From 93b9783f1addeac45539ab958171d7dca5e7037d Mon Sep 17 00:00:00 2001
From: Guillaume Leurquin <guillaume.g.leurquin.external@airbus.com>
Date: Wed, 14 Jun 2023 12:36:11 +0000
Subject: [PATCH] feat: add include and exclude cli options

---
 docs/index.md                  |  4 ++++
 src/yamlfix/entrypoints/cli.py | 41 +++++++++++++++++++++++++++++-----
 tests/e2e/test_cli.py          | 28 +++++++++++++++++++++++
 3 files changed, 68 insertions(+), 5 deletions(-)

diff --git a/docs/index.md b/docs/index.md
index 3ba9813..734bcbe 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -160,6 +160,10 @@ These provided arguments and environment variables would result in a merged runt
 none_representation = "~"
 ```
 
+## Configure include and exclude files
+
+Per default `yamlfix`, when run through cli, will include all `*.yaml` and `*.yml` files from the directories passed via the CLI. With `--exclude <glob>` and `--include <glob>` you can include or exclude specific files within those directories.
+
 ## Configuration Options
 
 All fields configured in the [YamlfixConfig class](./reference/#yamlfix.model.YamlfixConfig) can be provided through the means mentioned in [Configuration](#configuration). Here are the currently available configuration options with short examples on their impact to provided `yaml`-files.
diff --git a/src/yamlfix/entrypoints/cli.py b/src/yamlfix/entrypoints/cli.py
index d0158fd..38f447a 100644
--- a/src/yamlfix/entrypoints/cli.py
+++ b/src/yamlfix/entrypoints/cli.py
@@ -17,9 +17,20 @@
 log = logging.getLogger(__name__)
 
 
-def _find_all_yaml_files(dir_: Path) -> List[Path]:
-    files = [dir_.rglob(f"*.{ext}") for ext in ["yml", "yaml"]]
-    return [file for list_ in files for file in list_]
+def _matches_any_glob(file_to_test: Path, globs: Optional[List[str]]) -> bool:
+    return any(file_to_test.match(glob) for glob in (globs or []))
+
+
+def _find_all_yaml_files(
+    dir_: Path, include_globs: Optional[List[str]], exclude_globs: Optional[List[str]]
+) -> List[Path]:
+    files = [dir_.rglob(glob) for glob in (include_globs or [])]
+    return [
+        file
+        for list_ in files
+        for file in list_
+        if not _matches_any_glob(file, exclude_globs)
+    ]
 
 
 @click.command()
@@ -43,12 +54,32 @@ def _find_all_yaml_files(dir_: Path) -> List[Path]:
     default="YAMLFIX",
     help="Read yamlfix relevant environment variables starting with this prefix.",
 )
+@click.option(
+    "--exclude",
+    "-e",
+    multiple=True,
+    type=str,
+    help="Files matching this glob pattern will be ignored.",
+)
+@click.option(
+    "--include",
+    "-i",
+    multiple=True,
+    type=str,
+    default=["*.yaml", "*.yml"],
+    help=(
+        "Files matching this glob pattern will be included, "
+        "unless they are also excluded. Default to '*.yaml' and '*.yml'."
+    ),
+)
 @click.argument("files", type=str, required=True, nargs=-1)
-def cli(
+def cli(  # pylint: disable=too-many-arguments
     files: Tuple[str],
     verbose: bool,
     check: bool,
     config_file: Optional[List[str]],
+    include: Optional[List[str]],
+    exclude: Optional[List[str]],
     env_prefix: str,
 ) -> None:
     """Corrects the source code of the specified files.
@@ -67,7 +98,7 @@ def cli(
         real_files = []
         for provided_file in paths:
             if provided_file.is_dir():
-                real_files.extend(_find_all_yaml_files(provided_file))
+                real_files.extend(_find_all_yaml_files(provided_file, include, exclude))
             else:
                 real_files.append(provided_file)
         files_to_fix = [file.open("r+") for file in real_files]
diff --git a/tests/e2e/test_cli.py b/tests/e2e/test_cli.py
index 36e0e56..2db1e70 100644
--- a/tests/e2e/test_cli.py
+++ b/tests/e2e/test_cli.py
@@ -86,6 +86,34 @@ def test_corrects_code_from_stdin(runner: CliRunner) -> None:
     assert result.stdout == fixed_source
 
 
+def test_include_exclude_files(runner: CliRunner, tmp_path: Path) -> None:
+    """Correct only files matching include, and ignore files matching exclude."""
+    include1 = tmp_path / "source_1.yaml"
+    exclude1 = tmp_path / "source_2.txt"
+    (tmp_path / "foo").mkdir()
+    exclude2 = tmp_path / "foo" / "source_3.yaml"
+    test_files = [include1, exclude1, exclude2]
+    init_source = "program: yamlfix"
+    for test_file in test_files:
+        test_file.write_text(init_source)
+    fixed_source = dedent(
+        """\
+        ---
+        program: yamlfix
+        """
+    )
+
+    result = runner.invoke(
+        cli,
+        [str(tmp_path)] + ["--include", "*.yaml", "--exclude", "foo/*.yaml"],
+    )
+
+    assert result.exit_code == 0
+    assert include1.read_text() == fixed_source
+    assert exclude1.read_text() == init_source
+    assert exclude2.read_text() == init_source
+
+
 @pytest.mark.secondary()
 @pytest.mark.parametrize(
     ("verbose", "requires_fixing"), product([0, 1, 2], [True, False])