Normalize HTML poster wrappers

src/vk_parser_app/text_utils.py

@@ -1,6 +1,9 @@
 from __future__ import annotations
 
+import html
+from html.parser import HTMLParser
 import re
+from urllib.parse import urlparse
 
 
 def normalize_hash_tag(value: str, fallback: str = "source") -> str:
@@ -47,7 +50,91 @@ def publication_hashtags(category_tag: str, source_tag: str, common_tags: object
     return " ".join(f"#{tag}" for tag in tags if tag)
 
 
-import html
+def _unescape_repeated(value: str, limit: int = 3) -> str:
+    current = str(value or "")
+    for _ in range(limit):
+        next_value = html.unescape(current)
+        if next_value == current:
+            break
+        current = next_value
+    return current
+
+
+def _safe_href(value: str) -> str:
+    href = _unescape_repeated(value).strip()
+    parsed = urlparse(href)
+    if parsed.scheme not in {"http", "https"} or not parsed.netloc:
+        return ""
+    return href
+
+
+class _HTMLWrapperNormalizer(HTMLParser):
+    allowed_format_tags = {"b", "strong", "i", "em", "u", "s", "code"}
+
+    def __init__(self) -> None:
+        super().__init__(convert_charrefs=False)
+        self.parts: list[str] = []
+        self.open_tags: list[str] = []
+        self.skip_depth = 0
+
+    def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
+        tag = tag.lower()
+        if tag in {"script", "style"}:
+            self.skip_depth += 1
+            return
+        if self.skip_depth:
+            return
+        if tag == "br":
+            self.parts.append("\n")
+            return
+        if tag == "a":
+            href = _safe_href(next((value or "" for name, value in attrs if name.lower() == "href"), ""))
+            if not href:
+                return
+            self.parts.append(f'<a href="{html.escape(href, quote=True)}">')
+            self.open_tags.append("a")
+            return
+        if tag in self.allowed_format_tags:
+            self.parts.append(f"<{tag}>")
+            self.open_tags.append(tag)
+
+    def handle_endtag(self, tag: str) -> None:
+        tag = tag.lower()
+        if tag in {"script", "style"} and self.skip_depth:
+            self.skip_depth -= 1
+            return
+        if self.skip_depth or tag not in self.open_tags:
+            return
+        while self.open_tags:
+            opened = self.open_tags.pop()
+            self.parts.append(f"</{opened}>")
+            if opened == tag:
+                break
+
+    def handle_data(self, data: str) -> None:
+        if not self.skip_depth:
+            self.parts.append(html.escape(data))
+
+    def handle_entityref(self, name: str) -> None:
+        self.handle_data(html.unescape(f"&{name};"))
+
+    def handle_charref(self, name: str) -> None:
+        self.handle_data(html.unescape(f"&#{name};"))
+
+    def normalized(self) -> str:
+        while self.open_tags:
+            self.parts.append(f"</{self.open_tags.pop()}>")
+        return re.sub(r"\n{3,}", "\n\n", "".join(self.parts)).strip()
+
+
+def normalize_html_wrapper_text(value: str) -> str:
+    raw = str(value or "").strip()
+    if not raw:
+        return ""
+    parser = _HTMLWrapperNormalizer()
+    parser.feed(_unescape_repeated(raw))
+    parser.close()
+    return parser.normalized()
 
 
 def build_publication_text(
@@ -77,8 +164,12 @@ def build_publication_text(
             title_idx = idx
             break
 
-    header = str(header_text or "").strip()
-    footer = str(footer_text or "").strip()
+    if parse_mode == "html":
+        header = normalize_html_wrapper_text(header_text)
+        footer = normalize_html_wrapper_text(footer_text)
+    else:
+        header = str(header_text or "").strip()
+        footer = str(footer_text or "").strip()
     hashtags = publication_hashtags(category_tag, source_tag, common_tags)
     if title_idx is None:
         return "\n\n".join(part for part in [header, footer, hashtags] if part)