#!/usr/bin/env python3
"""Wake digest — cheap, computed session-start context.

Emits ~600 tokens of CURRENT state at SessionStart, replacing the ~1,000-token
session-handoff.md snapshot that had gone five months stale.

The governing property: this is DERIVED, never CACHED. A stored digest drifts
silently from the substrate — that is precisely how MemPalace's boot context
failed. A computation over the substrate cannot drift; it can only have bugs,
and bugs are found by positive controls. Every section here degrades honestly:
if it cannot be computed it says so rather than emitting nothing.

  wake-digest.py             emit the digest
  wake-digest.py --selftest  prove each extractor detects presence AND absence

Provenance: 2026-07-28, steward-authorized alongside the PENDING split and the
CLAUDE.md doctrine annotation. Sibling of governance-drift-check.py.
"""
import os, re, subprocess, sys, time

HOME = os.path.expanduser("~")
D = os.path.join(HOME, "dotfiles")
MEM = os.path.join(HOME, ".claude", "projects", "-Users-davidglidden", "memory")
PENDING = os.path.join(D, "PENDING.md")
REVIEWED = os.path.join(D, "REVIEWED.md")
DRIFT = os.path.join(D, "scripts", "governance-drift-check.py")
REPOS = ["CapableMind-AI", "BetterMemories.io", "chamber-library",
         "animal-davidglidden-eu", "studium-engine"]

warn = []


def sh(args, cwd=None, timeout=8):
    try:
        r = subprocess.run(args, cwd=cwd, capture_output=True, text=True, timeout=timeout)
        return r.stdout.strip() if r.returncode == 0 else None
    except Exception:
        return None


def read(p):
    try:
        return open(p, encoding="utf-8").read()
    except Exception:
        return None


# ---------- extractors (pure functions of text, so --selftest can exercise them) ----------

def open_items(text):
    """-> [(header, line_no)] for every '## ' item lacking closure evidence.
    Closure evidence = a REVIEWED-N (resolved by the caller) or CLOSED/COMPLETED
    in the header. Any '## ' line is an item — assuming one naming family missed
    twenty items on 2026-07-28."""
    out = []
    for i, l in enumerate(text.split("\n"), 1):
        if l.startswith("## ") and "CLOSED" not in l and not l.startswith("## COMPLETED"):
            out.append((l[3:].strip(), i))
    return out


def tag_of(text, header):
    m = re.search(re.escape("## " + header) + r".*?\n\*\*Tag:\*\*\s*(\[[A-Z]+\])",
                  text, re.S)
    return m.group(1) if m else ""


def strip_frontmatter(text):
    """YAML frontmatter restates the thread inside `description:`, and matching
    there runs the extraction on into `metadata:`. The body is the record."""
    m = re.match(r"^---\n.*?\n---\n", text, re.S)
    return text[m.end():] if m else text


def extract_anchor(text, anchor, limit=900):
    """Verbatim text following an anchor. Never summarised — a digest that
    paraphrases the steward's own words is the lossy extraction this whole
    architecture is designed to avoid.

    Handles both wrap shapes: `**ANCHOR: content**` (label and content share the
    bold run) and `**ANCHOR for next-Claude:** content` (label closes first)."""
    text = strip_frontmatter(text)
    i = text.find(anchor)
    if i < 0:
        return None
    s = text[i + len(anchor): i + len(anchor) + limit + 120]
    c = s.find(":")
    if 0 <= c <= 40:                      # remainder of the label, not content
        s = s[c + 1:]
    cut = re.search(r"\n\s*\n", s)
    if cut:
        s = s[: cut.start()]
    s = re.sub(r"\s+", " ", s.replace("**", "")).strip(" *—-")
    return (s[:limit].rstrip() + " …[truncated]") if len(s) > limit else s


# ---------- sections ----------

def sec_pause():
    files = [f for f in os.listdir(MEM)
             if f.startswith("session-") and not f.startswith("session-ledger-")]
    if not files:
        warn.append("no session memory files found")
        return None, None
    newest = max(files, key=lambda f: os.path.getmtime(os.path.join(MEM, f)))
    age = time.time() - os.path.getmtime(os.path.join(MEM, newest))
    h = age / 3600
    span = f"{age/60:.0f} min" if h < 1 else (f"{h:.1f} h" if h < 48 else f"{h/24:.1f} days")
    return newest, span


def sec_pending():
    t = read(PENDING)
    if t is None:
        warn.append("PENDING.md unreadable")
        return []
    rt = read(REVIEWED) or ""
    rev = {m for m in re.findall(r"^## REVIEWED-(\S+)", rt, re.M)}
    items = []
    for h, ln in open_items(t):
        m = re.match(r"PENDING-(\S+?)\s*—", h)
        if m and m.group(1) in rev:
            continue
        items.append((h, ln, tag_of(t, h)))
    if not items:
        warn.append("PENDING.md parsed but yielded no open items — check the parser")
    return items


def sec_reviewed(n=3):
    t = read(REVIEWED)
    if t is None:
        warn.append("REVIEWED.md unreadable")
        return []
    hs = re.findall(r"^## (REVIEWED-\S+ —.*)$", t, re.M)
    out = []
    for h in hs[-n:]:
        m = re.search(re.escape("## " + h) + r".*?\n\*\*Decision:\*\*\s*(\w+)", t, re.S)
        out.append((h[:88], m.group(1) if m else "?"))
    return out


def sec_drift():
    o = sh([sys.executable, DRIFT])
    if o is None:
        warn.append("drift check did not run")
        return None
    if "INSTRUMENT NOT VERIFIED" in o:
        return "INSTRUMENT NOT VERIFIED — treat as unestablished"
    m = re.search(r"(\d+) claim", o)
    return m.group(1) if m else "0"


def sec_repos():
    rows = []
    for r in REPOS:
        p = os.path.join(HOME, "_Dev", r)
        if not os.path.isdir(os.path.join(p, ".git")):
            continue
        sb = sh(["git", "-C", p, "status", "-sb", "--porcelain"]) or ""
        lines = sb.split("\n")
        head = lines[0][3:] if lines and lines[0].startswith("##") else "?"
        dirty = sum(1 for l in lines[1:] if l.strip())
        subj = sh(["git", "-C", p, "log", "-1", "--format=%s"]) or "?"
        rows.append((r, head, dirty, subj[:60]))
    if not rows:
        warn.append("no repos resolved")
    return rows


def sec_thread(newest):
    t = read(os.path.join(MEM, newest)) if newest else None
    if t is None:
        warn.append("session file unreadable — no thread or question")
        return None, None
    th = extract_anchor(t, "PULLING THREAD", 400)
    q = extract_anchor(t, "LITERAL QUESTION", 900)
    if th is None:
        warn.append("no PULLING THREAD anchor in the last wrap")
    if q is None:
        warn.append("no LITERAL QUESTION anchor in the last wrap")
    return th, q


# ---------- the .app brief ----------
# The jurist reads Claude.app's personal preferences and has NO filesystem access, so
# unlike CLAUDE.md that document cannot compute its state — it can only cache it. A cache
# drifts; the mitigations are to keep it small, generate rather than hand-write it, and
# date it so staleness is visible rather than misleading (Constitutional Constraint #4
# across a boundary we cannot check). This emits a paste-ready block and records when it
# was generated, so the wake can report the age.
BRIEF_PATH = os.path.join(D, "claude", "app-brief.md")


def trackers():
    t = read(os.path.join(MEM, "MEMORY.md"))
    if t is None:
        warn.append("MEMORY.md unreadable — no workstreams for the brief")
        return []
    m = re.search(r"^## Canonical Workstream Trackers\s*$(.*?)^## ", t, re.S | re.M)
    if not m:
        warn.append("MEMORY.md has no 'Canonical Workstream Trackers' heading")
        return []
    def clip(s, n):
        s = re.sub(r"\s+", " ", s).strip()
        if len(s) <= n:
            return s
        return s[:n].rsplit(" ", 1)[0] + "…"

    out = []
    for line in m.group(1).split("\n"):
        if not line.startswith("- "):
            continue
        body = re.sub(r"\[([^\]]+)\]\([^)]*\)", r"\1", line[2:])   # unlink
        body = re.sub(r"[*`]", "", body)
        name, _, gist = body.partition(" — ")
        out.append((clip(name, 38), clip(gist, 72)))
    return out


