Drop the doctype check (#10906)

Co-authored-by: Pradyun Gedam <[email protected]>

Author: q0w

news/10903.removal.rst

@@ -0,0 +1 @@
+Drop the doctype check, that presented a warning for index pages that use non-compliant HTML 5.

src/pip/_internal/exceptions.py

@@ -181,30 +181,6 @@ class UninstallationError(PipError):
     """General exception during uninstallation"""
 
 
-class BadHTMLDoctypeDeclaration(DiagnosticPipError):
-    reference = "bad-index-doctype"
-
-    def __init__(self, *, url: str) -> None:
-        super().__init__(
-            kind="warning",
-            message=(
-                "The package index page being used does not have a proper HTML "
-                "doctype declaration."
-            ),
-            context=f"Problematic URL: {escape(url)}",
-            note_stmt="This is an issue with the page at the URL mentioned above.",
-            hint_stmt=(
-                "You might need to reach out to the owner of that package index, "
-                "to get this fixed. "
-                "See https://github.com/pypa/pip/issues/10825 for context."
-            ),
-        )
-
-
-class MissingHTMLDoctypeDeclaration(BadHTMLDoctypeDeclaration):
-    reference = "missing-index-doctype"
-
-
 class MissingPyProjectBuildRequires(DiagnosticPipError):
     """Raised when pyproject.toml has `build-system`, but no `build-system.requires`."""
 

src/pip/_internal/index/collector.py

@@ -32,11 +32,7 @@
 from pip._vendor.requests import Response
 from pip._vendor.requests.exceptions import RetryError, SSLError
 
-from pip._internal.exceptions import (
-    BadHTMLDoctypeDeclaration,
-    MissingHTMLDoctypeDeclaration,
-    NetworkConnectionError,
-)
+from pip._internal.exceptions import NetworkConnectionError
 from pip._internal.models.link import Link
 from pip._internal.models.search_scope import SearchScope
 from pip._internal.network.session import PipSession
@@ -401,33 +397,12 @@ class HTMLLinkParser(HTMLParser):
 
     def __init__(self, url: str) -> None:
         super().__init__(convert_charrefs=True)
-        self._dealt_with_doctype_issues = False
 
         self.url: str = url
         self.base_url: Optional[str] = None
         self.anchors: List[Dict[str, Optional[str]]] = []
 
-    def handle_decl(self, decl: str) -> None:
-        self._dealt_with_doctype_issues = True
-        match = re.match(
-            r"""doctype\s+html\s*(?:SYSTEM\s+(["'])about:legacy-compat\1)?\s*$""",
-            decl,
-            re.IGNORECASE,
-        )
-        if match is None:
-            logger.warning(
-                "[present-diagnostic] %s",
-                BadHTMLDoctypeDeclaration(url=self.url),
-            )
-
     def handle_starttag(self, tag: str, attrs: List[Tuple[str, Optional[str]]]) -> None:
-        if not self._dealt_with_doctype_issues:
-            logger.warning(
-                "[present-diagnostic] %s",
-                MissingHTMLDoctypeDeclaration(url=self.url),
-            )
-            self._dealt_with_doctype_issues = True
-
         if tag == "base" and self.base_url is None:
             href = self.get_href(attrs)
             if href is not None:

tests/functional/test_install_index.py

@@ -1,4 +1,5 @@
 import os
+import shutil
 import textwrap
 import urllib.parse
 
@@ -24,6 +25,21 @@ def test_find_links_relative_path(script: PipTestEnvironment, data: TestData) ->
     result.did_create(initools_folder)
 
 
+def test_find_links_no_doctype(script: PipTestEnvironment, data: TestData) -> None:
+    shutil.copy(data.packages / "simple-1.0.tar.gz", script.scratch_path)
+    html = script.scratch_path.joinpath("index.html")
+    html.write_text('<a href="simple-1.0.tar.gz"></a>')
+    result = script.pip(
+        "install",
+        "simple==1.0",
+        "--no-index",
+        "--find-links",
+        script.scratch_path,
+        expect_stderr=True,
+    )
+    assert not result.stderr
+
+
 @pytest.mark.usefixtures("with_wheel")
 def test_find_links_requirements_file_relative_path(
     script: PipTestEnvironment, data: TestData

tests/unit/test_collector.py

@@ -551,44 +551,6 @@ def test_parse_link_handles_deprecated_usage_properly() -> None:
     assert "pkg1-2.0" in parsed_links[1].url
 
 
-def test_parse_links_presents_warning_on_missing_doctype(
-    caplog: pytest.LogCaptureFixture,
-) -> None:
-    html = b'<a href="/pkg1-1.0.tar.gz"></a><a href="/pkg1-2.0.tar.gz"></a>'
-    url = "https://example.com/simple/"
-    page = HTMLPage(html, encoding=None, url=url, cache_link_parsing=False)
-
-    with caplog.at_level(logging.WARN):
-        parsed_links = list(parse_links(page, use_deprecated_html5lib=False))
-
-    assert len(parsed_links) == 2, parsed_links
-    assert "pkg1-1.0" in parsed_links[0].url
-    assert "pkg1-2.0" in parsed_links[1].url
-
-    assert len(caplog.records) == 1
-
-
-def test_parse_links_presents_warning_on_html4_doctype(
-    caplog: pytest.LogCaptureFixture,
-) -> None:
-    html = (
-        b'<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" '
-        b'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
-        b'<a href="/pkg1-1.0.tar.gz"></a><a href="/pkg1-2.0.tar.gz"></a>'
-    )
-    url = "https://example.com/simple/"
-    page = HTMLPage(html, encoding=None, url=url, cache_link_parsing=False)
-
-    with caplog.at_level(logging.WARN):
-        parsed_links = list(parse_links(page, use_deprecated_html5lib=False))
-
-    assert len(parsed_links) == 2, parsed_links
-    assert "pkg1-1.0" in parsed_links[0].url
-    assert "pkg1-2.0" in parsed_links[1].url
-
-    assert len(caplog.records) == 1
-
-
 @mock.patch("pip._internal.index.collector.raise_for_status")
 def test_request_http_error(
     mock_raise_for_status: mock.Mock, caplog: pytest.LogCaptureFixture