github TobikoData/sqlmesh v0.17.0

latest releases: v0.219.0, v0.218.0, v0.209.3...
2 years ago

✨ Features

🪲 Bug fixes

  • Use correct quote policy in dbt projects. Create correct relation instances in the adapter (#1058) @izeigerman
  • More bigquery table api issues (#1059) @tobymao
  • Parse cluster_by in dbt models before converting to SQLMesh models (#1060) @izeigerman
  • Map relations to snapshot physical tables in the dbt adapter (#1064) @izeigerman
  • Bigquery catalog ast @tobymao
  • Prevent creation of schema for symbolic models (#1061) @izeigerman
  • Always set global variables for dbt packages. Manually extract macro references in additional to the ones fetched from the manifest (#1069) @izeigerman
  • Support int64 range partitioning for BQ dbt projects (#1070) @izeigerman
  • Allow multiple statements after rendering jinja for pre- / post-statements (#1072) @izeigerman
  • Handle optional model and audit paths in the API (#1071) @vchan
  • Propagate table properties when replacing query (#1076) @izeigerman
  • Make sure to skip a node only once when concurrently processing a DAG (#1079) @izeigerman
  • Support legacy property names in the dbt BQ target (#1080) @izeigerman
  • Fallback to model dialect when rendering an audit query (#1081) @izeigerman
  • Support dbt_utils.star prefix and suffix using dbt version of the macro (#1085) @crericha
  • Fetch columns-to-types mapping from the target DW when determining a type of the time column (#1083) @izeigerman
  • Make all BigQuery db calls retryable (#1084) @eakmanrq
  • Deterministic order when computing fingerprint for jinja macro packages (#1088) @izeigerman
  • Fix bigquery nested types (#1089) @tobymao
  • Update views only for snapshots that are new to the target environment. Close state sync connections properly (#1093) @izeigerman
  • (docs) Fix integration docs links @tobymao
  • Circular ref debug_mode @tobymao
  • Handle corrupt snapshots due to upgrades closes #1063 (#1091) @tobymao
  • Support older versions of BigQuery python library (#1096) @izeigerman
  • Preserve the model query cache after fetching snapshots from state sync (#1098) @izeigerman
  • Cli failure due to missing package (#1099) @z3z1ma
  • Include database in dbt source fqns when it differs from the project's database (#1101) @crericha
  • Logging import in CLI @izeigerman
  • Jinja rendering comments (#1105) @tobymao
  • Remove redundant get_snapshots call when promoting to a dev environments (#1107) @izeigerman
  • Determine duckdb database from db filename if not specified. (#1106) @crericha
  • Duckdb default catalog name is main for dbt version < 1.5. (#1109) @crericha
  • Use schema for table diff summary table (#1112) @izeigerman
  • Snapshot intervals erroneously classified as dev (#1113) @izeigerman
  • Invalid AST when constructing BQ parition by expressions for dbt models (#1114) @izeigerman
  • The skip-tests flag in CLI @izeigerman
  • Treat end date as inclusive only for models with a daily cadency (#1115) @izeigerman

🛠️ Refactor

  • Macro reference extraction doesn't need dialect specific sql. using generic for performance (#1067) @tobymao

🧹 Chore

👽 Other

Don't miss a new sqlmesh release

NewReleases is sending notifications on new releases.