Sanitized mirror from private repository - 2026-04-20 01:32:01 UTC

scripts/changelog-generator.py

@@ -0,0 +1,143 @@
+#!/usr/bin/env python3
+"""Generate changelogs from git commits using Ollama LLM.
+
+Reads recent git commits, groups them via LLM into Features/Fixes/Infrastructure/Documentation,
+and prepends the result to docs/CHANGELOG.md.
+
+Cron: Monday 8am
+  0 8 * * 1 cd /home/homelab/organized/repos/homelab && python3 scripts/changelog-generator.py
+"""
+
+import argparse
+import logging
+import sqlite3
+import subprocess
+import sys
+from datetime import datetime, timedelta
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).parent))
+from lib.ollama import ollama_generate, ollama_available, OllamaUnavailableError
+
+log = logging.getLogger(__name__)
+
+REPO_DIR = Path("/home/homelab/organized/repos/homelab")
+CHANGELOG_PATH = REPO_DIR / "docs" / "CHANGELOG.md"
+DB_PATH = Path(__file__).parent / "changelog.db"
+
+
+def init_db(db_path: Path) -> sqlite3.Connection:
+    """Initialise the SQLite DB and return a connection."""
+    conn = sqlite3.connect(str(db_path))
+    conn.execute(
+        "CREATE TABLE IF NOT EXISTS runs "
+        "(id INTEGER PRIMARY KEY AUTOINCREMENT, last_commit TEXT, run_at TEXT)"
+    )
+    conn.commit()
+    return conn
+
+
+def get_last_commit_date(conn: sqlite3.Connection) -> str | None:
+    """Return the run_at timestamp of the most recent run, or None."""
+    row = conn.execute("SELECT run_at FROM runs ORDER BY id DESC LIMIT 1").fetchone()
+    return row[0] if row else None
+
+
+def fetch_commits(since: str) -> list[str]:
+    """Return git log --oneline entries since *since* (ISO date string)."""
+    result = subprocess.run(
+        ["git", "-C", str(REPO_DIR), "log", "--oneline", f"--since={since}", "main"],
+        capture_output=True,
+        text=True,
+    )
+    if result.returncode != 0:
+        log.error("git log failed: %s", result.stderr.strip())
+        return []
+    lines = [line.strip() for line in result.stdout.strip().splitlines() if line.strip()]
+    return lines
+
+
+def generate_changelog(commits: list[str]) -> str:
+    """Send commits to Ollama and return a formatted changelog."""
+    prompt = (
+        "Summarize these git commits into a changelog grouped by: "
+        "Features, Fixes, Infrastructure, Documentation. "
+        "Use bullet points. Be concise. Omit any empty groups.\n\n"
+        + "\n".join(commits)
+    )
+    return ollama_generate(prompt, num_predict=1500)
+
+
+def prepend_changelog(content: str, date_header: str) -> None:
+    """Prepend a dated changelog entry to CHANGELOG.md."""
+    CHANGELOG_PATH.parent.mkdir(parents=True, exist_ok=True)
+    existing = CHANGELOG_PATH.read_text() if CHANGELOG_PATH.exists() else ""
+    entry = f"## {date_header}\n\n{content}\n\n---\n\n"
+    CHANGELOG_PATH.write_text(entry + existing)
+    log.info("Prepended changelog entry to %s", CHANGELOG_PATH)
+
+
+def record_run(conn: sqlite3.Connection, last_commit: str) -> None:
+    """Record the run in the DB."""
+    conn.execute(
+        "INSERT INTO runs (last_commit, run_at) VALUES (?, ?)",
+        (last_commit, datetime.now().isoformat()),
+    )
+    conn.commit()
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Generate changelog from git commits via LLM")
+    parser.add_argument("--dry-run", action="store_true", help="Print changelog without writing")
+    parser.add_argument("--since", type=str, help="Override start date (YYYY-MM-DD)")
+    parser.add_argument("--verbose", "-v", action="store_true", help="Enable debug logging")
+    args = parser.parse_args()
+
+    logging.basicConfig(
+        level=logging.DEBUG if args.verbose else logging.INFO,
+        format="%(asctime)s [%(levelname)s] %(name)s: %(message)s",
+    )
+
+    if not ollama_available():
+        log.error("Ollama is not reachable — aborting")
+        sys.exit(1)
+
+    conn = init_db(DB_PATH)
+
+    # Determine the since date
+    if args.since:
+        since = args.since
+    else:
+        last_run = get_last_commit_date(conn)
+        if last_run:
+            since = last_run[:10]  # YYYY-MM-DD
+        else:
+            since = (datetime.now() - timedelta(days=7)).strftime("%Y-%m-%d")
+
+    log.info("Fetching commits since %s", since)
+    commits = fetch_commits(since)
+
+    if not commits:
+        log.info("No new commits since %s — nothing to do", since)
+        return
+
+    log.info("Found %d commits, generating changelog...", len(commits))
+    changelog = generate_changelog(commits)
+
+    date_header = datetime.now().strftime("%Y-%m-%d")
+
+    if args.dry_run:
+        print(f"## {date_header}\n")
+        print(changelog)
+        return
+
+    prepend_changelog(changelog, date_header)
+
+    # Record the first commit hash (most recent) in the DB
+    first_hash = commits[0].split()[0]
+    record_run(conn, first_hash)
+    log.info("Done — recorded commit %s", first_hash)
+
+
+if __name__ == "__main__":
+    main()