#!/usr/bin/env python3
"""Draft CHANGELOG_PENDING.md entries from Conventional Commit messages
pushed since the last run. Deterministic -- no LLM call, no network access.
Human review still required before folding an entry into CHANGELOG.md.
"""
import re
import subprocess
import sys
from datetime import datetime, timezone

INCLUDE_TYPES = {
    "feat": "Added",
    "fix": "Fixed",
    "refactor": "Changed",
    "perf": "Changed",
    "security": "Fixed",
}
SKIP_TYPES = {"chore", "ci", "test", "docs", "style", "merge"}

# Anything mentioning internal infra by name/address is dropped rather than
# drafted -- better a missing entry (caught in human review) than a leaked one.
INTERNAL_PATTERN = re.compile(
    r"\b(erik|gitea\.context-x|192\.168\.|10\.10\.0\.|opnsense|ssh |systemd|pm2 |crontab)\b",
    re.IGNORECASE,
)

PENDING_HEADER = (
    "# Pending Changelog Entries\n\n"
    "Drafted automatically from commit messages since the last processed "
    "push. Review, edit as needed, and fold into CHANGELOG.md -- then clear "
    "this file.\n\n"
    "<!-- PENDING_ENTRIES -->\n"
)


def run(cmd):
    return subprocess.check_output(cmd, shell=True, text=True).strip()


def main():
    before = sys.argv[1] if len(sys.argv) > 1 else ""
    after = sys.argv[2] if len(sys.argv) > 2 else "HEAD"
    if not before or set(before) == {"0"}:
        print("No usable commit range (first push on branch) -- skipping.")
        return

    try:
        log = run("git log {}..{} --format='%s'".format(before, after))
    except subprocess.CalledProcessError:
        print("Could not diff commit range -- skipping.")
        return
    if not log:
        print("No new commits.")
        return

    buckets = {"Added": [], "Fixed": [], "Changed": []}
    for subject in log.splitlines():
        m = re.match(r"^(\w+)(\([\w./-]+\))?!?:\s*(.+)$", subject)
        if not m:
            continue
        ctype, _, desc = m.groups()
        ctype = ctype.lower()
        if ctype in SKIP_TYPES or ctype not in INCLUDE_TYPES:
            continue
        if INTERNAL_PATTERN.search(desc):
            continue
        buckets[INCLUDE_TYPES[ctype]].append(desc.strip())

    if not any(buckets.values()):
        print("Nothing changelog-worthy in this push.")
        return

    date = datetime.now(timezone.utc).strftime("%Y-%m-%d")
    lines = ["## Pending -- {}\n".format(date)]
    for section in ("Added", "Fixed", "Changed"):
        if buckets[section]:
            lines.append("### {}".format(section))
            for item in buckets[section]:
                lines.append("- {}".format(item))
            lines.append("")
    entry = "\n".join(lines).rstrip() + "\n\n"

    try:
        with open("CHANGELOG_PENDING.md", "r") as f:
            existing = f.read()
        if not existing.strip():
            existing = PENDING_HEADER  # pre-existing but empty placeholder file
    except FileNotFoundError:
        existing = PENDING_HEADER

    marker = "<!-- PENDING_ENTRIES -->\n"
    if marker in existing:
        head, tail = existing.split(marker, 1)
        updated = head + marker + entry + tail.lstrip("\n")
    else:
        updated = existing.rstrip("\n") + "\n\n" + entry

    with open("CHANGELOG_PENDING.md", "w") as f:
        f.write(updated)

    print("Wrote entry:\n" + entry)


if __name__ == "__main__":
    main()