feature: add branch option to ingest and CLI for repository cloning

requirements-dev.txt

@@ -5,3 +5,4 @@ pre-commit
 pylint
 pytest
 pytest-asyncio
+uvicorn

src/gitingest/cli.py

@@ -14,12 +14,14 @@
 @click.option("--max-size", "-s", default=MAX_FILE_SIZE, help="Maximum file size to process in bytes")
 @click.option("--exclude-pattern", "-e", multiple=True, help="Patterns to exclude")
 @click.option("--include-pattern", "-i", multiple=True, help="Patterns to include")
+@click.option("--branch", "-b", default=None, help="Branch to clone and ingest")
 def main(
     source: str,
     output: str | None,
     max_size: int,
     exclude_pattern: tuple[str, ...],
     include_pattern: tuple[str, ...],
+    branch: str | None,
 ) -> None:
     """
     Analyze a directory or repository and create a text dump of its contents.
@@ -41,6 +43,8 @@ def main(
         A tuple of patterns to exclude during the analysis. Files matching these patterns will be ignored.
     include_pattern : tuple[str, ...]
         A tuple of patterns to include during the analysis. Only files matching these patterns will be processed.
+    branch : str | None
+        The branch to clone (optional).
 
     Raises
     ------
@@ -54,7 +58,7 @@ def main(
 
         if not output:
             output = "digest.txt"
-        summary, _, _ = ingest(source, max_size, include_patterns, exclude_patterns, output=output)
+        summary, _, _ = ingest(source, max_size, include_patterns, exclude_patterns, branch, output=output)
 
         click.echo(f"Analysis complete! Output written to: {output}")
         click.echo("\nSummary:")

src/gitingest/ingest.py

@@ -14,6 +14,7 @@ def ingest(
     max_file_size: int = 10 * 1024 * 1024,  # 10 MB
     include_patterns: list[str] | str | None = None,
     exclude_patterns: list[str] | str | None = None,
+    branch: str | None = None,
     output: str | None = None,
 ) -> tuple[str, str, str]:
     """
@@ -34,6 +35,8 @@ def ingest(
         Pattern or list of patterns specifying which files to include. If `None`, all files are included.
     exclude_patterns : list[str] | str | None, optional
         Pattern or list of patterns specifying which files to exclude. If `None`, no files are excluded.
+    branch : str | None, optional
+        The branch to clone and ingest. If `None`, the default branch is used.
     output : str | None, optional
         File path where the summary and content should be written. If `None`, the results are not written to a file.
 
@@ -65,7 +68,7 @@ def ingest(
                 url=query["url"],
                 local_path=query["local_path"],
                 commit=query.get("commit"),
-                branch=query.get("branch"),
+                branch=branch,
             )
             clone_result = clone_repo(clone_config)
 
@@ -82,7 +85,7 @@ def ingest(
 
         return summary, tree, content
     finally:
-        # Clean up the temporary directory if it was created
+        # Clean up temporary directories
         if query["url"]:
             # Clean up the temporary directory under /tmp/gitingest
             cleanup_path = "/tmp/gitingest"

src/process_query.py

@@ -21,6 +21,7 @@ async def process_query(
     slider_position: int,
     pattern_type: str = "exclude",
     pattern: str = "",
+    branch: str = None,
     is_index: bool = False,
 ) -> _TemplateResponse:
     """
@@ -41,6 +42,8 @@ async def process_query(
         Type of pattern to use, either "include" or "exclude" (default is "exclude").
     pattern : str
         Pattern to include or exclude in the query, depending on the pattern type.
+    branch : str
+        The branch to clone, by default None.
     is_index : bool
         Flag indicating whether the request is for the index page (default is False).
 
@@ -88,14 +91,13 @@ async def process_query(
             url=query["url"],
             local_path=query["local_path"],
             commit=query.get("commit"),
-            branch=query.get("branch"),
+            branch=branch or query.get("branch"),
         )
         await clone_repo(clone_config)
         summary, tree, content = ingest_from_query(query)
         with open(f"{clone_config.local_path}.txt", "w", encoding="utf-8") as f:
             f.write(tree + "\n" + content)
     except Exception as e:
-        # hack to print error message when query is not defined
         if "query" in locals() and query is not None and isinstance(query, dict):
             _print_error(query["url"], e, max_file_size, pattern_type, pattern)
         else:

src/static/js/utils.js

@@ -56,6 +56,12 @@ function handleSubmit(event, showLoading = false) {
         formData.append('pattern', pattern.value);
     }
 
+    const branch = document.getElementById('branch');
+    if (branch) {
+        formData.delete('branch');
+        formData.append('branch', branch.value);
+    }
+
     const originalContent = submitButton.innerHTML;
     const currentStars = document.getElementById('github-stars')?.textContent;
 

src/templates/components/github_form.jinja

@@ -25,6 +25,15 @@
             </div>
             <input type="hidden" name="pattern_type" value="exclude">
             <input type="hidden" name="pattern" value="">
+            <div class="relative w-full h-full">
+                <div class="w-full h-full rounded bg-gray-900 translate-y-1 translate-x-1 absolute inset-0 z-10"></div>
+                <input type="text"
+                       name="branch"
+                       id="branch"
+                       placeholder="Branch (optional)"
+                       value="{{ branch if branch else '' }}"
+                       class="border-[3px] w-full relative z-20 border-gray-900 placeholder-gray-600 text-lg font-medium focus:outline-none py-3.5 px-6 rounded">
+            </div>
         </form>
         <div class="mt-4 relative z-20 flex flex-wrap gap-4 items-start">
             <!-- Pattern selector -->

tests/test_clone.py

@@ -1,5 +1,6 @@
 """ Tests for the clone module. """
 
+import os
 from unittest.mock import AsyncMock, patch
 
 import pytest
@@ -171,3 +172,43 @@ async def test_check_repo_exists_with_redirect() -> None:
         mock_exec.return_value = mock_process
 
         assert await _check_repo_exists(url)
+
+
+@pytest.mark.asyncio
+async def test_clone_specific_branch(tmp_path):
+    repo_url = "https://github.com/cyclotruc/gitingest.git"
+    branch_name = "main"
+    local_path = tmp_path / "gitingest"
+
+    config = CloneConfig(url=repo_url, local_path=str(local_path), branch=branch_name)
+    await clone_repo(config)
+
+    # Assertions
+    assert local_path.exists(), "The repository was not cloned successfully."
+    assert local_path.is_dir(), "The cloned repository path is not a directory."
+
+    # Check the current branch
+    current_branch = os.popen(f"git -C {local_path} branch --show-current").read().strip()
+    assert current_branch == branch_name, f"Expected branch '{branch_name}', got '{current_branch}'."
+
+
+@pytest.mark.asyncio
+async def test_clone_branch_with_slashes(tmp_path):
+    repo_url = "https://github.com/cyclotruc/gitingest.git"
+    branch_name = "feat/logo"
+    local_path = tmp_path / "gitingest"
+
+    config = CloneConfig(url=repo_url, local_path=str(local_path), branch=branch_name)
+    with patch("gitingest.clone._check_repo_exists", return_value=True):
+        with patch("gitingest.clone._run_git_command", new_callable=AsyncMock) as mock_exec:
+            await clone_repo(config)
+            mock_exec.assert_called_once_with(
+                "git",
+                "clone",
+                "--depth=1",
+                "--single-branch",
+                "--branch",
+                branch_name,
+                repo_url,
+                str(local_path),
+            )