def commits_30d(path):
    o = sh(["git", "-C", path, "log", "--since=30.days", "--oneline"])
    return len(o.split("\n")) if o else 0


def brief(stamp):
    """Replaces ONLY the projects half of the .app preferences' §Standing Context.
    The personal entries there (the orchestra conflict, the divorce, fraternal practice)
    are steward-held, unverifiable from any substrate, and must never be generated."""
    items = sec_pending()
    o = [f"### Standing Context — Projects   *(generated {stamp}; do not hand-edit)*",
         "",
         "Regenerate: `python3 ~/dotfiles/scripts/wake-digest.py --brief`. This is a snapshot,",
         "not a live view — the reader of this document has no filesystem access, so it cannot",
         "be computed here. If today is more than ~30 days after the date above, treat every",
         "line as unverified rather than current. Personal standing context is kept separately",
         "and by hand.", "",
         "TRACKER INDEX (from MEMORY.md — what exists, not what is hot; see repo activity"
         " below for that)"]
    for n, tail in trackers():
        o.append(f"  {n:<38}  {tail}")
    o.append(f"\nOPEN AUTHORIZATION ITEMS ({len(items)}) — full text in ~/PENDING.md; "
             "closed items in ~/PENDING-archive.md")
    for h, _ln, tag in items:
        o.append(f"  {tag:<12} {h[:92]}")
    o.append("\nLAST RULINGS")
    for h, d in sec_reviewed(4):
        o.append(f"  {d:<10} {h}")
    o.append("\nREPO ACTIVITY (commits, last 30 days)")
    for r in REPOS:
        p = os.path.join(HOME, "_Dev", r)
        if os.path.isdir(os.path.join(p, ".git")):
            o.append(f"  {r:<24} {commits_30d(p):>4}")
    o.append(f"\nGOVERNANCE DRIFT — ~/CLAUDE.md: {sec_drift()} substrate-contradicted claim(s)")
    o.append("\n*(end generated block)*")
    return "\n".join(o) + "\n"


def brief_age_days():
    """Age from the date written INSIDE the brief, never from mtime: git does not
    preserve mtime, so a fresh clone would make a stale brief read as newly
    generated — false freshness inside the freshness instrument."""
    t = read(BRIEF_PATH)
    if t is None:
        return None
    m = re.search(r"generated (\d{4})-(\d{2})-(\d{2})", t)
    if not m:
        warn.append("app-brief.md carries no 'generated <date>' stamp — age unknown")
        return None
    y, mo, d = (int(x) for x in m.groups())
    return (time.time() - time.mktime((y, mo, d, 12, 0, 0, 0, 0, -1))) / 86400


# ---------- self-test ----------

def selftest():
    ok = True

    def chk(name, cond):
        nonlocal ok
        ok = ok and cond
        print(f"  [{'ok  ' if cond else 'FAIL'}] {name}")

    print("extractor controls (presence AND absence):")
    chk("open_items finds a plain item",
        open_items("## PENDING-1 — x") == [("PENDING-1 — x", 1)])
    chk("open_items finds a non-numeric family  [v1 splitter's blind spot]",
        open_items("## PENDING-S2 — x\n## PENDING — ICP-19") and
        len(open_items("## PENDING-S2 — x\n## PENDING — ICP-19")) == 2)
    chk("open_items excludes CLOSED / COMPLETED",
        open_items("## PENDING-1 — x (CLOSED)\n## COMPLETED — y") == [])
    chk("open_items returns empty on empty input", open_items("") == [])
    chk("tag_of reads a tag",
        tag_of("## PENDING-9 — t\n**Date:** d\n**Tag:** [ESCALATE]\n", "PENDING-9 — t") == "[ESCALATE]")
    chk("tag_of returns '' when absent", tag_of("## PENDING-9 — t\n", "PENDING-9 — t") == "")
    chk("extract_anchor pulls verbatim",
        extract_anchor("noise\n**PULLING THREAD:** do the thing.\n\nmore", "PULLING THREAD")
        == "do the thing.")
    chk("extract_anchor handles label-inside-bold  (**A: content**)",
        extract_anchor("**PULLING THREAD: do the thing.**\n\nmore", "PULLING THREAD")
        == "do the thing.")
    chk("extract_anchor drops a label remnant  (**A for next-Claude:** content)",
        extract_anchor("**LITERAL QUESTION for next-Claude:** Is it?\n\nx", "LITERAL QUESTION")
        == "Is it?")
    chk("extract_anchor skips YAML frontmatter",
        extract_anchor('---\ndescription: "PULLING THREAD: the wrong one."\ntype: x\n---\n'
                       "**PULLING THREAD: the right one.**\n\nz", "PULLING THREAD")
        == "the right one.")
    chk("extract_anchor marks truncation",
        extract_anchor("**A:** " + "w " * 200, "A", limit=50).endswith("…[truncated]"))
    chk("extract_anchor returns None when the anchor is absent",
        extract_anchor("no anchor here", "PULLING THREAD") is None)
    print("\nlive substrate:")
    chk("PENDING.md readable", read(PENDING) is not None)
    chk("REVIEWED.md readable", read(REVIEWED) is not None)
    chk("drift check runs", sec_drift() is not None)
    chk("memory dir resolves", os.path.isdir(MEM))
    print("\nSELFTEST", "PASS" if ok else "FAIL")
    return 0 if ok else 2


