v12.1.0 - 2023-05-05
0b46fa6- remove bigquery workaround for values type inference (commit by @tobymao)dfae784- sqlite no table options closes #1553 (commit by @tobymao)aef9cfa- double json spark closes #1547 (commit by @tobymao)
v12.0.0 - 2023-05-04
862cbeb- Use alternative transform for dialects that do not support distinct on (PR #1524 by @crericha)52c80e0- Support regex function in Starrocks (PR #1528 by @acreux)00b4779- spark: new Spark2 dialect, improve DATEDIFF sql generation BREAKING (PR #1529 by @GeorgeSittas)e2593ba- Add BYTES type to BigQuery dialect (PR #1536 by @relud)911e4e9- distinct on builder (commit by @tobymao)
5c26f56- windowspec generator (commit by @tobymao)e7111ba- trino: wrap SEQUENCE in an UNNEST call if used as a source (PR #1527 by @GeorgeSittas)3d964c6- use preprocess instead of expanding transform dicts BREAKING (PR #1525 by @GeorgeSittas)5fc27d3- trim with double pipes (commit by @tobymao)9778c16- create table options for bigquery closes #1531 (commit by @tobymao)55dc509- bigquery: allow 2nd argument for PERCENTILE_[CONT|DISC] (PR #1537 by @GeorgeSittas)5c59747- qualifying correlated subqueries (commit by @tobymao)19c0490- parse unnest array type closes #1532 (commit by @tobymao)51ca411- expand laterals first if no schema is present (commit by @tobymao)abfbce2- bigquery udf existing func clash closes #1535 (commit by @tobymao)444dd94- Spark: Add DOUBLE, FLOAT cast functions (PR #1530 by @vegarsti)0578d6d- oracle: allow parsing of @dblink in table names (PR #1540 by @GeorgeSittas)b7e08cc- duckdb: parse DATEDIFF correctly (PR #1546 by @GeorgeSittas)f21abb7- oracle: set post_tablesample_alias=True to fix alias parsing (PR #1548 by @GeorgeSittas)8ebba48- binary_double/float types closes #1543 (commit by @tobymao)2a6a3e7- make some SQL builders pure (PR #1526 by @GeorgeSittas)99310c4- duckdb: remove parentheses from CurrentTimestamp, CurrentDate (PR #1551 by @GeorgeSittas)↘️ fixes issue #1550 opened by @BTheunissen
a113685- spark: cast UnixToTime to TIMESTAMP BREAKING (PR #1549 by @GeorgeSittas)↘️ fixes issue #1545 opened by @joshmarcus
v11.7.0 - 2023-05-02
20cacba- tsql, oracle: add support for NEXT VALUE FOR clause (PR #1521 by @GeorgeSittas)96bb150- builder methods for basic ops (PR #1516 by @tobymao)e11a5ce- add case when change func to parse BREAKING (commit by @tobymao)455b9e9- oracle: support KEEP (.. [FIRST|LAST] ..) window function syntax (PR #1522 by @GeorgeSittas)
94fed8c- call _parse_bitwise as a fallback for nested type args (PR #1515 by @GeorgeSittas)52ab03d- array with method calls (commit by @tobymao)c3db2b8- allow parsing 'if' as an identifier (PR #1517 by @GeorgeSittas)
d5d360a- Remove older CHANGELOG entries and let CI handle it from now on (commit by @GeorgeSittas)
v11.6.3 - 2023-05-01
80287dd- presto: transpile explode/posexplode into (cross join) unnest (PR #1501 by @GeorgeSittas)
efd8c05- redshift doesn't support locks (commit by @tobymao)f4ece7c- comment after paren closes #1504 (commit by @tobymao)85b2c00- postgres doesn't support plural interval closes #1503 (commit by @tobymao)2e0eee6- postgres date_part type closes #1506 (commit by @tobymao)2dcbc7f- tsql hashbytes closes #1508 (commit by @tobymao)5347c7a- change str to Expression in alias_ isinstance check (PR #1510 by @GeorgeSittas)f137815- BigQueryTIMESTAMPandTIMESTAMPTZtypes (PR #1511 by @plaflamme)6143491- allow $ to appear in postgres/redshift identifiers (PR #1512 by @GeorgeSittas)
adc526c- add note about conventional commit naming in CONTRIBUTING.md (commit by @GeorgeSittas)9421650- set action for automatic conventional changelog generation (PR #1513 by @GeorgeSittas)