html5lib: Add various types (#11429)

I started out investigating comments in #11411 and ended up adding a few other
types that were reasonably obvious from the source code. For reference:
https://github.com/html5lib/html5lib-python/tree/master/html5lib

stubs/html5lib/html5lib/_inputstream.pyi

@@ -1,5 +1,10 @@
-from _typeshed import Incomplete
-from typing import Any
+from _typeshed import Incomplete, SupportsRead
+from typing import Any, overload
+from typing_extensions import TypeAlias
+
+_UnicodeInputStream: TypeAlias = str | SupportsRead[str]
+_BinaryInputStream: TypeAlias = bytes | SupportsRead[bytes]
+_InputStream: TypeAlias = _UnicodeInputStream  # noqa: Y047  # used in other files
 
 spaceCharactersBytes: Any
 asciiLettersBytes: Any
@@ -20,14 +25,26 @@ class BufferedStream:
     def seek(self, pos) -> None: ...
     def read(self, bytes): ...
 
-def HTMLInputStream(source, **kwargs): ...
+@overload
+def HTMLInputStream(source: _UnicodeInputStream) -> HTMLUnicodeInputStream: ...
+@overload
+def HTMLInputStream(
+    source: _BinaryInputStream,
+    *,
+    override_encoding: str | bytes | None = None,
+    transport_encoding: str | bytes | None = None,
+    same_origin_parent_encoding: str | bytes | None = None,
+    likely_encoding: str | bytes | None = None,
+    default_encoding: str = "windows-1252",
+    useChardet: bool = True,
+) -> HTMLBinaryInputStream: ...
 
 class HTMLUnicodeInputStream:
     reportCharacterErrors: Any
     newLines: Any
     charEncoding: Any
     dataStream: Any
-    def __init__(self, source) -> None: ...
+    def __init__(self, source: _UnicodeInputStream) -> None: ...
     chunk: str
     chunkSize: int
     chunkOffset: int
@@ -56,11 +73,11 @@ class HTMLBinaryInputStream(HTMLUnicodeInputStream):
     charEncoding: Any
     def __init__(
         self,
-        source,
-        override_encoding: Incomplete | None = None,
-        transport_encoding: Incomplete | None = None,
-        same_origin_parent_encoding: Incomplete | None = None,
-        likely_encoding: Incomplete | None = None,
+        source: _BinaryInputStream,
+        override_encoding: str | bytes | None = None,
+        transport_encoding: str | bytes | None = None,
+        same_origin_parent_encoding: str | bytes | None = None,
+        likely_encoding: str | bytes | None = None,
         default_encoding: str = "windows-1252",
         useChardet: bool = True,
     ) -> None: ...
@@ -108,4 +125,4 @@ class ContentAttrParser:
     def __init__(self, data) -> None: ...
     def parse(self): ...
 
-def lookupEncoding(encoding): ...
+def lookupEncoding(encoding: str | bytes | None) -> str | None: ...

stubs/html5lib/html5lib/_tokenizer.pyi

@@ -1,6 +1,8 @@
 from _typeshed import Incomplete
 from typing import Any
 
+from ._inputstream import _InputStream
+
 entitiesTrie: Any
 attributeMap = dict
 
