docs: fix readthedocs timeouts (#4350)

Author: rchowell

.readthedocs.yaml

@@ -7,10 +7,7 @@ build:
 
 python:
   install:
-  - requirements: requirements-dev.txt
+  - requirements: requirements-doc.txt
 
 mkdocs:
   configuration: mkdocs.yml
-
-formats:
-- htmlzip

daft/catalog/__init__.py

@@ -364,10 +364,8 @@ def from_glue(
 
         Args:
             name (str): glue database name
-            type (Literal["iceberg"]): optional catalog type
             client: optional boto3 client
             session: optional boto3 session
-            **options: additional options for boto3 client creation
 
         Returns:
             Catalog: new daft catalog instance backed by AWS Glue.
@@ -410,9 +408,6 @@ def create_namespace(self, identifier: Identifier | str):
 
         Args:
             identifier (Identifier | str): namespace identifier
-
-        Returns:
-            None
         """
         raise NotImplementedError
 
@@ -421,9 +416,6 @@ def create_namespace_if_not_exists(self, identifier: Identifier | str):
 
         Args:
             identifier (Identifier | str): namespace identifier
-
-        Returns:
-            None
         """
         if not self.has_namespace(identifier):
             self.create_namespace(identifier)
@@ -587,13 +579,9 @@ class Identifier(Sequence):
     def __init__(self, *parts: str):
         """Creates an Identifier from its parts.
 
-        Returns:
-            Identifier: A new identifier.
-
         Examples:
             >>> from daft.catalog import Identifier
             >>> Identifier("namespace", "table")
-
         """
         if len(parts) < 1:
             raise ValueError("Identifier requires at least one part.")
@@ -770,7 +758,7 @@ def from_unity(table: object) -> Table:
         """Returns a Daft Table instance from a Unity table.
 
         Args:
-            table
+            table (object): unity table instance.
         """
         try:
             from daft.catalog.__unity import UnityTable
@@ -813,11 +801,11 @@ def _validate_options(method: str, input: dict[str, Any], valid: set[str]):
     ###
 
     @abstractmethod
-    def read(self, **options) -> DataFrame:
+    def read(self, **options: dict[str, Any]) -> DataFrame:
         """Creates a new DataFrame from this table.
 
         Args:
-            **options: additional format-dependent read options
+            **options (dict[str,Any]): additional format-dependent read options
 
         Returns:
             DataFrame: new DataFrame instance
@@ -850,25 +838,25 @@ def show(self, n: int = 8) -> None:
     ###
 
     @abstractmethod
-    def write(self, df: DataFrame, mode: Literal["append", "overwrite"] = "append", **options) -> None:
+    def write(self, df: DataFrame, mode: Literal["append", "overwrite"] = "append", **options: dict[str, Any]) -> None:
         """Writes the DataFrame to this table.
 
         Args:
             df (DataFrame): datafram to write
             mode (str): write mode such as 'append' or 'overwrite'
-            **options: additional format-dependent write options
+            **options (dict[str,Any]): additional format-dependent write options
         """
 
-    def append(self, df: DataFrame, **options) -> None:
+    def append(self, df: DataFrame, **options: dict[str, Any]) -> None:
         """Appends the DataFrame to this table.
 
         Args:
             df (DataFrame): dataframe to append
-            **options: additional format-dependent write options
+            **options (dict[str,Any]): additional format-dependent write options
         """
         self.write(df, mode="append", **options)
 
-    def overwrite(self, df: DataFrame, **options) -> None:
+    def overwrite(self, df: DataFrame, **options: dict[str, Any]) -> None:
         """Overwrites this table with the given DataFrame.
 
         Args:

daft/io/__shim.py

@@ -34,7 +34,7 @@ def name(self) -> str:
         return self._source.name
 
     def display_name(self) -> str:
-        return f"DataFrameSource({self.name()})"
+        return f"DataSource({self.name()})"
 
     def partitioning_keys(self) -> list[PartitionField]:
         return []

daft/io/pushdowns.py

@@ -29,7 +29,7 @@ class Pushdowns:
     to convert the rust expressions to this python pushdowns class.
 
