ENH: Improve read_{csv/fwf/table}

Author: bashtage

pandas-stubs/_libs/lib.pyi

@@ -1,3 +1,8 @@
+from enum import Enum
+
 no_default = None
 
+class NoDefault(Enum):
+    no_default: int
+
 def infer_dtype(value: object, skipna: bool = ...) -> str: ...

pandas-stubs/_typing.pyi

@@ -14,6 +14,7 @@ from typing import (
     AnyStr,
     Callable,
     Hashable,
+    Iterator,
     Literal,
     Mapping,
     Optional,
@@ -65,6 +66,12 @@ class BaseBuffer(Protocol): ...
 class ReadBuffer(BaseBuffer, Protocol[AnyStr_cov]): ...
 class WriteBuffer(BaseBuffer, Protocol[AnyStr_cov]): ...
 
+class ReadCsvBuffer(ReadBuffer[AnyStr_cov], Protocol[AnyStr_cov]):
+    def __iter__(self) -> Iterator[AnyStr_cov]: ...
+    def readline(self) -> AnyStr_cov: ...
+    @property
+    def closed(self) -> bool: ...
+
 FilePath = Union[str, PathLike[str]]
 
 Buffer = Union[IO[AnyStr], RawIOBase, BufferedIOBase, TextIOBase, TextIOWrapper, mmap]

pandas-stubs/core/frame.pyi

@@ -1862,55 +1862,6 @@ class DataFrame(NDFrame, OpsMixin):
         self, excel: _bool = ..., sep: _str | None = ..., **kwargs
     ) -> None: ...
     @overload
-    def to_csv(
-        self,
-        path_or_buf: FilePathOrBuffer | None,
-        sep: _str = ...,
-        na_rep: _str = ...,
-        float_format: _str | None = ...,
-        columns: Sequence[Hashable] | None = ...,
-        header: _bool | list[_str] = ...,
-        index: _bool = ...,
-        index_label: _bool | _str | Sequence[Hashable] | None = ...,
-        mode: _str = ...,
-        encoding: _str | None = ...,
-        compression: _str | Mapping[_str, _str] = ...,
-        quoting: int | None = ...,
-        quotechar: _str = ...,
-        line_terminator: _str | None = ...,
-        chunksize: int | None = ...,
-        date_format: _str | None = ...,
-        doublequote: _bool = ...,
-        escapechar: _str | None = ...,
-        decimal: _str = ...,
-        errors: _str = ...,
-        storage_options: dict[_str, Any] | None = ...,
-    ) -> None: ...
-    @overload
-    def to_csv(
-        self,
-        sep: _str = ...,
-        na_rep: _str = ...,
-        float_format: _str | None = ...,
-        columns: Sequence[Hashable] | None = ...,
-        header: _bool | list[_str] = ...,
-        index: _bool = ...,
-        index_label: _bool | _str | Sequence[Hashable] | None = ...,
-        mode: _str = ...,
-        encoding: _str | None = ...,
-        compression: _str | Mapping[_str, _str] = ...,
-        quoting: int | None = ...,
-        quotechar: _str = ...,
-        line_terminator: _str | None = ...,
-        chunksize: int | None = ...,
-        date_format: _str | None = ...,
-        doublequote: _bool = ...,
-        escapechar: _str | None = ...,
-        decimal: _str = ...,
-        errors: _str = ...,
-        storage_options: dict[_str, Any] | None = ...,
-    ) -> _str: ...
-    @overload
     def to_json(
         self,
         path_or_buf: FilePathOrBuffer | None,

pandas-stubs/core/generic.pyi

@@ -33,6 +33,7 @@ from pandas._typing import (
     Scalar,
     SeriesAxisType,
     SortKind,
+    StorageOptions,
     T,
 )
 
@@ -204,17 +205,19 @@ class NDFrame(PandasObject, indexing.IndexingMixin):
     @overload
     def to_csv(
         self,
-        path_or_buf: FilePathOrBuffer | None,
+        path_or_buf: FilePathOrBuffer,
         sep: _str = ...,
         na_rep: _str = ...,
-        float_format: _str | None = ...,
-        columns: Sequence[Hashable] | None = ...,
+        float_format: _str | Callable | None = ...,
+        columns: list[HashableT] | None = ...,
         header: _bool | list[_str] = ...,
         index: _bool = ...,
-        index_label: _bool | _str | Sequence[Hashable] | None = ...,
-        mode: _str = ...,
+        index_label: Literal[False] | _str | list[HashableT] | None = ...,
+        mode: Literal[
+            "a", "w", "x", "at", "wt", "xt", "ab", "wb", "xb", "w+", "w+b", "a+", "a+b"
+        ] = ...,
         encoding: _str | None = ...,
-        compression: _str | Mapping[_str, _str] = ...,
+        compression: CompressionOptions = ...,
         quoting: int | None = ...,
         quotechar: _str = ...,
         line_terminator: _str | None = ...,
@@ -224,21 +227,24 @@ class NDFrame(PandasObject, indexing.IndexingMixin):
         escapechar: _str | None = ...,
         decimal: _str = ...,
         errors: _str = ...,
-        storage_options: dict[_str, Any] | None = ...,
+        storage_options: StorageOptions = ...,
     ) -> None: ...
     @overload
     def to_csv(
         self,
+        path_or_buf: None = ...,
         sep: _str = ...,
         na_rep: _str = ...,
-        float_format: _str | None = ...,
-        columns: Sequence[Hashable] | None = ...,
+        float_format: _str | Callable | None = ...,
+        columns: list[HashableT] | None = ...,
         header: _bool | list[_str] = ...,
         index: _bool = ...,
-        index_label: _bool | _str | Sequence[Hashable] | None = ...,
-        mode: _str = ...,
+        index_label: Literal[False] | _str | list[HashableT] | None = ...,
+        mode: Literal[
+            "a", "w", "x", "at", "wt", "xt", "ab", "wb", "xb", "w+", "w+b", "a+", "a+b"
+        ] = ...,
         encoding: _str | None = ...,
-        compression: _str | Mapping[_str, _str] = ...,
+        compression: CompressionOptions = ...,
         quoting: int | None = ...,
         quotechar: _str = ...,
         line_terminator: _str | None = ...,
@@ -248,7 +254,7 @@ class NDFrame(PandasObject, indexing.IndexingMixin):
         escapechar: _str | None = ...,
         decimal: _str = ...,
         errors: _str = ...,
-        storage_options: dict[_str, Any] | None = ...,
+        storage_options: StorageOptions = ...,
     ) -> _str: ...
     def take(
         self, indices, axis=..., is_copy: _bool | None = ..., **kwargs