improve IO docs ordering

pola-rs · Mar 7, 2024 · cac54df · cac54df
1 parent a5332bc
commit cac54df
Show file tree

Hide file tree

Showing 5 changed files with 13 additions and 9 deletions.
diff --git a/py-polars/docs/source/reference/io.rst b/py-polars/docs/source/reference/io.rst
@@ -45,8 +45,8 @@ Delta Lake
 .. autosummary::
    :toctree: api/
 
-   scan_delta
    read_delta
+   scan_delta
    DataFrame.write_delta
 
 Excel / ODS
@@ -64,9 +64,9 @@ Feather / IPC
    :toctree: api/
 
    read_ipc
+   read_ipc_schema
    read_ipc_stream
    scan_ipc
-   read_ipc_schema
    DataFrame.write_ipc
    DataFrame.write_ipc_stream
    LazyFrame.sink_ipc
@@ -96,8 +96,8 @@ Parquet
    :toctree: api/
 
    read_parquet
-   scan_parquet
    read_parquet_schema
+   scan_parquet
    DataFrame.write_parquet
    LazyFrame.sink_parquet
 

diff --git a/py-polars/polars/io/__init__.py b/py-polars/polars/io/__init__.py
@@ -21,8 +21,8 @@
     "read_delta",
     "read_excel",
     "read_ipc",
-    "read_ipc_stream",
     "read_ipc_schema",
+    "read_ipc_stream",
     "read_json",
     "read_ndjson",
     "read_ods",

diff --git a/py-polars/polars/io/csv/__init__.py b/py-polars/polars/io/csv/__init__.py
@@ -1,6 +1,8 @@
+from polars.io.csv.batched_reader import BatchedCsvReader
 from polars.io.csv.functions import read_csv, read_csv_batched, scan_csv
 
 __all__ = [
+    "BatchedCsvReader",
     "read_csv",
     "read_csv_batched",
     "scan_csv",

diff --git a/py-polars/polars/io/csv/batched_reader.py b/py-polars/polars/io/csv/batched_reader.py
@@ -110,14 +110,12 @@ def next_batches(self, n: int) -> list[DataFrame] | None:
         """
         Read `n` batches from the reader.
 
-        The `n` chunks will be parallelized over the
-        available threads.
+        These batches will be parallelized over the available threads.
 
         Parameters
         ----------
         n
-            Number of chunks to fetch.
-            This is ideally >= number of threads
+            Number of chunks to fetch; ideally this is >= number of threads.
 
         Examples
         --------

diff --git a/py-polars/tests/unit/io/test_csv.py b/py-polars/tests/unit/io/test_csv.py
@@ -16,6 +16,7 @@
 import polars as pl
 from polars._utils.various import normalize_filepath
 from polars.exceptions import ComputeError, NoDataError
+from polars.io.csv import BatchedCsvReader
 from polars.testing import assert_frame_equal, assert_series_equal
 
 if TYPE_CHECKING:
@@ -1414,8 +1415,9 @@ def test_csv_categorical_categorical_merge() -> None:
 
 def test_batched_csv_reader(foods_file_path: Path) -> None:
     reader = pl.read_csv_batched(foods_file_path, batch_size=4)
-    batches = reader.next_batches(5)
+    assert isinstance(reader, BatchedCsvReader)
 
+    batches = reader.next_batches(5)
     assert batches is not None
     assert len(batches) == 5
     assert batches[0].to_dict(as_series=False) == {
@@ -1431,10 +1433,12 @@ def test_batched_csv_reader(foods_file_path: Path) -> None:
         "sugars_g": [25, 0, 5, 11],
     }
     assert_frame_equal(pl.concat(batches), pl.read_csv(foods_file_path))
+
     # the final batch of the low-memory variant is different
     reader = pl.read_csv_batched(foods_file_path, batch_size=4, low_memory=True)
     batches = reader.next_batches(5)
     assert len(batches) == 5  # type: ignore[arg-type]
+
     batches += reader.next_batches(5)  # type: ignore[operator]
     assert_frame_equal(pl.concat(batches), pl.read_csv(foods_file_path))