docs: add lancedb integration page

daft/dataframe/dataframe.py

@@ -1195,7 +1195,7 @@ def write_lance(
           **kwargs: Additional keyword arguments to pass to the Lance writer.
 
         Note:
-            write_lance` requires python 3.9 or higher
+            `write_lance` requires python 3.9 or higher
 
         Examples:
             >>> import daft
@@ -1210,6 +1210,7 @@ def write_lance(
             ╰───────────────┴──────────────────┴─────────────────┴─────────╯
             <BLANKLINE>
             (Showing first 1 of 1 rows)
+            <BLANKLINE>
             >>> daft.read_lance("/tmp/lance/my_table.lance").collect()  # doctest: +SKIP
             ╭───────╮
             │ a     │
@@ -1226,6 +1227,7 @@ def write_lance(
             ╰───────╯
             <BLANKLINE>
             (Showing first 4 of 4 rows)
+            <BLANKLINE>
             >>> # Pass additional keyword arguments to the Lance writer
             >>> # All additional keyword arguments are passed to `lance.write_fragments`
             >>> df.write_lance("/tmp/lance/my_table.lance", mode="overwrite", max_bytes_per_file=1024)  # doctest: +SKIP

docs/integrations/lance.md

@@ -0,0 +1,65 @@
+# LanceDB
+
+[LanceDB](https://github.com/lancedb/lancedb) is an open-source database for vector-search built with persistent storage, which greatly simplifies retrieval, filtering and management of embeddings.
+
+Daft currently supports reading from and writing to a LanceDB table. To use Daft with LanceDB, you will need to install Daft with the 'lance' option specific like so:
+
+```bash
+pip install daft[lance]
+```
+
+## Create a DataFrame from a LanceDB table
+
+You can create a Daft DataFrame by reading from a local LanceDB table with [`read_lance`][daft.read_lance]:
+
+```python
+df = daft.read_lance("/tmp/lance/my_table.lance")
+df.show()
+```
+``` {title=Output}
+╭───────╮
+│ a     │
+│ ---   │
+│ Int64 │
+╞═══════╡
+│ 1     │
+├╌╌╌╌╌╌╌┤
+│ 2     │
+├╌╌╌╌╌╌╌┤
+│ 3     │
+├╌╌╌╌╌╌╌┤
+│ 4     │
+╰───────╯
+
+(Showing first 4 of 4 rows)
+```
+
+Likewise, you can also create a Daft DataFrame from reading a LanceDB table from a public S3 bucket:
+
+```python
+from daft.io import S3Config
+s3_config = S3Config(region="us-west-2", anonymous=True)
+df = daft.read_lance("s3://daft-public-data/lance/words-test-dataset", io_config=s3_config)
+df.show()
+```
+
+## Write to a LanceDB table
+
+You can write a Daft DataFrame to a LanceDB table with [`write_lance`][daft.DataFrame.write_lance]
+
+```python
+import daft
+df = daft.from_pydict({"a": [1, 2, 3, 4]})
+df.write_lance("/tmp/lance/my_table.lance")
+```
+``` {title=Output}
+╭───────────────┬──────────────────┬─────────────────┬─────────╮
+│ num_fragments ┆ num_deleted_rows ┆ num_small_files ┆ version │
+│ ---           ┆ ---              ┆ ---             ┆ ---     │
+│ Int64         ┆ Int64            ┆ Int64           ┆ Int64   │
+╞═══════════════╪══════════════════╪═════════════════╪═════════╡
+│ 1             ┆ 0                ┆ 1               ┆ 1       │
+╰───────────────┴──────────────────┴─────────────────┴─────────╯
+
+(Showing first 1 of 1 rows)
+```

mkdocs.yml

@@ -35,6 +35,7 @@ nav:
       - Apache Iceberg: integrations/iceberg.md
       - AWS Glue: integrations/glue.md
       - Delta Lake: integrations/delta_lake.md
+      - LanceDB: integrations/lance.md
       - S3 Tables: integrations/s3tables.md
       - Unity Catalog: integrations/unity_catalog.md
     - Storage: