BigUpdate

md2gost/bibliography.py

@@ -0,0 +1,57 @@
+"""Bibliography parsing helpers (no docx dependency) and optional renderable."""
+
+from __future__ import annotations
+
+import re
+from dataclasses import dataclass
+
+
+BIBLIO_LINE_RE = re.compile(
+    r"^\[(\d+(?:\.\d+)?)\]:\s*(.+)$"
+)
+CITE_RE = re.compile(
+    r"\[(\d+(?:\.\d+)?(?:\s*,\s*\d+(?:\.\d+)?)*)"
+    r"(?:\s*,\s*[cс]\.\s*[\d\-–]+)?"
+    r"\]"
+)
+
+
+@dataclass
+class BiblioEntry:
+    key: str
+    text: str
+    year: int | None = None
+
+
+def extract_year(text: str) -> int | None:
+    years = re.findall(r"(?:^|[^\d])((?:19|20)\d{2})(?:[^\d]|$)", text)
+    if not years:
+        return None
+    return max(int(y) for y in years)
+
+
+def parse_biblio_block(lines: list[str]) -> list[BiblioEntry]:
+    entries = []
+    for line in lines:
+        m = BIBLIO_LINE_RE.match(line.strip())
+        if m:
+            key, text = m.group(1), m.group(2).strip()
+            entries.append(BiblioEntry(key=key, text=text, year=extract_year(text)))
+    return entries
+
+
+def find_citations(text: str) -> list[str]:
+    seen: list[str] = []
+    for m in CITE_RE.finditer(text):
+        inner = m.group(1)
+        for part in re.split(r"\s*,\s*", inner):
+            part = part.strip()
+            if part and part not in seen:
+                seen.append(part)
+    return seen
+
+
+def Bibliography(*args, **kwargs):
+    """Lazy wrapper to avoid importing docx/freetype at checker import time."""
+    from .bibliography_renderable import Bibliography as _Bibliography
+    return _Bibliography(*args, **kwargs)