Merge branch 'main' into fix/windows-support

Author: filipchristiansen

README.md

@@ -122,6 +122,9 @@ gitingest https://github.com/username/private-repo --token github_pat_...
 # Or set it as an environment variable
 export GITHUB_TOKEN=github_pat_...
 gitingest https://github.com/username/private-repo
+
+# Include repository submodules
+gitingest https://github.com/username/repo-with-submodules --include-submodules
 ```
 
 By default, files listed in `.gitignore` are skipped. Use `--include-gitignored` if you
@@ -163,6 +166,9 @@ summary, tree, content = ingest("https://github.com/username/private-repo", toke
 import os
 os.environ["GITHUB_TOKEN"] = "github_pat_..."
 summary, tree, content = ingest("https://github.com/username/private-repo")
+
+# Include repository submodules
+summary, tree, content = ingest("https://github.com/username/repo-with-submodules", include_submodules=True)
 ```
 
 By default, this won't write a file but can be enabled with the `output` argument.

src/gitingest/cli.py

@@ -20,6 +20,7 @@ class _CLIArgs(TypedDict):
     include_pattern: tuple[str, ...]
     branch: str | None
     include_gitignored: bool
+    include_submodules: bool
     token: str | None
     output: str | None
 
@@ -47,6 +48,12 @@ class _CLIArgs(TypedDict):
     default=False,
     help="Include files matched by .gitignore and .gitingestignore",
 )
+@click.option(
+    "--include-submodules",
+    is_flag=True,
+    help="Include repository's submodules in the analysis",
+    default=False,
+)
 @click.option(
     "--token",
     "-t",
@@ -94,6 +101,9 @@ def main(**cli_kwargs: Unpack[_CLIArgs]) -> None:
         $ gitingest https://github.com/user/private-repo -t ghp_token
         $ GITHUB_TOKEN=ghp_token gitingest https://github.com/user/private-repo
 
+    Include submodules:
+        $ gitingest https://github.com/user/repo --include-submodules
+
     """
     asyncio.run(_async_main(**cli_kwargs))
 
@@ -106,6 +116,7 @@ async def _async_main(
     include_pattern: tuple[str, ...] | None = None,
     branch: str | None = None,
     include_gitignored: bool = False,
+    include_submodules: bool = False,
     token: str | None = None,
     output: str | None = None,
 ) -> None:
@@ -129,6 +140,8 @@ async def _async_main(
         Git branch to ingest. If ``None``, the repository's default branch is used.
     include_gitignored : bool
         If ``True``, also ingest files matched by ``.gitignore`` or ``.gitingestignore`` (default: ``False``).
+    include_submodules : bool
+        If ``True``, recursively include all Git submodules within the repository (default: ``False``).
     token : str | None
         GitHub personal access token (PAT) for accessing private repositories.
         Can also be set via the ``GITHUB_TOKEN`` environment variable.
@@ -155,14 +168,15 @@ async def _async_main(
             click.echo(f"Analyzing source, output will be written to '{output_target}'...", err=True)
 
         summary, _, _ = await ingest_async(
-            source=source,
+            source,
             max_file_size=max_size,
             include_patterns=include_patterns,
             exclude_patterns=exclude_patterns,
             branch=branch,
-            output=output_target,
             include_gitignored=include_gitignored,
+            include_submodules=include_submodules,
             token=token,
+            output=output_target,
         )
     except Exception as exc:
         # Convert any exception into Click.Abort so that exit status is non-zero

src/gitingest/clone.py

@@ -63,7 +63,9 @@ async def clone_repo(config: CloneConfig, *, token: str | None = None) -> None:
         clone_cmd += ["-c", create_git_auth_header(token, url=url)]
 
     clone_cmd += ["clone", "--single-branch"]
-    # TODO: Re-enable --recurse-submodules when submodule support is needed
+
+    if config.include_submodules:
+        clone_cmd += ["--recurse-submodules"]
 
     if partial_clone:
         clone_cmd += ["--filter=blob:none", "--sparse"]
@@ -86,15 +88,28 @@ async def clone_repo(config: CloneConfig, *, token: str | None = None) -> None:
 
     # Checkout the subpath if it is a partial clone
     if partial_clone:
-        subpath = config.subpath.lstrip("/")
-        if config.blob:
-            # When ingesting from a file url (blob/branch/path/file.txt), we need to remove the file name.
-            subpath = str(Path(subpath).parent.as_posix())
-
-        checkout_cmd = create_git_command(["git"], local_path, url, token)
-        await run_command(*checkout_cmd, "sparse-checkout", "set", subpath)
+        await _checkout_partial_clone(config, token)
 
     # Checkout the commit if it is provided
     if commit:
         checkout_cmd = create_git_command(["git"], local_path, url, token)
         await run_command(*checkout_cmd, "checkout", commit)
+
+
+async def _checkout_partial_clone(config: CloneConfig, token: str | None) -> None:
+    """Configure sparse-checkout for a partially cloned repository.
+
+    Parameters
+    ----------
+    config : CloneConfig
+        The configuration for cloning the repository, including subpath and blob flag.
+    token : str | None
+        GitHub personal access token (PAT) for accessing private repositories.
+
+    """
+    subpath = config.subpath.lstrip("/")
+    if config.blob:
+        # Remove the file name from the subpath when ingesting from a file url (e.g. blob/branch/path/file.txt)
+        subpath = str(Path(subpath).parent.as_posix())
+    checkout_cmd = create_git_command(["git"], config.local_path, config.url, token)
+    await run_command(*checkout_cmd, "sparse-checkout", "set", subpath)

src/gitingest/entrypoint.py

@@ -27,6 +27,7 @@ async def ingest_async(
     branch: str | None = None,
     tag: str | None = None,
     include_gitignored: bool = False,
+    include_submodules: bool = False,
     token: str | None = None,
     output: str | None = None,
 ) -> tuple[str, str, str]:
@@ -52,6 +53,8 @@ async def ingest_async(
         The tag to clone and ingest. If ``None``, no tag is used.
     include_gitignored : bool
         If ``True``, include files ignored by ``.gitignore`` and ``.gitingestignore`` (default: ``False``).
+    include_submodules : bool
+        If ``True``, recursively include all Git submodules within the repository (default: ``False``).
     token : str | None
         GitHub personal access token (PAT) for accessing private repositories.
         Can also be set via the ``GITHUB_TOKEN`` environment variable.
@@ -86,6 +89,8 @@ async def ingest_async(
     if query.url:
         _override_branch_and_tag(query, branch=branch, tag=tag)
 
+    query.include_submodules = include_submodules
+
     async with _clone_repo_if_remote(query, token=token):
         summary, tree, content = ingest_query(query)
         await _write_output(tree, content=content, target=output)
@@ -101,6 +106,7 @@ def ingest(
     branch: str | None = None,
     tag: str | None = None,
     include_gitignored: bool = False,
+    include_submodules: bool = False,
     token: str | None = None,
     output: str | None = None,
 ) -> tuple[str, str, str]:
@@ -126,6 +132,8 @@ def ingest(
         The tag to clone and ingest. If ``None``, no tag is used.
     include_gitignored : bool
         If ``True``, include files ignored by ``.gitignore`` and ``.gitingestignore`` (default: ``False``).
+    include_submodules : bool
+        If ``True``, recursively include all Git submodules within the repository (default: ``False``).
     token : str | None
         GitHub personal access token (PAT) for accessing private repositories.
         Can also be set via the ``GITHUB_TOKEN`` environment variable.
@@ -156,6 +164,7 @@ def ingest(
             branch=branch,
             tag=tag,
             include_gitignored=include_gitignored,
+            include_submodules=include_submodules,
             token=token,
             output=output,
         ),

src/gitingest/schemas/ingestion.py

@@ -11,7 +11,7 @@
 
 
 @dataclass
-class CloneConfig:
+class CloneConfig:  # pylint: disable=too-many-instance-attributes
     """Configuration for cloning a Git repository.
 
     This class holds the necessary parameters for cloning a repository to a local path, including
@@ -33,6 +33,8 @@ class CloneConfig:
         The subpath to clone from the repository (default: ``"/"``).
     blob: bool
         Whether the repository is a blob (default: ``False``).
+    include_submodules: bool
+        Whether to clone submodules (default: ``False``).
 
     """
 
@@ -43,6 +45,7 @@ class CloneConfig:
     tag: str | None = None
     subpath: str = "/"
     blob: bool = False
+    include_submodules: bool = False
 
 
 class IngestionQuery(BaseModel):  # pylint: disable=too-many-instance-attributes
@@ -78,6 +81,8 @@ class IngestionQuery(BaseModel):  # pylint: disable=too-many-instance-attributes
         The patterns to ignore (default: ``set()``).
     include_patterns : set[str] | None
         The patterns to include.
+    include_submodules : bool
+        Whether to include all Git submodules within the repository. (default: ``False``)
 
     """
 
@@ -95,6 +100,7 @@ class IngestionQuery(BaseModel):  # pylint: disable=too-many-instance-attributes
     max_file_size: int = Field(default=MAX_FILE_SIZE)
     ignore_patterns: set[str] = set()  # TODO: ignore_patterns and include_patterns have the same type
     include_patterns: set[str] | None = None
+    include_submodules: bool = False
 
     def extract_clone_config(self) -> CloneConfig:
         """Extract the relevant fields for the CloneConfig object.
@@ -122,6 +128,7 @@ def extract_clone_config(self) -> CloneConfig:
             tag=self.tag,
             subpath=self.subpath,
             blob=self.type == "blob",
+            include_submodules=self.include_submodules,
         )
 
     def ensure_url(self) -> None:

tests/query_parser/test_git_host_agnostic.py

@@ -68,6 +68,7 @@ async def test_parse_query_without_host(
         "commit": None,
         "max_file_size": 50,
         "include_patterns": None,
+        "include_submodules": False,
     }
 
     assert actual == expected

tests/test_cli.py

@@ -27,6 +27,7 @@
                 "tests/",
                 "--include-pattern",
                 "src/",
+                "--include-submodules",
             ],
             True,
             id="custom-options",

tests/test_clone.py

@@ -181,10 +181,10 @@ async def test_clone_default_shallow_clone(run_command_mock: AsyncMock) -> None:
 
 
 @pytest.mark.asyncio
-async def test_clone_commit_without_branch(run_command_mock: AsyncMock) -> None:
-    """Test cloning when a commit hash is provided but no branch is specified.
+async def test_clone_commit(run_command_mock: AsyncMock) -> None:
+    """Test cloning when a commit hash is provided.
 
-    Given a valid URL and a commit hash (but no branch):
+    Given a valid URL and a commit hash:
     When ``clone_repo`` is called,
     Then the repository should be cloned and checked out at that commit.
     """
@@ -414,3 +414,28 @@ async def test_clone_with_commit_and_subpath(run_command_mock: AsyncMock) -> Non
     )
 
     assert run_command_mock.call_count == expected_call_count
+
+
+@pytest.mark.asyncio
+async def test_clone_with_include_submodules(run_command_mock: AsyncMock) -> None:
+    """Test cloning a repository with submodules included.
+
+    Given a valid URL and ``include_submodules=True``:
+    When ``clone_repo`` is called,
+    Then the repository should be cloned with ``--recurse-submodules`` in the git command.
+    """
+    expected_call_count = 1  # No commit and no partial clone
+    clone_config = CloneConfig(url=DEMO_URL, local_path=LOCAL_REPO_PATH, branch="main", include_submodules=True)
+
+    await clone_repo(clone_config)
+
+    assert run_command_mock.call_count == expected_call_count
+    run_command_mock.assert_called_once_with(
+        "git",
+        "clone",
+        "--single-branch",
+        "--recurse-submodules",
+        "--depth=1",
+        clone_config.url,
+        clone_config.local_path,
+    )