Update bleach stubs (#5631)

pyrightconfig.stricter.json

@@ -26,6 +26,7 @@
         "stdlib/xml/sax",
         "stubs/backports",
         "stubs/backports_abc",
+        "stubs/bleach",
         "stubs/boto",
         "stubs/cryptography",
         "stubs/docutils",

stubs/bleach/@tests/stubtest_allowlist.txt

@@ -1,2 +1 @@
-bleach.linkifier.LinkifyFilter.__init__
-bleach.sanitizer.BleachSanitizerFilter.__init__
+bleach.html5lib_shim.*

stubs/bleach/METADATA.toml

@@ -1,2 +1,2 @@
-version = "0.1"
+version = "3.3"
 python2 = true

stubs/bleach/bleach/__init__.pyi

@@ -7,8 +7,11 @@ from bleach.sanitizer import (
     ALLOWED_STYLES as ALLOWED_STYLES,
     ALLOWED_TAGS as ALLOWED_TAGS,
     Cleaner as Cleaner,
+    _Attributes,
 )
 
+__all__ = ["clean", "linkify"]
+
 __releasedate__: Text
 __version__: Text
 VERSION: Any  # packaging.version.Version
@@ -16,7 +19,7 @@ VERSION: Any  # packaging.version.Version
 def clean(
     text: Text,
     tags: Container[Text] = ...,
-    attributes: Any = ...,
+    attributes: _Attributes = ...,
     styles: Container[Text] = ...,
     protocols: Container[Text] = ...,
     strip: bool = ...,

stubs/bleach/bleach/html5lib_shim.pyi

@@ -0,0 +1,26 @@
+from typing import Any, Generator, Iterable, List, Optional, Text
+
+class HTMLParser(object):  # actually html5lib.HTMLParser
+    def __getattr__(self, __name: Text) -> Any: ...  # incomplete
+
+class Filter(object):  # actually html5lib.filters.base.Filter
+    def __getattr__(self, __name: Text) -> Any: ...  # incomplete
+
+class SanitizerFilter(object):  # actually html5lib.filters.sanitizer.Filter
+    def __getattr__(self, __name: Text) -> Any: ...  # incomplete
+
+class HTMLSerializer(object):  # actually html5lib.serializer.HTMLSerializer
+    def __getattr__(self, __name: Text) -> Any: ...  # incomplete
+
+class BleachHTMLParser(HTMLParser):
+    tags: Optional[List[Text]]
+    strip: bool
+    consume_entities: bool
+    def __init__(self, tags: Optional[Iterable[Text]], strip: bool, consume_entities: bool, **kwargs) -> None: ...
+
+class BleachHTMLSerializer(HTMLSerializer):
+    escape_rcdata: bool
+    def escape_base_amp(self, stoken: Text) -> Generator[Text, None, None]: ...
+    def serialize(self, treewalker, encoding: Optional[Text] = ...) -> Generator[Text, None, None]: ...
+
+def __getattr__(__name: Text) -> Any: ...  # incomplete

stubs/bleach/bleach/linkifier.pyi

@@ -1,5 +1,7 @@
 from typing import Any, Container, Iterable, List, MutableMapping, Optional, Pattern, Protocol, Text
 
+from .html5lib_shim import Filter
+
 _Attrs = MutableMapping[Any, Text]
 
 class _Callback(Protocol):
@@ -13,6 +15,9 @@ def build_url_re(tlds: Iterable[Text] = ..., protocols: Iterable[Text] = ...) ->
 
 URL_RE: Pattern[Text]
 PROTO_RE: Pattern[Text]
+
+def build_email_re(tlds: Iterable[Text] = ...) -> Pattern[Text]: ...
+
 EMAIL_RE: Pattern[Text]
 
 class Linker(object):
@@ -27,5 +32,13 @@ class Linker(object):
     ) -> None: ...
     def linkify(self, text: Text) -> Text: ...
 
-class LinkifyFilter(object):  # TODO: derives from html5lib.Filter
-    def __getattr__(self, item: str) -> Any: ...  # incomplete
+class LinkifyFilter(Filter):
+    callbacks: Any
+    skip_tags: Container[Text]
+    parse_email: bool
+    url_re: Any
+    email_re: Any
+    def __init__(
+        self, source, callbacks=..., skip_tags: Container[Text] | None = ..., parse_email: bool = ..., url_re=..., email_re=...
+    ) -> None: ...
+    def __getattr__(self, item: Text) -> Any: ...  # incomplete

stubs/bleach/bleach/sanitizer.pyi

@@ -1,5 +1,7 @@
 from typing import Any, Callable, Container, Dict, Iterable, List, Optional, Pattern, Text, Union
 
+from .html5lib_shim import BleachHTMLParser, BleachHTMLSerializer, SanitizerFilter
+
 ALLOWED_TAGS: List[Text]
 ALLOWED_ATTRIBUTES: Dict[Text, List[Text]]
 ALLOWED_STYLES: List[Text]
@@ -13,10 +15,20 @@ INVISIBLE_REPLACEMENT_CHAR: Text
 _Filter = Any
 
 class Cleaner(object):
+    tags: Container[Text]
+    attributes: _Attributes
+    styles: Container[Text]
+    protocols: Container[Text]
+    strip: bool
+    strip_comments: bool
+    filters: Iterable[_Filter]
+    parser: BleachHTMLParser
+    walker: Any
+    serializer: BleachHTMLSerializer
     def __init__(
         self,
         tags: Container[Text] = ...,
-        attributes: Any = ...,
+        attributes: _Attributes = ...,
         styles: Container[Text] = ...,
         protocols: Container[Text] = ...,
         strip: bool = ...,
@@ -27,8 +39,28 @@ class Cleaner(object):
 
 _AttributeFilter = Callable[[Text, Text, Text], bool]
 _AttributeDict = Dict[Text, Union[Container[Text], _AttributeFilter]]
+_Attributes = Union[_AttributeFilter, _AttributeDict, List[Text]]
 
-def attribute_filter_factory(attributes: Union[_AttributeFilter, _AttributeDict, Container[Text]]) -> _AttributeFilter: ...
+def attribute_filter_factory(attributes: _Attributes) -> _AttributeFilter: ...
 
-class BleachSanitizerFilter(object):  # TODO: derives from html5lib.sanitizer.Filter
-    def __getattr__(self, item: str) -> Any: ...  # incomplete
+class BleachSanitizerFilter(SanitizerFilter):
+    attr_filter: _AttributeFilter
+    strip_disallowed_elements: bool
+    strip_html_comments: bool
+    def __init__(
+        self,
+        source,
+        attributes: _Attributes = ...,
+        strip_disallowed_elements: bool = ...,
+        strip_html_comments: bool = ...,
+        **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): ...