 
@@ -473,6 +481,9 @@ class Dialect(metaclass=_Dialect):
     For example, such columns may be excluded from `SELECT *` queries.
     """
 
+    SUPPORTS_POSITIONAL_COLUMN_REFS = False
+    """Whether qualified `$N` references the Nth column of their source."""
+
     PREFER_CTE_ALIAS_COLUMN = False
     """
     Some dialects, such as Snowflake, allow you to reference a CTE column alias in the
@@ -1079,15 +1090,22 @@ class Dialect(metaclass=_Dialect):
                 )
             )
         ):
-            normalized = (
-                expression.this.upper()
-                if self.normalization_strategy
-                in (
-                    NormalizationStrategy.UPPERCASE,
-                    NormalizationStrategy.CASE_INSENSITIVE_UPPERCASE,
+            if self.normalization_strategy in (
+                NormalizationStrategy.UPPERCASE,
+                NormalizationStrategy.CASE_INSENSITIVE_UPPERCASE,
+            ):
+                normalized = (
+                    expression.this.translate(ASCII_UPPER)
+                    if self.ASCII_ONLY_NORMALIZATION
+                    else expression.this.upper()
                 )
-                else expression.this.lower()
-            )
+            else:
+                normalized = (
+                    expression.this.translate(ASCII_LOWER)
+                    if self.ASCII_ONLY_NORMALIZATION
+                    else expression.this.lower()
+                )
+
             expression.set("this", normalized)
 
         return expression
@@ -1157,7 +1175,7 @@ class Dialect(metaclass=_Dialect):
                 path_text = f"[{path_text}]"
             try:
                 return parse_json_path(path_text, self)
-            except ParseError as e:
+            except (ParseError, TokenError) as e:
                 if self.STRICT_JSON_PATH_SYNTAX and not path_text.lstrip().startswith(
                     ("lax", "strict")
                 ):
@@ -1247,12 +1265,20 @@ def if_sql(
     return _if_sql
 
 
-def arrow_json_extract_sql(self: Generator, expression: JSON_EXTRACT_TYPE) -> str:
+def arrow_json_extract_sql(
+    self: Generator,
+    expression: JSON_EXTRACT_TYPE,
+    op: str | None = None,
+) -> str:
     this = expression.this
     if self.JSON_TYPE_REQUIRED_FOR_EXTRACTION and isinstance(this, exp.Literal) and this.is_string:
         this.replace(exp.cast(this, exp.DType.JSON))
 
-    return self.binary(expression, "->" if isinstance(expression, exp.JSONExtract) else "->>")
+    if isinstance(expression.expression, (exp.Binary, exp.Predicate, exp.Not)):
+        expression.set("expression", exp.paren(expression.expression, copy=False))
+
+    op = op or ("->" if isinstance(expression, exp.JSONExtract) else "->>")
+    return self.binary(expression, op)
 
 
 def inline_array_sql(self: Generator, expression: exp.Expr) -> str:
@@ -1838,7 +1864,9 @@ def trim_sql(self: Generator, expression: exp.Trim, default_trim_type: str = "")
 
 
 def concat_to_dpipe_sql(self: Generator, expression: exp.Concat) -> str:
-    return self.sql(reduce(lambda x, y: exp.DPipe(this=x, expression=y), expression.expressions))
+    return self.sql(
+        reduce(lambda x, y: exp.DPipe(this=x, expression=y), self.convert_concat_args(expression))
+    )
 
 
 def concat_ws_to_dpipe_sql(self: Generator, expression: exp.ConcatWs) -> str:
@@ -1901,7 +1929,7 @@ def binary_from_function(expr_type: Type[B]) -> t.Callable[[BuilderArgs], B]:
 
 # Used to represent DATE_TRUNC in Doris, Postgres and Starrocks dialects
 def build_timestamp_trunc(args: BuilderArgs) -> exp.TimestampTrunc:
-    return exp.TimestampTrunc(this=seq_get(args, 1), unit=seq_get(args, 0))
+    return exp.TimestampTrunc(this=seq_get(args, 1), unit=seq_get(args, 0), zone=seq_get(args, 2))
 
 
 def build_trunc(
@@ -2623,6 +2651,16 @@ def regexp_replace_global_modifier(expression: exp.RegexpReplace) -> exp.Expr |
     return modifiers
 
