mirror of
https://github.com/kuhyx/utils.git
synced 2026-11-10 18:47:20 +01:00
Krzysztof kuhy Rudnicki
a0fb279b1b
New agent_ready/ package (dist and live-url modes, 100% branch coverage) checks robots, sitemap, llms.txt, ai-catalog, 404, no-JS text, markdown mirrors and JSON-LD. repo_contract gains web-repo detection and installs the agent-ready pre-commit hook, CI workflow and check_agent_ready.sh shim; scripts/check_agent_ready.sh is the shared entry point. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NBScELfMUiieK9ManFbS3Y
122 lines
4.1 KiB
Python
122 lines
4.1 KiB
Python
"""One-pass HTML scan: visible body text, links, JSON-LD and code blocks."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
import re
|
|
from dataclasses import dataclass, field
|
|
from html.parser import HTMLParser
|
|
|
|
_HIDDEN = {"script", "style", "noscript", "template", "svg", "head", "title"}
|
|
_LANG = re.compile(r"\b(language|lang)-\S+")
|
|
|
|
|
|
@dataclass
|
|
class PageScan:
|
|
"""What an agent that does not run JavaScript sees in a page."""
|
|
|
|
text: str = ""
|
|
links: list[tuple[set[str], str, str]] = field(default_factory=list)
|
|
anchors: list[str] = field(default_factory=list)
|
|
scripts: list[str] = field(default_factory=list)
|
|
jsonld_ok: int = 0
|
|
jsonld_bad: int = 0
|
|
code_blocks: int = 0
|
|
untagged_code: int = 0
|
|
|
|
def link_href(self, rel: str, mime: str | None = None) -> str | None:
|
|
"""First ``<link>`` href with ``rel`` (and ``type`` if given)."""
|
|
for rels, typ, href in self.links:
|
|
if rel in rels and (mime is None or typ == mime):
|
|
return href
|
|
return None
|
|
|
|
|
|
def _tagged(attrs: dict[str, str]) -> bool:
|
|
"""True when an element carries a code-language marker."""
|
|
return bool(_LANG.search(attrs.get("class", ""))) or bool(
|
|
attrs.get("data-language") or attrs.get("data-lang")
|
|
)
|
|
|
|
|
|
class _Scanner(HTMLParser):
|
|
def __init__(self) -> None:
|
|
super().__init__(convert_charrefs=True)
|
|
self.out = PageScan()
|
|
self._hidden: list[str] = []
|
|
self._text: list[str] = []
|
|
self._pre: list[list[bool]] = [] # [pre_tagged, code_seen]
|
|
self._ld: list[str] | None = None
|
|
|
|
def handle_starttag(
|
|
self, tag: str, attr_list: list[tuple[str, str | None]]
|
|
) -> None:
|
|
attrs = {k: v or "" for k, v in attr_list}
|
|
if tag == "body": # </head> is optional in HTML; <body> always ends it
|
|
self._hidden = [h for h in self._hidden if h not in ("head", "title")]
|
|
if tag == "link" and attrs.get("href"):
|
|
rels = set(attrs.get("rel", "").lower().split())
|
|
self.out.links.append((rels, attrs.get("type", "").lower(), attrs["href"]))
|
|
elif tag == "a" and attrs.get("href"):
|
|
self.out.anchors.append(attrs["href"])
|
|
elif tag == "script":
|
|
if attrs.get("src"):
|
|
self.out.scripts.append(attrs["src"])
|
|
if attrs.get("type", "").lower() == "application/ld+json":
|
|
self._ld = []
|
|
elif tag == "pre":
|
|
self._pre.append([_tagged(attrs), False])
|
|
elif tag == "code" and self._pre and not self._pre[-1][1]:
|
|
self._pre[-1][1] = True
|
|
self.out.code_blocks += 1
|
|
self.out.untagged_code += not (self._pre[-1][0] or _tagged(attrs))
|
|
if tag in _HIDDEN:
|
|
self._hidden.append(tag)
|
|
|
|
def handle_endtag(self, tag: str) -> None:
|
|
if tag == "script" and self._ld is not None:
|
|
self._finish_ld("".join(self._ld))
|
|
self._ld = None
|
|
if tag == "pre" and self._pre:
|
|
self._pre.pop()
|
|
if tag in _HIDDEN and tag in self._hidden:
|
|
while self._hidden and self._hidden.pop() != tag:
|
|
pass
|
|
|
|
def handle_data(self, data: str) -> None:
|
|
if self._ld is not None:
|
|
self._ld.append(data)
|
|
elif not self._hidden:
|
|
self._text.append(data)
|
|
|
|
def _finish_ld(self, raw: str) -> None:
|
|
try:
|
|
doc = json.loads(raw)
|
|
except ValueError:
|
|
self.out.jsonld_bad += 1
|
|
return
|
|
nodes = doc if isinstance(doc, list) else [doc]
|
|
ok = (
|
|
all(
|
|
isinstance(n, dict) and ("@context" in n or "@graph" in n)
|
|
for n in nodes
|
|
)
|
|
and nodes
|
|
)
|
|
if ok:
|
|
self.out.jsonld_ok += 1
|
|
else:
|
|
self.out.jsonld_bad += 1
|
|
|
|
def result(self) -> PageScan:
|
|
self.out.text = re.sub(r"\s+", " ", " ".join(self._text)).strip()
|
|
return self.out
|
|
|
|
|
|
def scan(html: str) -> PageScan:
|
|
"""Scan an HTML document."""
|
|
parser = _Scanner()
|
|
parser.feed(html)
|
|
parser.close()
|
|
return parser.result()
|