Get Pr Comments

#!/usr/bin/env python3
"""
Fetch and organize all PR comments (issue-level, review bodies, inline review comments).

Usage:
  fetch_pr_comments.py <pr-number> [--repo OWNER/REPO] [--output json|text]

Examples:
  fetch_pr_comments.py 31
  fetch_pr_comments.py 33 --repo gupsammy/Claudest --output json
"""

from __future__ import annotations

import argparse
import json
import re
import subprocess
import sys


BOT_SUFFIXES = ("[bot]",)


def run_gh(args: list[str]) -> str:
    result = subprocess.run(
        ["gh"] + args,
        capture_output=True, text=True, timeout=30
    )
    if result.returncode != 0:
        print(f"gh command failed: {result.stderr.strip()}", file=sys.stderr)
        sys.exit(2)
    return result.stdout.strip()


def _parse_slurped(raw: str) -> list:
    """Parse --slurp output which wraps each page's array in an outer array."""
    if not raw:
        return []
    pages = json.loads(raw)
    # --slurp produces [[page1...], [page2...], ...] — flatten to single list
    return [item for page in pages for item in page]


def is_bot(login: str) -> bool:
    return any(login.endswith(s) for s in BOT_SUFFIXES)


def _api_prefix(repo: str | None) -> str:
    """Return the repos/OWNER/REPO or repos/{owner}/{repo} prefix for gh api."""
    if repo:
        return f"repos/{repo}"
    return "repos/{owner}/{repo}"


def fetch_issue_comments(pr_number: int, repo: str | None) -> list[dict]:
    prefix = _api_prefix(repo)
    raw = run_gh(["api", f"{prefix}/issues/{pr_number}/comments",
                  "--paginate", "--slurp"])
    items = _parse_slurped(raw)
    return [
        {
            "type": "issue_comment",
            "id": c["id"],
            "user": c["user"]["login"],
            "is_bot": is_bot(c["user"]["login"]),
            "body": c["body"],
            "created_at": c["created_at"],
            "url": c.get("html_url", ""),
        }
        for c in items
    ]


def fetch_reviews(pr_number: int, repo: str | None) -> list[dict]:
    prefix = _api_prefix(repo)
    raw = run_gh(["api", f"{prefix}/pulls/{pr_number}/reviews",
                  "--paginate", "--slurp"])
    items = _parse_slurped(raw)
    return [
        {
            "type": "review",
            "id": r["id"],
            "user": r["user"]["login"],
            "is_bot": is_bot(r["user"]["login"]),
            "state": r["state"],
            "body": r["body"],
            "submitted_at": r.get("submitted_at", ""),
            "url": r.get("html_url", ""),
        }
        for r in items
        if r["body"].strip()  # skip empty review bodies
    ]


def fetch_inline_comments(pr_number: int, repo: str | None) -> list[dict]:
    prefix = _api_prefix(repo)
    raw = run_gh(["api", f"{prefix}/pulls/{pr_number}/comments",
                  "--paginate", "--slurp"])
    items = _parse_slurped(raw)
    return [
        {
            "type": "inline_comment",
            "id": c["id"],
            "user": c["user"]["login"],
            "is_bot": is_bot(c["user"]["login"]),
            "body": c["body"],
            "path": c.get("path", ""),
            "line": c.get("line") or c.get("original_line"),
            "side": c.get("side", ""),
            "diff_hunk": c.get("diff_hunk", ""),
            "created_at": c["created_at"],
            "url": c.get("html_url", ""),
            "in_reply_to_id": c.get("in_reply_to_id"),
            "commit_id": c.get("commit_id", ""),
        }
        for c in items
    ]


# --- Actionable item extraction ---

MUST_FIX_PATTERNS = [
    re.compile(r"^###?\s*must.fix", re.IGNORECASE | re.MULTILINE),
    re.compile(r"\*\*must.fix\*\*", re.IGNORECASE),
    re.compile(r"^###?\s*(?:required|blocking|critical)", re.IGNORECASE | re.MULTILINE),
]

OPTIONAL_PATTERNS = [
    re.compile(r"^###?\s*optional", re.IGNORECASE | re.MULTILINE),
    re.compile(r"\*\*optional\*\*", re.IGNORECASE),
    re.compile(r"^###?\s*(?:suggestions?|nit|minor|non-blocking)", re.IGNORECASE | re.MULTILINE),
]


INLINE_SE