# ---------- render ----------

def main():
    newest, span = sec_pause()
    th, q = sec_thread(newest)
    items, revs, drift, repos = sec_pending(), sec_reviewed(), sec_drift(), sec_repos()

    o = ["=== WAKE DIGEST (computed now — not a stored snapshot) ==="]
    o.append(f"Last wrap: {span} ago  ({newest})" if span else "Last wrap: UNKNOWN")
    if th:
        o.append(f"\nPULLING THREAD — {th}")
    if q:
        o.append(f"\nOPEN QUESTION — {q}")

    o.append(f"\nOPEN AUTHORIZATION ITEMS ({len(items)}) — bodies in ~/PENDING.md at the line given;"
             " closed items are in ~/PENDING-archive.md")
    for h, ln, tag in items:
        o.append(f"  L{ln:<5} {tag:<12} {h[:78]}")

    if revs:
        o.append("\nLAST RULINGS")
        for h, d in revs:
            o.append(f"  {d:<10} {h}")

    o.append(f"\nGOVERNANCE DRIFT — ~/CLAUDE.md: {drift} substrate-contradicted claim(s)"
             " (detection only; correction needs [ESCALATE])")

    age = brief_age_days()
    if age is None:
        o.append("\n.APP BRIEF — never generated. The jurist's §Standing Context cannot be"
                 " checked from here; run `wake-digest.py --brief` and paste it.")
    elif age > 30:
        o.append(f"\n.APP BRIEF — last GENERATED {age:.0f} days ago. This tracks generation,"
                 " not pasting: it is a lower bound on the jurist's staleness, never a"
                 " guarantee of freshness.")

    if repos:
        o.append("\nREPOS")
        for r, head, dirty, subj in repos:
            flag = f" ~{dirty} dirty" if dirty else ""
            o.append(f"  {r:<24} {head:<34}{flag}  {subj}")

    if warn:
        o.append("\n⚠ DIGEST DEGRADED — these sections could not be computed:")
        for w in warn:
            o.append(f"  - {w}")
    o.append("=== END WAKE DIGEST ===")
    print("\n".join(o))


def emit_brief():
    stamp = sh(["date", "+%Y-%m-%d"]) or "date-unavailable"
    text = brief(stamp)
    os.makedirs(os.path.dirname(BRIEF_PATH), exist_ok=True)
    open(BRIEF_PATH, "w", encoding="utf-8").write(text)
    print(text)
    print(f"[written to {BRIEF_PATH} — the wake reports this file's age]", file=sys.stderr)
    if warn:
        print("⚠ brief degraded: " + "; ".join(warn), file=sys.stderr)


if __name__ == "__main__":
    try:
        if "--brief" in sys.argv:
            sys.exit(emit_brief())
        sys.exit(selftest() if "--selftest" in sys.argv else main())
    except Exception as e:                                   # never break session start
        print(f"=== WAKE DIGEST UNAVAILABLE — {type(e).__name__}: {e} ===")