 
+def nth_value_from_sql(self: Generator, expression: exp.NthValue) -> str:
+    this = self.func("NTH_VALUE", expression.this, expression.args.get("offset"))
+    from_first = expression.args.get("from_first")
+
+    if from_first is None:
+        return this
+
+    return f"{this} FROM {'FIRST' if from_first else 'LAST'}"
+
+
 def getbit_sql(self: Generator, expression: exp.Getbit) -> str:
     """
     Generates SQL for Getbit according to DuckDB and Postgres, transpiling it if either:

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/duckdb.py

@@ -28,6 +28,7 @@ class DuckDB(Dialect):
 
     # https://duckdb.org/docs/sql/introduction.html#creating-a-new-table
     NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE
+    ASCII_ONLY_NORMALIZATION = True
 
     DATE_PART_MAPPING = {
         **Dialect.DATE_PART_MAPPING,

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/mysql.py

@@ -44,8 +44,11 @@ class MySQL(Dialect):
         "%u": "%W",
         "%k": "%-H",
         "%l": "%-I",
+        "%r": "%I:%M:%S %p",
         "%T": "%H:%M:%S",
         "%W": "%A",
+        "%x": "%G",
+        # %v (ISO week) is unmapped due to collision with %V (roundtrip issue)
     }
 
     VALID_INTERVAL_UNITS = {

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/postgres.py

@@ -11,6 +11,7 @@ from sqlglot.typing.postgres import EXPRESSION_METADATA
 class Postgres(Dialect):
     EXPRESSION_METADATA = EXPRESSION_METADATA.copy()
     INDEX_OFFSET = 1
+    ASCII_ONLY_NORMALIZATION = True
     # Normalizing `x IS NOT NULL` to `NOT x IS NULL` is unsafe due to row values,
     # e.g. `ROW(1, NULL) IS NOT NULL` is false whereas `NOT ROW(1, NULL) IS NULL` is true
     NORMALIZE_NOT_NULL = False
@@ -79,6 +80,8 @@ class Postgres(Dialect):
         HEREDOC_TAG_IS_IDENTIFIER = True
         HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER
 
+        COMMANDS = {*tokens.Tokenizer.COMMANDS, TokenType.LOCK}
+
         KEYWORDS = {
             **tokens.Tokenizer.KEYWORDS,
             "~": TokenType.RLIKE,

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/snowflake.py

@@ -22,6 +22,7 @@ class Snowflake(Dialect):
     TIME_FORMAT = "'YYYY-MM-DD HH24:MI:SS'"
     SUPPORTS_USER_DEFINED_TYPES = False
     PREFER_CTE_ALIAS_COLUMN = True
+    SUPPORTS_POSITIONAL_COLUMN_REFS = True
     TABLESAMPLE_SIZE_IS_PERCENT = True
     COPY_PARAMS_ARE_CSV = False
     ARRAY_AGG_INCLUDES_NULLS = None

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/sqlite.py

@@ -13,6 +13,7 @@ from sqlglot.tokens import TokenType
 class SQLite(Dialect):
     # https://sqlite.org/forum/forumpost/5e575586ac5c711b?raw
     NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE
+    ASCII_ONLY_NORMALIZATION = True
     TYPED_DIVISION = True
     SAFE_DIVISION = True
     SAFE_TO_ELIMINATE_DOUBLE_NEGATION = False

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/starrocks.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from sqlglot import exp
 from sqlglot.dialects.mysql import MySQL
 from sqlglot.generators.starrocks import StarRocksGenerator
 from sqlglot.parsers.starrocks import StarRocksParser
@@ -10,6 +11,10 @@ class StarRocks(MySQL):
     STRICT_JSON_PATH_SYNTAX = False
     INDEX_OFFSET = 1
 
+    DEFAULT_FUNCTIONS_COLUMN_NAMES = {
+        exp.GenerateSeries: "generate_series",
+    }
+
     class Tokenizer(MySQL.Tokenizer):
         KEYWORDS = {
             **MySQL.Tokenizer.KEYWORDS,

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/dialects/tsql.py

@@ -19,6 +19,7 @@ class TSQL(Dialect):
     TYPED_DIVISION = True
     CONCAT_COALESCE = True
     CONCAT_WS_COALESCE = True
+    JSON_EXTRACT_SCALAR_SCALAR_ONLY = True
     NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE
     ALTER_TABLE_ADD_REQUIRED_FOR_EACH_COLUMN = False
 

