rewrite clone_forks in python

teach_utils/clone_forks.py

+#!/usr/bin/env python
+# Reads JSON output from collect-forks (either as a file or in stdin)
+# and clone all target forks into provided directory at the target commit.
+
+# usage ./clone_forks_from_json.sh OUT_DIR forks.json
+#       ./collect_forks.py --token token.asc URL | ./clone_forks_from_json.sh OUT_DIR
+
+import sys
+from pathlib import Path
+import json
+import click
+import git
+
+
+def clone_fork(url: str, out_path: Path, commit: str):
+    """
+    Clone input repo in target directory
+    and checkout desired commit."""
+    cloned = git.Repo.clone_from(url, out_path)
+    cloned.git.checkout(commit)
+
+
+@click.command()
+@click.argument("out_dir", type=click.Path(file_okay=False, exists=False))
+@click.argument("forks", type=click.File(mode="r"), default=sys.stdin)
+def main(out_dir, forks, token):
+
+    forks = json.load(forks)
+    for fork in forks:
+        url, group, commit = [fork.get(key) for key in ["url", "group", "commit"]]
+        path = Path(out_dir) / group
+        clone_fork(url, path, commit)

teach_utils/clone_forks_from_json.sh

-#!/usr/bin/env bash
-# Reads JSON output from collect_forks.py (either as a file or in stdin)
-# and clone all target forks into provided directory at the target commit.
-
-# usage ./clone_forks_from_json.sh OUT_DIR forks.json
-#       ./collect_forks.py --token token.asc URL | ./clone_forks_from_json.sh OUT_DIR
-
-# Help message
-function usage () {
-   cat <<EOF
-Usage:
-    ./$(basename $0) [outdir] [in_file.json]
-    ./collect_forks.py --token token.asc URL | ./$(basename $0) [outdir]
-
-Reads json output from collect_forks.py and clone all target forks into provided directory
-at the target commit. Repositories are cloned into outdir/namespace.
-
-Arguments:
-    outdir: Directory where all forks will be cloned [default: .]
-    in_file.json: JSON output of collect_forks.py containing fork metadata [default: stdin]
-EOF
-   exit 0
-}
-
-# Parsing CL arguments
-OUT_DIR=${1:-.}
-JSON=${2:-/dev/stdin}
-if [[ $# -gt 2 ]] || [[ $1 == '-h' ]] || [[ $1 == '--help' ]]; then
-    usage
-fi
-
-mkdir -p "${OUT_DIR}"
-
-jq ".[] | \"git clone \(.url) $OUT_DIR/\(.group) && cd ${OUT_DIR}/\(.group) && git checkout \(.commit)\"" \
-    < "${JSON}" \
-    | xargs -L 1 -I {} sh -c "{}"

teach_utils/collect_forks.py

@@ -49,7 +49,7 @@ def collect_forks(project_url: str, header=Dict[str, str]) -> List[Dict]:
 
 def filter_group_forks(forks: List[Dict]) -> List[Dict]:
     """Given a list of forks' metadata, only keep those that belong to a group
-    
+
     Examples
     --------
     >>> d1 = {'id': 1, 'namespace': {'kind': 'user'}}
@@ -93,9 +93,9 @@ def get_last_commit_hash(
 def format_fork_metadata(
     fork: Dict, header: Dict[str, str], deadline: Optional[str] = None
 ) -> Dict:
-    """Format and add fields to a fork's metadata. The 
+    """Format and add fields to a fork's metadata. The
     resulting metadata will have the following fields:
-    id, http_url_to_repo, autostart_url, commit, members, visibility, group"""
+    id, url, autostart_url, commit, members, visibility, group"""
     meta = {
         "id": fork["id"],
         "url": fork["http_url_to_repo"],