     Attributes:
-        projections (list[Term] | None): Optional list of Terms (typically column references) to project.
+        projections (list[Reference | None): Optional list of references to project.
         predicate (Term | None): Optional filter predicate to apply to rows.
         limit (int | None): Optional limit on the number of rows to return.
     """

daft/session.py

@@ -169,9 +169,6 @@ def detach_catalog(self, alias: str):
 
         Args:
             alias (str): catalog alias to detach
-
-        Returns:
-            None
         """
         return self._session.detach_catalog(alias)
 
@@ -180,9 +177,6 @@ def detach_table(self, alias: str):
 
         Args:
             alias (str): catalog alias to detach
-
-        Returns:
-            None
         """
         return self._session.detach_table(alias)
 
@@ -294,9 +288,6 @@ def drop_namespace(self, identifier: Identifier | str):
 
         Args:
             identifier (Identifier|str): table identifier
-
-        Returns:
-            None
         """
         if not (catalog := self.current_catalog()):
             raise ValueError("Cannot drop a namespace without a current catalog")
@@ -307,9 +298,6 @@ def drop_table(self, identifier: Identifier | str):
 
         Args:
             identifier (Identifier|str): table identifier
-
-        Returns:
-            None
         """
         if not (catalog := self.current_catalog()):
             raise ValueError("Cannot drop a table without a current catalog")
@@ -337,9 +325,6 @@ def use(self, identifier: Identifier | str | None = None):
     def current_catalog(self) -> Catalog | None:
         """Get the session's current catalog or None.
 
-        Args:
-            None
-
         Returns:
             Catalog: current catalog or None if one is not set
         """
@@ -348,9 +333,6 @@ def current_catalog(self) -> Catalog | None:
     def current_namespace(self) -> Identifier | None:
         """Get the session's current namespace or None.
 
-        Args:
-            None
-
         Returns:
             Identifier: current namespace or none if one is not set
         """
@@ -474,9 +456,6 @@ def set_catalog(self, identifier: str | None):
         Args:
             identifier (str): sets the current catalog
 
-        Returns:
-            None
-
         Raises:
             ValueError: If the catalog does not exist.
         """
@@ -501,18 +480,15 @@ def write_table(
         identifier: Identifier | str,
         df: DataFrame,
         mode: Literal["append", "overwrite"] = "append",
-        **options,
+        **options: dict[str, Any],
     ):
         """Writes the DataFrame to the table specified by the identifier.
 
         Args:
             identifier (Identifier|str): table identifier
             df (DataFrame): dataframe to write
             mode ("append"|"overwrite"): write mode, defaults to "append"
-            options**: additional, format-specific write options
-
-        Returns:
-            None
+            **options (dict[str,Any]): additional, format-specific write options
         """
         if isinstance(identifier, str):
             identifier = Identifier.from_str(identifier)

docs/api/dataframe_creation.md

@@ -0,0 +1,81 @@
+# DataFrame Creation
+
+Daft offers a variety of approaches to creating a DataFrame from various data sources like in-memory data, files, data catalogs, and integrations. Learn more about [Creating DataFrames](../core_concepts.md#creating-a-dataframe) in Daft User Guide.
+
+## SQL
+
+::: daft.sql.sql.sql
+    options:
+        heading_level: 3
+
+## In-Memory Data
+
+::: daft.from_pylist
+    options:
+        heading_level: 3
+
+::: daft.from_pydict
+    options:
+        heading_level: 3
+
+::: daft.from_arrow
+    options:
+        heading_level: 3
+
+::: daft.from_pandas
+    options:
+        heading_level: 3
+
+## Files
+
+::: daft.read_parquet
+    options:
+        heading_level: 3
+
+::: daft.read_csv
+    options:
+        heading_level: 3
+
+::: daft.read_json
+    options:
+        heading_level: 3
+
+::: daft.read_warc
+    options:
+        heading_level: 3
+
+::: daft.from_glob_path
+    options:
+        heading_level: 3
+
+## Data Catalogs
+
+::: daft.read_iceberg
+    options:
+        heading_level: 3
+
+::: daft.read_deltalake
+    options:
+        heading_level: 3
+
+::: daft.read_hudi
+    options:
+        heading_level: 3
+
+## Integrations
+
+::: daft.from_ray_dataset
+    options:
+        heading_level: 3
+
+::: daft.from_dask_dataframe
+    options:
+        heading_level: 3
+
+::: daft.read_sql
+    options:
+        heading_level: 3
+
+::: daft.read_lance
+    options:
+        heading_level: 3

docs/api/expressions.md

@@ -106,13 +106,3 @@ Daft Expressions allow you to express some computation that needs to happen in a
         filters: ["!^_"]
         toc_label: Expression.embedding
         heading: Expression.embedding
-
-## Visitor
-
-::: daft.expressions.visitor.ExpressionVisitor
-    options:
-        filters: ["!^_", "!_$"]
-
-::: daft.expressions.visitor.PredicateVisitor
-    options:
-        filters: ["!^_"]