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/executor/context.py

@@ -19,16 +19,27 @@ class Context:
     evaluation of aggregation functions.
     """
 
-    def __init__(self, tables: dict[str, Table], env: dict | None = None) -> None:
+    def __init__(
+        self, tables: dict[str, Table], env: dict | None = None, outer: dict | None = None
+    ) -> None:
         """
         Args
             tables: representing the scope of the current execution context.
             env: dictionary of functions within the execution context.
+            outer: readers of an enclosing query, for a subquery run once per outer row.
+                Own tables shadow them, as an inner alias shadows an outer one in SQL.
         """
         self.tables = tables
         self._table: Table | None = None
-        self.range_readers = {name: table.range_reader for name, table in self.tables.items()}
-        self.row_readers = {name: table.reader for name, table in tables.items()}
+        range_readers = {name: table.range_reader for name, table in self.tables.items()}
+        row_readers = {name: table.reader for name, table in tables.items()}
+
+        if outer:
+            range_readers = {**outer, **range_readers}
+            row_readers = {**outer, **row_readers}
+
+        self.range_readers = range_readers
+        self.row_readers = row_readers
         self.env = {**ENV, **(env or {}), "scope": self.row_readers}
 
     def eval(self, code):
@@ -56,7 +67,7 @@ class Context:
 
     @property
     def columns(self) -> tuple:
-        return self.table.columns
+        return self.table.columns if self.tables else ()
 
     def __iter__(self):
         self.env["scope"] = self.row_readers

{sqlglot-30.17.0 → sqlglot-30.18.0}/sqlglot/executor/env.py

@@ -179,10 +179,15 @@ def cast(this, to):
     raise NotImplementedError(f"Casting {this} to '{to}' not implemented.")
 
 
+FIRST = 0
+LAST = 1
+NULL_PLACEHOLDER = 0
+
+
 def ordered(this, desc, nulls_first):
-    if desc:
-        return reverse_key(this)
-    return this
+    if this is None:
+        return (FIRST if nulls_first else LAST, NULL_PLACEHOLDER)
+    return (LAST if nulls_first else FIRST, reverse_key(this) if desc else this)
 
 
 def _like(this, e, flags=0):
@@ -199,6 +204,14 @@ def interval(this, unit):
     return datetime.timedelta(**{unit.lower(): float(this)})
 
 
+@null_if_any
+def arrayconcat(*args):
+    result = []
+    for arg in args:
+        result.extend(arg if isinstance(arg, list) else [arg])
+    return result
+
+
 @null_if_any("this", "expression")
 def arraytostring(this, expression, null=None):
     return expression.join(x for x in (x if x is not None else null for x in this) if x is not None)
@@ -235,6 +248,7 @@ ENV = {
     "ABS": null_if_any(lambda this: abs(this)),
     "ADD": null_if_any(lambda e, this: e + this),
     "ARRAYANY": null_if_any(lambda arr, func: any(func(e) for e in arr)),
+    "ARRAYCONCAT": arrayconcat,
     "ARRAYTOSTRING": arraytostring,
     "BETWEEN": null_if_any(lambda this, low, high: low <= this and this <= high),
     "BITWISEAND": null_if_any(lambda this, e: this & e),
@@ -257,6 +271,7 @@ ENV = {
     "GTE": null_if_any(lambda this, e: this >= e),
     "IF": lambda predicate, true, false: true if predicate else false,
     "IN": sql_in,
+    "INT": null_if_any(int),
     "INTDIV": null_if_any(lambda e, this: e // this),
     "INTERVAL": interval,
     "JSONEXTRACT": jsonextract,
@@ -275,7 +290,8 @@ ENV = {
     "NOT": sql_not,
     "OR": sql_or,
     "ORDERED": ordered,
-    "POW": pow,
+    "POW": null_if_any(pow),
+    "REVERSE": null_if_any(lambda this: this[::-1]),
     "RIGHT": null_if_any(lambda this, e: this[-e:]),
     "ROUND": null_if_any(lambda this, decimals=None, truncate=None: round(this, ndigits=decimals)),
     "STRPOSITION": str_position,