Add --diff option to show diff when running with --check

Author: tdgroot

src/mdformat/_cli.py

@@ -3,6 +3,7 @@
 import argparse
 from collections.abc import Callable, Generator, Iterable, Mapping, Sequence
 import contextlib
+from datetime import datetime
 import itertools
 import logging
 from pathlib import Path
@@ -17,6 +18,7 @@
 from mdformat._util import atomic_write, detect_newline_type, is_md_equal
 import mdformat.plugins
 import mdformat.renderer
+from src.mdformat._output import diff
 
 # Match "\r" and "\n" characters that are not part of a "\r\n" sequence
 RE_NON_CRLF_LINE_END = re.compile(r"(?:[^\r]|^)\n|\r(?:[^\n]|\Z)")
@@ -92,6 +94,17 @@ def run(cli_args: Sequence[str]) -> int:  # noqa: C901
             ):
                 format_errors_found = True
                 print_error(f'File "{path_str}" is not formatted.')
+
+                if opts["diff"]:
+                    then = datetime.utcfromtimestamp(path.stat().st_mtime)
+                    now = datetime.utcnow()
+                    src_name = f"{path}\t{then} +0000"
+                    dst_name = f"{path}\t{now} +0000"
+
+                    diff_contents = diff(
+                        original_str, formatted_str, src_name, dst_name
+                    )
+                    print(diff_contents)
         else:
             if not changes_ast and not is_md_equal(
                 original_str,
@@ -146,6 +159,11 @@ def make_arg_parser(
     parser.add_argument(
         "--check", action="store_true", help="do not apply changes to files"
     )
+    parser.add_argument(
+        "--diff",
+        action="store_true",
+        help="show diff of what would be changed when running with --check",
+    )
     version_str = f"mdformat {mdformat.__version__}"
     if plugin_versions_str:
         version_str += f" ({plugin_versions_str})"

src/mdformat/_output.py

@@ -0,0 +1,22 @@
+import difflib
+
+
+def diff(a: str, b: str, a_name: str, b_name: str) -> str:
+    """Return a unified diff string between strings `a` and `b`.
+
+    Highly inspired by Black's diff function.
+    """
+    a_lines = a.splitlines(keepends=True)
+    b_lines = b.splitlines(keepends=True)
+
+    diff_lines = []
+    for line in difflib.unified_diff(
+        a_lines, b_lines, fromfile=a_name, tofile=b_name, n=5
+    ):
+        if line[-1] == "\n":
+            diff_lines.append(line)
+        else:
+            diff_lines.append(line + "\n")
+            diff_lines.append("\\ No newline at end of file\n")
+
+    return "".join(diff_lines)

tests/test_cli.py

@@ -82,6 +82,21 @@ def test_check__fail(tmp_path):
     assert run((str(file_path), "--check")) == 1
 
 
+def test_check_fail_diff(capsys, tmp_path):
+    """Test for --check flag and --diff flag combined on unformatted files.
+
+    Test that when an unformatted file fails, a diff is writtin to
+    stdout.
+    """
+
+    file_path = tmp_path / "test_markdown.md"
+    file_path.write_text(UNFORMATTED_MARKDOWN)
+    assert run((str(file_path), "--check", "--diff")) == 1
+    captured = capsys.readouterr()
+    assert str(file_path) in captured.out
+    assert "-\n-\n # A header\n-\n" in captured.out
+
+
 def test_check__multi_fail(capsys, tmp_path):
     """Test for --check flag when multiple files are unformatted.