Mark bleach as complete

Author: sobolevn

pyrightconfig.stricter.json

@@ -21,7 +21,6 @@
         "stdlib/xml/sax",
         "stubs/aws-xray-sdk",
         "stubs/babel",
-        "stubs/bleach",
         "stubs/boto",
         "stubs/beautifulsoup4",
         "stubs/braintree",

stubs/bleach/@tests/stubtest_allowlist.txt

@@ -1,2 +1,13 @@
-bleach.css_sanitizer  # Requires tinycss2 to be installed
-bleach.html5lib_shim.*
+# Internal private stuff:
+bleach._vendor.*
+
+# Hacks:
+bleach.html5lib_shim.InputStreamWithMemory.changeEncoding
+bleach.html5lib_shim.InputStreamWithMemory.reset
+
+# Re-exports:
+bleach.html5lib_shim.allowed_svg_properties
+bleach.html5lib_shim.allowed_protocols
+bleach.html5lib_shim.allowed_css_properties
+bleach.html5lib_shim.namespaces
+bleach.html5lib_shim.prefixes

stubs/bleach/METADATA.toml

@@ -1 +1,6 @@
 version = "5.0.*"
+requires = ["types-html5lib"]
+
+[tool.stubtest]
+ignore_missing_stub = false
+extras = ["css"]

stubs/bleach/bleach/html5lib_shim.pyi

@@ -1,17 +1,46 @@
-from _typeshed import Incomplete
-from collections.abc import Generator, Iterable
+from collections.abc import Iterator, Iterable
+from typing import Any
+import re
 
-class HTMLParser:  # actually html5lib.HTMLParser
-    def __getattr__(self, __name: str) -> Incomplete: ...
+# We don't re-export any `html5lib` types / values here, because they are not
+# really public and may change at any time. This is just a helper module,
+# import things directly from `html5lib` instead!
+from html5lib import HTMLParser
+from html5lib.serializer import HTMLSerializer
+from html5lib._tokenizer import HTMLTokenizer
+from html5lib._inputstream import HTMLUnicodeInputStream
 
-class Filter:  # actually html5lib.filters.base.Filter
-    def __getattr__(self, __name: str) -> Incomplete: ...
+HTML_TAGS: list[str]
+HTML_TAGS_BLOCK_LEVEL: frozenset[str]
+AMP_SPLIT_RE: re.Pattern[str]
+ENTITIES: dict[str, str]
+TAG_TOKEN_TYPES: set[int]
+TAG_TOKEN_TYPE_CHARACTERS: int
+TAG_TOKEN_TYPE_END: int
+TAG_TOKEN_TYPE_PARSEERROR: int
+TAG_TOKEN_TYPE_START: int
 
-class SanitizerFilter:  # actually html5lib.filters.sanitizer.Filter
-    def __getattr__(self, __name: str) -> Incomplete: ...
+class InputStreamWithMemory:
+    position: int
+    def __init__(self, inner_stream: HTMLUnicodeInputStream) -> None: ...
+    def reset(self) -> None: ...
+    @property
+    def errors(self) -> list[str]: ...
+    @property
+    def charEncoding(self) -> tuple[str, str]: ...
+    # Is a property returning a method, simplified:
+    def changeEncoding(self, newEncoding: str) -> None: ...
+    def char(self) -> str: ...
+    def charsUntil(self, characters: str, opposite: bool = ...) -> str: ...
+    def unget(self, char: str | None) -> None: ...
+    def get_tag(self) -> str: ...
+    def start_tag(self) -> None: ...
 
-class HTMLSerializer:  # actually html5lib.serializer.HTMLSerializer
-    def __getattr__(self, __name: str) -> Incomplete: ...
+class BleachHTMLTokenizer(HTMLTokenizer):
+    consume_entities: bool
+    stream: InputStreamWithMemory
+    emitted_last_token: dict[str, Any] | None
+    def __init__(self, consume_entities: bool = ..., **kwargs): ...
 
 class BleachHTMLParser(HTMLParser):
     tags: list[str] | None
@@ -21,7 +50,10 @@ class BleachHTMLParser(HTMLParser):
 
 class BleachHTMLSerializer(HTMLSerializer):
     escape_rcdata: bool
-    def escape_base_amp(self, stoken: str) -> Generator[str, None, None]: ...
-    def serialize(self, treewalker, encoding: str | None = ...) -> Generator[str, None, None]: ...
+    def escape_base_amp(self, stoken: str) -> Iterator[str]: ...
+    def serialize(self, treewalker, encoding: str | None = ...) -> Iterator[str]: ...  # type: ignore[override]
 
-def __getattr__(__name: str) -> Incomplete: ...
+def convert_entity(value: str) -> str | None: ...
+def convert_entities(text: str) -> str: ...
+def match_entity(stream: str) -> str | None: ...
+def next_possible_entity(text: str) -> Iterator[str]: ...

stubs/bleach/bleach/linkifier.pyi

@@ -1,12 +1,12 @@
-from _typeshed import Incomplete
-from collections.abc import Container, Iterable, MutableMapping
+import re
+from collections.abc import Container, Iterable, MutableMapping, Sequence, Iterator
 from re import Pattern
 from typing import Any, Protocol
 from typing_extensions import TypeAlias
 
-from .html5lib_shim import Filter
+from html5lib.filters.base import Filter
 
-_Attrs: TypeAlias = MutableMapping[Any, str]
+from .callbacks import _Attrs
 
 class _Callback(Protocol):
     def __call__(self, attrs: _Attrs, new: bool = ...) -> _Attrs: ...
@@ -36,13 +36,22 @@ class Linker:
     ) -> None: ...
     def linkify(self, text: str) -> str: ...
 
