Changes:
- Breaking : New add_table method in Schema ABC.
- New: SQLGlot now supports the PySpark dataframe API. This is still relatively experimental.
Changes:
- Improvement: Pretty generator now takes max_text_width which breaks segments into new lines
- New: exp.to_table helper to turn table names into table expression objects
- New: int[] type parsers
- New: annotations are now generated in sql
Changes:
-
Breaking: DISTINCT within functions now take in multiple values eg. COUNT(DISTINCT a, b). exp.Distinct no longer uses
thisand now uses the expressions property -
New: Expression False kwargs are now excluded from equality checks
-
New: Parse DESCRIBE and CREATE SCHEMA
-
New: DELETE and VALUES builder
-
New: Unused CTE and JOINS are now removed in the optimizer
Changes:
-
New: Snowflake table literals
-
New: Anti and semi joins
-
New: Vacuum as a command
-
New: Stored procedures
-
New: Reweriting derived tables as CTES
-
Improvement: Various clickhouse improvements
-
Improvement: Optimizer predicate pushdown
-
Breaking: DATE_DIFF default renamed to DATEDIFF
Changes:
-
New: TSQL support
-
Breaking: Removed $ from tokenizer, added @ placeholders
-
Improvement: Nodes can now be removed in transform and replace 8cd81c3
-
Improvement: Snowflake timestamp support
-
Improvement: Property conversion for CTAS Builder
-
Improvement: Tokenizers are now unique per dialect instance
Changes: