TYP: read_sas

doc/source/whatsnew/v1.5.0.rst

@@ -461,6 +461,7 @@ Other API changes
   October 2022. (:issue:`46312`)
 - :func:`read_json` now raises ``FileNotFoundError`` (previously ``ValueError``) when input is a string ending in ``.json``, ``.json.gz``, ``.json.bz2``, etc. but no such file exists. (:issue:`29102`)
 - Operations with :class:`Timestamp` or :class:`Timedelta` that would previously raise ``OverflowError`` instead raise ``OutOfBoundsDatetime`` or ``OutOfBoundsTimedelta`` where appropriate (:issue:`47268`)
+- When :func:`read_sas` previously returned ``None``, it now returns an empty :class:`DataFrame` (:issue:`47410`)
 -
 
 .. ---------------------------------------------------------------------------

pandas/io/sas/sas7bdat.py

@@ -163,12 +163,12 @@ def __init__(
         self,
         path_or_buf: FilePath | ReadBuffer[bytes],
         index=None,
-        convert_dates=True,
-        blank_missing=True,
-        chunksize=None,
-        encoding=None,
-        convert_text=True,
-        convert_header_text=True,
+        convert_dates: bool = True,
+        blank_missing: bool = True,
+        chunksize: int | None = None,
+        encoding: str | None = None,
+        convert_text: bool = True,
+        convert_header_text: bool = True,
         compression: CompressionOptions = "infer",
     ) -> None:
 
@@ -361,9 +361,9 @@ def _get_properties(self) -> None:
                     self.encoding or self.default_encoding
                 )
 
-    def __next__(self):
+    def __next__(self) -> DataFrame:
         da = self.read(nrows=self.chunksize or 1)
-        if da is None:
+        if da.empty:
             self.close()
             raise StopIteration
         return da
@@ -732,7 +732,7 @@ def _process_format_subheader(self, offset: int, length: int) -> None:
         self.column_formats.append(column_format)
         self.columns.append(col)
 
-    def read(self, nrows: int | None = None) -> DataFrame | None:
+    def read(self, nrows: int | None = None) -> DataFrame:
 
         if (nrows is None) and (self.chunksize is not None):
             nrows = self.chunksize
@@ -744,7 +744,7 @@ def read(self, nrows: int | None = None) -> DataFrame | None:
             raise EmptyDataError("No columns to parse from file")
 
         if nrows > 0 and self._current_row_in_file_index >= self.row_count:
-            return None
+            return DataFrame()
 
         m = self.row_count - self._current_row_in_file_index
         if nrows > m:

pandas/io/sas/sas_xport.py

@@ -280,7 +280,7 @@ def __init__(
             self.close()
             raise
 
-    def close(self):
+    def close(self) -> None:
         self.handles.close()
 
     def _get_row(self):
@@ -463,7 +463,7 @@ def _missing_double(self, vec):
         return miss
 
     @Appender(_read_method_doc)
-    def read(self, nrows=None):
+    def read(self, nrows: int | None = None) -> pd.DataFrame:
 
         if nrows is None:
             nrows = self.nobs

pandas/io/sas/sasreader.py

@@ -38,17 +38,17 @@ class ReaderBase(metaclass=ABCMeta):
     """
 
     @abstractmethod
-    def read(self, nrows=None):
+    def read(self, nrows: int | None = None) -> DataFrame:
         pass
 
     @abstractmethod
-    def close(self):
+    def close(self) -> None:
         pass
 
-    def __enter__(self):
+    def __enter__(self) -> ReaderBase:
         return self
 
-    def __exit__(self, exc_type, exc_value, traceback):
+    def __exit__(self, exc_type, exc_value, traceback) -> None:
         self.close()
 
 

pyright_reportGeneralTypeIssues.json

@@ -105,8 +105,7 @@
         "pandas/io/parsers/base_parser.py",
         "pandas/io/parsers/c_parser_wrapper.py",
         "pandas/io/pytables.py",
-        "pandas/io/sas/sas7bdat.py",
-        "pandas/io/sas/sasreader.py",
+        "pandas/io/sas/sas_xport.py",
         "pandas/io/sql.py",
         "pandas/io/stata.py",
         "pandas/io/xml.py",