+# TODO: `_Token` might be converted into `TypedDict`
+# or `html5lib` token might be reused
+_Token: TypeAlias = dict[str, Any]
+
 class LinkifyFilter(Filter):
-    callbacks: Any
+    callbacks: list[_Callback]
     skip_tags: Container[str]
     parse_email: bool
-    url_re: Any
-    email_re: Any
+    url_re: re.Pattern[str]
+    email_re: re.Pattern[str]
     def __init__(
-        self, source, callbacks=..., skip_tags: Container[str] | None = ..., parse_email: bool = ..., url_re=..., email_re=...
+        self, source, callbacks=..., skip_tags: Container[str] | None = ..., parse_email: bool = ..., url_re: re.Pattern[str] = ..., email_re: re.Pattern[str] = ...
     ) -> None: ...
-    def __getattr__(self, item: str) -> Incomplete: ...
+    def apply_callbacks(self, attrs: _Attrs, is_new: bool) -> _Attrs: ...
+    def extract_character_data(self, token_list: _Token) -> str: ...
+    def handle_a_tag(self, token_buffer: Sequence[_Token]) -> Iterator[_Token]: ...
+    def handle_email_addresses(self, src_iter: _Token) -> Iterator[_Token]: ...
+    def handle_links(self, src_iter: Iterable[_Token]) -> Iterator[_Token]: ...
+    def strip_non_url_bits(self, fragment: str) -> tuple[str, str, str]: ...

stubs/bleach/bleach/parse_shim.pyi

@@ -0,0 +1 @@
+from urllib import parse as parse

stubs/bleach/bleach/sanitizer.pyi

@@ -1,10 +1,13 @@
-from collections.abc import Callable, Container, Iterable
+from collections.abc import Callable, Container, Iterable, Iterator
 from re import Pattern
 from typing import Any
 from typing_extensions import TypeAlias
 
 from .css_sanitizer import CSSSanitizer
-from .html5lib_shim import BleachHTMLParser, BleachHTMLSerializer, SanitizerFilter
+from .html5lib_shim import BleachHTMLParser, BleachHTMLSerializer
+from .linkifier import _Token
+from html5lib.filters.sanitizer import Filter as SanitizerFilter
+from html5lib.filters.base import Filter
 
 ALLOWED_TAGS: list[str]
 ALLOWED_ATTRIBUTES: dict[str, list[str]]
@@ -14,16 +17,13 @@ INVISIBLE_CHARACTERS: str
 INVISIBLE_CHARACTERS_RE: Pattern[str]
 INVISIBLE_REPLACEMENT_CHAR: str
 
-# A html5lib Filter class
-_Filter: TypeAlias = Any
-
 class Cleaner:
     tags: Container[str]
     attributes: _Attributes
     protocols: Container[str]
     strip: bool
     strip_comments: bool
-    filters: Iterable[_Filter]
+    filters: Iterable[Filter]
     css_sanitizer: CSSSanitizer | None
     parser: BleachHTMLParser
     walker: Any
@@ -35,7 +35,7 @@ class Cleaner:
         protocols: Container[str] = ...,
         strip: bool = ...,
         strip_comments: bool = ...,
-        filters: Iterable[_Filter] | None = ...,
+        filters: Iterable[Filter] | None = ...,
         css_sanitizer: CSSSanitizer | None = ...,
     ) -> None: ...
     def clean(self, text: str) -> str: ...
@@ -61,12 +61,7 @@ class BleachSanitizerFilter(SanitizerFilter):
         css_sanitizer: CSSSanitizer | None = ...,
         **kwargs,
     ) -> None: ...
-    def sanitize_stream(self, token_iterator): ...
-    def merge_characters(self, token_iterator): ...
-    def __iter__(self): ...
-    def sanitize_token(self, token): ...
-    def sanitize_characters(self, token): ...
-    def sanitize_uri_value(self, value, allowed_protocols): ...
-    def allow_token(self, token): ...
-    def disallowed_token(self, token): ...
-    def sanitize_css(self, style): ...
+    def sanitize_stream(self, token_iterator: Iterable[_Token]) -> Iterator[_Token]: ...
+    def merge_characters(self, token_iterator: Iterable[_Token]) -> Iterator[_Token]: ...
+    def sanitize_characters(self, token: _Token) -> _Token | None: ...
+    def sanitize_uri_value(self, value: str, allowed_protocols: Container[str]) -> str | None: ...

stubs/html5lib/html5lib/_inputstream.pyi

@@ -24,13 +24,13 @@ def HTMLInputStream(source, **kwargs): ...
 class HTMLUnicodeInputStream:
     reportCharacterErrors: Any
     newLines: Any
-    charEncoding: Any
+    charEncoding: tuple[str, str]
     dataStream: Any
     def __init__(self, source) -> None: ...
     chunk: str
     chunkSize: int
     chunkOffset: int
-    errors: Any
+    errors: list[str]
     prevNumLines: int
     prevNumCols: int
     def reset(self) -> None: ...