1 files changed, 107 insertions, 0 deletions
diff --git a/wflib/refs.py b/wflib/refs.py
new file mode 100644
index 0000000..d1e853d
--- /dev/null
+++ b/wflib/refs.py
@@ -0,0 +1,107 @@
+"""Links, anchors and doc sections: what `[[id]]` and `Ref: path#anchor` point at."""
+from __future__ import annotations
+
+import re
+from dataclasses import dataclass
+from pathlib import Path
+
+LINK_RE = re.compile(r"\[\[([^\]|#]+)(?:#[^\]|]+)?(?:\|[^\]]+)?\]\]")
+HEADING_RE = re.compile(r"^(#{1,6})\s+(.*?)\s*$")
+ANCHOR_RE = re.compile(r'<a\s+id="([^"]+)"\s*>\s*</a>', re.I)
+FENCE_RE = re.compile(r"^\s*(```|~~~)")
+
+
+def strip_code(text: str) -> str:
+    text = re.sub(r"```.*?```", "", text, flags=re.S)
+    return re.sub(r"`[^`\n]*`", "", text)
+
+
+def links(text: str) -> list[str]:
+    return [m.group(1).strip() for m in LINK_RE.finditer(strip_code(text))]
+
+
+def slugify(heading: str) -> str:
+    text = ANCHOR_RE.sub("", heading).strip().lower()
+    text = re.sub(r"[^\w\s-]", "", text)
+    return re.sub(r"[\s_]+", "-", text).strip("-").replace("--", "-")
+
+
+@dataclass
+class DocSection:
+    level: int          # 0 = an anchor standing alone, not a heading
+    heading: str
+    anchors: set[str]
+    start: int          # 0-based line of the heading (or of the lone anchor)
+    end: int            # exclusive
+
+
+def _lines(text: str) -> list[str]:
+    lines = text.replace("\r\n", "\n").split("\n")
+    while lines and not lines[-1].strip():
+        lines.pop()
+    return lines
+
+
+def _is_lone_anchor(line: str) -> bool:
+    return bool(ANCHOR_RE.search(line)) and not ANCHOR_RE.sub("", line).strip()
+
+
+def sections(text: str) -> list[DocSection]:
+    lines = _lines(text)
+    out: list[DocSection] = []
+    fenced = False
+    for n, line in enumerate(lines):
+        if FENCE_RE.match(line):
+            fenced = not fenced
+        if fenced or FENCE_RE.match(line):
+            continue
+        m = HEADING_RE.match(line)
+        tags = set(ANCHOR_RE.findall(line))
+        if m:
+            heading = ANCHOR_RE.sub("", m.group(2)).strip()
+            above = set(ANCHOR_RE.findall(lines[n - 1])) if n and _is_lone_anchor(lines[n - 1]) else set()
+            out.append(DocSection(len(m.group(1)), heading, {slugify(heading)} | tags | above, n, len(lines)))
+        elif tags and not (_is_lone_anchor(line) and n + 1 < len(lines) and HEADING_RE.match(lines[n + 1])):
+            out.append(DocSection(0, "", tags, n, len(lines)))
+    headings = [s for s in out if s.level]
+    for s in out:
+        nxt = next((h for h in headings if h.start > s.start and (s.level == 0 or h.level <= s.level)), None)
+        if nxt:
+            above = nxt.start - 1
+            s.end = above if above > s.start and _is_lone_anchor(lines[above]) else nxt.start
+    return out
+
+
+def anchors(text: str) -> set[str]:
+    return {a for s in sections(text) for a in s.anchors}
+
+
+def find(text: str, anchor: str) -> DocSection | None:
+    found = sections(text)
+    return next((s for s in found if anchor in s.anchors and s.level), None) or \
+        next((s for s in found if anchor in s.anchors), None)
+
+
+def resolve(root: Path, path: str, anchor: str | None, limit: int = 80) -> str:
+    target = root / path
+    shown = path.rstrip("/")
+    if target.is_dir():
+        return f"{shown}/ (directory)"
+    if not target.is_file():
+        return f"(missing: {shown})"
+    text = target.read_text(encoding="utf-8", errors="replace")
+    lines = _lines(text)
+    if anchor is None:
+        heading = next((m.group(2) for l in lines if (m := HEADING_RE.match(l))), "")
+        goal = next((l[len("**Goal:**"):].strip() for l in lines if l.startswith("**Goal:**")), "")
+        return shown + (f": {heading}" if heading else "") + (f" — {goal}" if goal else "")
+    s = find(text, anchor)
+    if s is None:
+        return f"(no anchor '{anchor}' in {shown})"
+    body = lines[s.start:s.end]
+    while body and not body[-1].strip():
+        body.pop()
+    out = [f"===== {shown}#{anchor} (line {s.start + 1}) =====", *body[:limit]]
+    if len(body) > limit:
+        out.append(f"… ({len(body) - limit} more lines: {shown}:{s.start + limit + 1})")
+    return "\n".join(out)