address copilot comments

Signed-off-by: Abhishek Kumar <abhishek.mrt22@gmail.com>

Author: shwstppr

tools/contribution/README.md

@@ -30,7 +30,7 @@ rough signal; read the actual contributions before drawing conclusions.
 
 ## Requirements
 
-- Python 3.7+
+- Python 3.8+
 - [`requests`](https://pypi.org/project/requests/): `pip install requests`
 
 ## github_org_contributions.py

tools/contribution/apache_list_mentions.py

@@ -47,7 +47,6 @@
 import mailbox
 import re
 import sys
-from email.utils import parseaddr
 from pathlib import Path
 from typing import Dict, Iterable, List, Tuple
 
@@ -85,15 +84,26 @@ def download_mbox(list_name: str, domain: str, ym: str, out_dir: Path) -> Path:
     out_path = out_dir / f"{list_name}@{domain}-{ym}.mbox"
 
     headers = {"User-Agent": "apache-list-mentions/1.0"}
-    r = requests.get(url, headers=headers, timeout=120)
-
-    if r.status_code == 404 or not r.content.strip():
+    with requests.get(url, headers=headers, timeout=120, stream=True) as r:
+        if r.status_code == 404:
+            raise FileNotFoundError(f"No archive found for {ym}")
+        r.raise_for_status()
+
+        # Stream to a temp file so an interrupted download is never reused as a cached archive
+        tmp_path = out_path.with_suffix(".mbox.part")
+        has_content = False
+        try:
+            with tmp_path.open("wb") as f:
+                for chunk in r.iter_content(chunk_size=1024 * 1024):
+                    has_content = has_content or bool(chunk.strip())
+                    f.write(chunk)
+        except BaseException:
+            tmp_path.unlink(missing_ok=True)
+            raise
+
+    if not has_content:
+        tmp_path.unlink(missing_ok=True)
         raise FileNotFoundError(f"No archive found for {ym}")
-
-    r.raise_for_status()
-    # Write via a temp file so an interrupted download is never reused as a cached archive
-    tmp_path = out_path.with_suffix(".mbox.part")
-    tmp_path.write_bytes(r.content)
     tmp_path.replace(out_path)
     return out_path
 

tools/contribution/github_org_contributions.py

@@ -84,7 +84,7 @@ def shift_years(d: dt.datetime, years: int) -> dt.datetime:
 
 def year_chunks(from_dt: dt.datetime, to_dt: dt.datetime) -> Iterable[Tuple[dt.datetime, dt.datetime]]:
     """
-    Split [from_dt, to_dt] into consecutive windows of at most one year.
+    Split [from_dt, to_dt) into consecutive end-exclusive windows of at most one year.
     """
     start = from_dt
     while start < to_dt:
@@ -154,7 +154,7 @@ def count_user(token: str, login: str, org_id: str, org_login: str, repo_rx: re.
         })
         user = data.get("user")
         if not user:
-            raise RuntimeError("User not found")
+            raise RuntimeError(f"User not found: {login}")
         user_login = user["login"]
 
         cc = user["contributionsCollection"]