From e26fc0e30ba90e7d4febad243713597b84c582ad Mon Sep 17 00:00:00 2001 From: peasee <98815791+peasee@users.noreply.github.com> Date: Mon, 21 Oct 2024 14:38:35 +1000 Subject: [PATCH 1/2] docs: Update DuckDB limitations --- spiceaidocs/docs/components/data-accelerators/duckdb.md | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/spiceaidocs/docs/components/data-accelerators/duckdb.md b/spiceaidocs/docs/components/data-accelerators/duckdb.md index 8d7cd3f2..7e1421dd 100644 --- a/spiceaidocs/docs/components/data-accelerators/duckdb.md +++ b/spiceaidocs/docs/components/data-accelerators/duckdb.md @@ -46,7 +46,6 @@ datasets: - `SELECT MAP(['key1', 'key2', 'key3'], [10, 20, 30])` - `SELECT ['duck', 'goose', 'heron'];` - The DuckDB accelerator does not support `Decimal256` (76 digits), as it exceeds DuckDB's maximum Decimal width of 38 digits. -- The DuckDB accelerator does not support using the same `duckdb_file` for multiple accelerated datasets. - Updating a dataset with DuckDB acceleration while the Spice Runtime is running (hot-reload) will cause DuckDB accelerator query federation to disable until the Runtime is restarted. ::: @@ -57,4 +56,4 @@ When accelerating a dataset using `mode: memory` (the default), some or all of t In-memory limitations can be mitigated by storing acceleration data on disk, which is supported by [`duckdb`](./duckdb.md) and [`sqlite`](./sqlite.md) accelerators by specifying `mode: file`. -::: \ No newline at end of file +::: From 088c14e192ecdd53de6f6eef28fa05351ffcb07f Mon Sep 17 00:00:00 2001 From: peasee <98815791+peasee@users.noreply.github.com> Date: Mon, 21 Oct 2024 14:41:36 +1000 Subject: [PATCH 2/2] docs: Add boolean lists as limitation --- spiceaidocs/docs/components/data-accelerators/duckdb.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/spiceaidocs/docs/components/data-accelerators/duckdb.md b/spiceaidocs/docs/components/data-accelerators/duckdb.md index 7e1421dd..ab8e7ea2 100644 --- a/spiceaidocs/docs/components/data-accelerators/duckdb.md +++ b/spiceaidocs/docs/components/data-accelerators/duckdb.md @@ -36,7 +36,7 @@ datasets: :::warning[Limitations] -- The DuckDB accelerator does not support schemas with [field types](https://duckdb.org/docs/sql/data_types/overview): nested arrays/lists, UTF8/string arrays/lists, structs, nested structs or map fields. For example: +- The DuckDB accelerator does not support schemas with [field types](https://duckdb.org/docs/sql/data_types/overview): nested arrays/lists, UTF8/string arrays/lists, boolean arrays/lists, structs, nested structs or map fields. For example: - Supported: - `SELECT [1, 2, 3];` - `SELECT ['1992-09-20 11:30:00.123456789', 'epoch'::TIMESTAMP]` @@ -45,6 +45,7 @@ datasets: - `SELECT {'x': 1, 'y': 2, 'z': 3}` - `SELECT MAP(['key1', 'key2', 'key3'], [10, 20, 30])` - `SELECT ['duck', 'goose', 'heron'];` + - `SELECT [true, false];` - The DuckDB accelerator does not support `Decimal256` (76 digits), as it exceeds DuckDB's maximum Decimal width of 38 digits. - Updating a dataset with DuckDB acceleration while the Spice Runtime is running (hot-reload) will cause DuckDB accelerator query federation to disable until the Runtime is restarted.