@@ -12,7 +14,7 @@ class HTMLTokenizer:
     state: Any
     escape: bool
     currentToken: Any
-    def __init__(self, stream, parser: Incomplete | None = None, **kwargs) -> None: ...
+    def __init__(self, stream: _InputStream, parser: Incomplete | None = None, **kwargs) -> None: ...
     tokenQueue: Any
     def __iter__(self): ...
     def consumeNumberEntity(self, isHex): ...
@@ -36,23 +38,23 @@ class HTMLTokenizer:
     def rawtextLessThanSignState(self): ...
     def rawtextEndTagOpenState(self): ...
     def rawtextEndTagNameState(self): ...
-    def scriptDataLessThanSignState(self): ...
-    def scriptDataEndTagOpenState(self): ...
-    def scriptDataEndTagNameState(self): ...
-    def scriptDataEscapeStartState(self): ...
-    def scriptDataEscapeStartDashState(self): ...
-    def scriptDataEscapedState(self): ...
-    def scriptDataEscapedDashState(self): ...
-    def scriptDataEscapedDashDashState(self): ...
-    def scriptDataEscapedLessThanSignState(self): ...
-    def scriptDataEscapedEndTagOpenState(self): ...
-    def scriptDataEscapedEndTagNameState(self): ...
-    def scriptDataDoubleEscapeStartState(self): ...
-    def scriptDataDoubleEscapedState(self): ...
-    def scriptDataDoubleEscapedDashState(self): ...
-    def scriptDataDoubleEscapedDashDashState(self): ...
-    def scriptDataDoubleEscapedLessThanSignState(self): ...
-    def scriptDataDoubleEscapeEndState(self): ...
+    def scriptDataLessThanSignState(self) -> bool: ...
+    def scriptDataEndTagOpenState(self) -> bool: ...
+    def scriptDataEndTagNameState(self) -> bool: ...
+    def scriptDataEscapeStartState(self) -> bool: ...
+    def scriptDataEscapeStartDashState(self) -> bool: ...
+    def scriptDataEscapedState(self) -> bool: ...
+    def scriptDataEscapedDashState(self) -> bool: ...
+    def scriptDataEscapedDashDashState(self) -> bool: ...
+    def scriptDataEscapedLessThanSignState(self) -> bool: ...
+    def scriptDataEscapedEndTagOpenState(self) -> bool: ...
+    def scriptDataEscapedEndTagNameState(self) -> bool: ...
+    def scriptDataDoubleEscapeStartState(self) -> bool: ...
+    def scriptDataDoubleEscapedState(self) -> bool: ...
+    def scriptDataDoubleEscapedDashState(self) -> bool: ...
+    def scriptDataDoubleEscapedDashDashState(self) -> bool: ...
+    def scriptDataDoubleEscapedLessThanSignState(self) -> bool: ...
+    def scriptDataDoubleEscapeEndState(self) -> bool: ...
     def beforeAttributeNameState(self): ...
     def attributeNameState(self): ...
     def afterAttributeNameState(self): ...
@@ -64,17 +66,17 @@ class HTMLTokenizer:
     def selfClosingStartTagState(self): ...
     def bogusCommentState(self): ...
     def markupDeclarationOpenState(self): ...
-    def commentStartState(self): ...
-    def commentStartDashState(self): ...
-    def commentState(self): ...
-    def commentEndDashState(self): ...
-    def commentEndState(self): ...
-    def commentEndBangState(self): ...
-    def doctypeState(self): ...
-    def beforeDoctypeNameState(self): ...
-    def doctypeNameState(self): ...
-    def afterDoctypeNameState(self): ...
-    def afterDoctypePublicKeywordState(self): ...
+    def commentStartState(self) -> bool: ...
+    def commentStartDashState(self) -> bool: ...
+    def commentState(self) -> bool: ...
+    def commentEndDashState(self) -> bool: ...
+    def commentEndState(self) -> bool: ...
+    def commentEndBangState(self) -> bool: ...
+    def doctypeState(self) -> bool: ...
+    def beforeDoctypeNameState(self) -> bool: ...
+    def doctypeNameState(self) -> bool: ...
+    def afterDoctypeNameState(self) -> bool: ...
+    def afterDoctypePublicKeywordState(self) -> bool: ...
     def beforeDoctypePublicIdentifierState(self): ...
     def doctypePublicIdentifierDoubleQuotedState(self): ...
     def doctypePublicIdentifierSingleQuotedState(self): ...

stubs/html5lib/html5lib/html5parser.pyi

@@ -1,25 +1,25 @@
-from _typeshed import Incomplete, SupportsRead
+from _typeshed import Incomplete
 from typing import Any, Literal, overload
 from xml.etree.ElementTree import Element
 
+from ._inputstream import _InputStream
+from ._tokenizer import HTMLTokenizer
+
 @overload
 def parse(
-    doc: str | bytes | SupportsRead[str] | SupportsRead[bytes],
-    treebuilder: Literal["etree"] = "etree",
-    namespaceHTMLElements: bool = True,
-    **kwargs,
+    doc: _InputStream, treebuilder: Literal["etree"] = "etree", namespaceHTMLElements: bool = True, **kwargs
 ) -> Element: ...
 @overload
-def parse(
-    doc: str | bytes | SupportsRead[str] | SupportsRead[bytes], treebuilder: str, namespaceHTMLElements: bool = True, **kwargs
+def parse(doc: _InputStream, treebuilder: str, namespaceHTMLElements: bool = True, **kwargs): ...
+def parseFragment(
+    doc: _InputStream, container: str = "div", treebuilder: str = "etree", namespaceHTMLElements: bool = True, **kwargs
 ): ...
-def parseFragment(doc, container: str = "div", treebuilder: str = "etree", namespaceHTMLElements: bool = True, **kwargs): ...
 def method_decorator_metaclass(function): ...
 
 class HTMLParser:
-    strict: Any
+    strict: bool
     tree: Any
-    errors: Any
+    errors: list[Incomplete]
     phases: Any
     def __init__(
         self, tree: Incomplete | None = None, strict: bool = False, namespaceHTMLElements: bool = True, debug: bool = False
@@ -27,20 +27,21 @@ class HTMLParser:
     firstStartTag: bool
     log: Any
     compatMode: str
+    container: str
     innerHTML: Any
     phase: Any
     lastPhase: Any
     beforeRCDataPhase: Any
     framesetOK: bool
-    tokenizer: Any
+    tokenizer: HTMLTokenizer
     def reset(self) -> None: ...
     @property
     def documentEncoding(self) -> str | None: ...
-    def isHTMLIntegrationPoint(self, element) -> bool: ...
-    def isMathMLTextIntegrationPoint(self, element) -> bool: ...
+    def isHTMLIntegrationPoint(self, element: Element) -> bool: ...
+    def isMathMLTextIntegrationPoint(self, element: Element) -> bool: ...
     def mainLoop(self) -> None: ...
-    def parse(self, stream, scripting: bool = ..., **kwargs): ...
-    def parseFragment(self, stream, *args, **kwargs): ...
+    def parse(self, stream: _InputStream, scripting: bool = ..., **kwargs): ...
+    def parseFragment(self, stream: _InputStream, *args, **kwargs): ...
     def parseError(self, errorcode: str = "XXX-undefined-error", datavars: Incomplete | None = None) -> None: ...
     def adjustMathMLAttributes(self, token) -> None: ...
     def adjustSVGAttributes(self, token) -> None: ...