fix: improve git file discovery with untracked and removed files (#1650)

* fix: improve git file discovery with untracked and removed files (#1647)

Signed-off-by: Xabier Napal <xabiernapal@pm.me>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* test: update git ls-files assertions (#1647)

Signed-off-by: Xabier Napal <xabiernapal@pm.me>

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Sorin Sbarnea <ssbarnea@redhat.com>

src/ansiblelint/file_utils.py

@@ -206,14 +206,31 @@ def __repr__(self) -> str:
 def discover_lintables(options: Namespace) -> Dict[str, Any]:
     """Find all files that we know how to lint."""
     # git is preferred as it also considers .gitignore
-    git_command = ['git', 'ls-files', '-z']
+    git_command_present = [
+        'git',
+        'ls-files',
+        '--cached',
+        '--others',
+        '--exclude-standard',
+        '-z',
+    ]
+    git_command_absent = ['git', 'ls-files', '--deleted', '-z']
     out = None
 
     try:
-        out = subprocess.check_output(
-            git_command, stderr=subprocess.STDOUT, universal_newlines=True
+        out_present = subprocess.check_output(
+            git_command_present, stderr=subprocess.STDOUT, universal_newlines=True
         ).split("\x00")[:-1]
-        _logger.info("Discovered files to lint using: %s", ' '.join(git_command))
+        _logger.info(
+            "Discovered files to lint using: %s", ' '.join(git_command_present)
+        )
+
+        out_absent = subprocess.check_output(
+            git_command_absent, stderr=subprocess.STDOUT, universal_newlines=True
+        ).split("\x00")[:-1]
+        _logger.info("Excluded removed files using: %s", ' '.join(git_command_absent))
+
+        out = set(out_present) - set(out_absent)
     except subprocess.CalledProcessError as exc:
         if not (exc.returncode == 128 and 'fatal: not a git repository' in exc.output):
             _logger.warning(

test/TestUtils.py

@@ -354,7 +354,11 @@ def test_cli_auto_detect(capfd: CaptureFixture[str]) -> None:
     out, err = capfd.readouterr()
 
     # Confirmation that it runs in auto-detect mode
-    assert "Discovered files to lint using: git ls-files -z" in err
+    assert (
+        "Discovered files to lint using: git ls-files --cached --others --exclude-standard -z"
+        in err
+    )
+    assert "Excluded removed files using: git ls-files --deleted -z" in err
     # An expected rule match from our examples
     assert (
         "examples/playbooks/empty_playbook.yml:1: "