From e08257d4012b6df624c88b45ba28e8f9829ce2ad Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal+gh@gmail.com>
Date: Sun, 28 Jun 2020 14:07:32 +0200
Subject: [PATCH] Add example of how to do sparse-checkout

---
 spacy/cli/_git_sparse_checkout_example.py | 62 +++++++++++++++++++++++
 1 file changed, 62 insertions(+)
 create mode 100644 spacy/cli/_git_sparse_checkout_example.py

diff --git a/spacy/cli/_git_sparse_checkout_example.py b/spacy/cli/_git_sparse_checkout_example.py
new file mode 100644
index 000000000..2bb3d6734
--- /dev/null
+++ b/spacy/cli/_git_sparse_checkout_example.py
@@ -0,0 +1,62 @@
+import tempfile
+import typer
+from pathlib import Path
+import subprocess
+import shlex
+import shutil
+from contextlib import contextmanager
+
+
+@contextmanager
+def make_tempdir():
+    d = Path(tempfile.mkdtemp())
+    yield d
+    shutil.rmtree(str(d))
+
+
+
+def clone_repo(repo, temp_dir):
+    subprocess.check_call([
+        "git",
+        "clone",
+        repo,
+        temp_dir,
+        "--no-checkout",
+        "--depth", "1",
+        "--config", "core.sparseCheckout=true"
+    ])
+
+
+def checkout_and_fetch(temp_dir):
+    subprocess.check_call([
+        "git",
+        "-C", temp_dir,
+        "fetch"
+    ])
+    subprocess.check_call([
+        "git",
+        "-C", temp_dir,
+        "checkout"
+    ])
+
+
+def set_sparse_checkout_dir(temp_dir, subpath):
+    with (temp_dir / ".git" / "info" / "sparse-checkout").open("w") as file_:
+        file_.write(subpath)
+
+
+def main(repo: str, subpath: str, dest: Path):
+    with make_tempdir() as temp_dir:
+        clone_repo(repo, temp_dir)
+        print("After clone", list(temp_dir.iterdir()))
+        set_sparse_checkout_dir(temp_dir, subpath)
+        checkout_and_fetch(temp_dir)
+        print("After checkout", list(temp_dir.iterdir()))
+        assert (temp_dir / subpath) in list(temp_dir.iterdir())
+        shutil.copytree(temp_dir / subpath, dest / subpath, dirs_exist_ok=True)
+    print("Exists after cleanup?", temp_dir.exists())
+    print("Destination", list(dest.iterdir()))
+
+
+if __name__ == "__main__":
+    typer.run(main)