Include column number in pygrep output

pre_commit/languages/pygrep.py

@@ -17,13 +17,28 @@ healthy = helpers.basic_healthy
 install_environment = helpers.no_install
 
 
+def _column_marker(line: bytes, byte_offset: Optional[int]) -> str:
+    try:
+        char_offset = len(line[0:byte_offset].decode())
+    except UnicodeEncodeError:
+        retv = ''
+    else:
+        retv = f'{char_offset}:'
+
+    return retv
+
+
 def _process_filename_by_line(pattern: Pattern[bytes], filename: str) -> int:
     retv = 0
     with open(filename, 'rb') as f:
         for line_no, line in enumerate(f, start=1):
-            if pattern.search(line):
+            match = pattern.search(line)
+            if match:
                 retv = 1
-                output.write(f'{filename}:{line_no}:')
+                output.write(
+                    f'{filename}:{line_no}:' +
+                    _column_marker(line, match.start()),
+                )
                 output.write_line_b(line.rstrip(b'\r\n'))
     return retv
 
@@ -36,7 +51,11 @@ def _process_filename_at_once(pattern: Pattern[bytes], filename: str) -> int:
         if match:
             retv = 1
             line_no = contents[:match.start()].count(b'\n')
-            output.write(f'{filename}:{line_no + 1}:')
+            line_start = contents.rfind(b'\n', 0, match.start()) + 1
+            output.write(
+                f'{filename}:{line_no + 1}:' +
+                _column_marker(contents[line_start:match.start()], None),
+            )
 
             matched_lines = match[0].split(b'\n')
             matched_lines[0] = contents.split(b'\n')[line_no]

tests/languages/pygrep_test.py

@@ -17,10 +17,10 @@ def some_files(tmpdir):
     ('pattern', 'expected_retcode', 'expected_out'),
     (
         ('baz', 0, ''),
-        ('foo', 1, 'f1:1:foo\n'),
-        ('bar', 1, 'f1:2:bar\n'),
-        (r'(?i)\[info\]', 1, 'f2:1:[INFO] hi\n'),
-        ("h'q", 1, "f3:1:with'quotes\n"),
+        ('foo', 1, 'f1:1:0:foo\n'),
+        ('bar', 1, 'f1:2:0:bar\n'),
+        (r'(?i)\[info\]', 1, 'f2:1:0:[INFO] hi\n'),
+        ("h'q", 1, "f3:1:3:with'quotes\n"),
     ),
 )
 def test_main(some_files, cap_out, pattern, expected_retcode, expected_out):
@@ -34,32 +34,32 @@ def test_ignore_case(some_files, cap_out):
     ret = pygrep.main(('--ignore-case', 'info', 'f1', 'f2', 'f3'))
     out = cap_out.get()
     assert ret == 1
-    assert out == 'f2:1:[INFO] hi\n'
+    assert out == 'f2:1:1:[INFO] hi\n'
 
 
 def test_multiline(some_files, cap_out):
     ret = pygrep.main(('--multiline', r'foo\nbar', 'f1', 'f2', 'f3'))
     out = cap_out.get()
     assert ret == 1
-    assert out == 'f1:1:foo\nbar\n'
+    assert out == 'f1:1:0:foo\nbar\n'
 
 
 def test_multiline_line_number(some_files, cap_out):
     ret = pygrep.main(('--multiline', r'ar', 'f1', 'f2', 'f3'))
     out = cap_out.get()
     assert ret == 1
-    assert out == 'f1:2:bar\n'
+    assert out == 'f1:2:1:bar\n'
 
 
 def test_multiline_dotall_flag_is_enabled(some_files, cap_out):
     ret = pygrep.main(('--multiline', r'o.*bar', 'f1', 'f2', 'f3'))
     out = cap_out.get()
     assert ret == 1
-    assert out == 'f1:1:foo\nbar\n'
+    assert out == 'f1:1:1:foo\nbar\n'
 
 
 def test_multiline_multiline_flag_is_enabled(some_files, cap_out):
     ret = pygrep.main(('--multiline', r'foo$.*bar', 'f1', 'f2', 'f3'))
     out = cap_out.get()
     assert ret == 1
-    assert out == 'f1:1:foo\nbar\n'
+    assert out == 'f1:1:0:foo\nbar\n'