line numbers should never be 0

pydiffchecker/helper.py

@@ -3,7 +3,12 @@ from typing import Iterator
 
 
 def subprocess_readlines(cmd, cwd=None) -> Iterator[str]:
-    process = subprocess.Popen(cmd, cwd=cwd, stdout=subprocess.PIPE, text=True)
+    process = subprocess.Popen(cmd,
+                               cwd=cwd,
+                               stdout=subprocess.PIPE,
+                               text=True,
+                               encoding='utf-8',
+                               errors='replace')
 
     for line in process.stdout:
         line = line.rstrip('\n')
@@ -13,3 +18,36 @@ def subprocess_readlines(cmd, cwd=None) -> Iterator[str]:
 
     if process.returncode != 0:
         raise subprocess.CalledProcessError(process.returncode, cmd)
+
+
+def get_changed_files(since, until, diff_filter=None, cwd=None):
+    command = ['git', 'diff-tree', '-r', '-M']
+    if diff_filter:
+        command.extend([f'--diff-filter={diff_filter}'])
+    command.extend([since, until])
+
+    raw_diff = subprocess_readlines(command, cwd=cwd)
+    return [parse_raw_file_info(raw_diff_entry)
+            for raw_diff_entry in raw_diff]
+
+
+def parse_raw_file_info(raw_diff_entry):
+    diff_entry = raw_diff_entry.lstrip(':').split()
+
+    mode_src = diff_entry[0]
+    mode_dst = diff_entry[1]
+    sha1_src = diff_entry[2]
+    sha1_dst = diff_entry[3]
+    status = diff_entry[4]
+    src = diff_entry[5]
+    dst = diff_entry[6] if len(diff_entry) > 6 else src
+
+    return {
+        'mode_src': mode_src,
+        'mode_dst': mode_dst,
+        'sha1_src': sha1_src,
+        'sha1_dst': sha1_dst,
+        'status': status,
+        'src': src,
+        'dst': dst
+    }

pydiffchecker/line_shift_checker.py

@@ -1,6 +1,6 @@
 import re
 from typing import List, Dict, Tuple, Sized, Iterable, Iterator
-from .helper import subprocess_readlines
+from .helper import subprocess_readlines, get_changed_files
 
 
 class ShiftedLines(Sized, Iterable):
@@ -28,7 +28,8 @@ class ShiftedLines(Sized, Iterable):
 
 
 class LineShiftChecker:
-    DIFF_BLOCK_REGEX = r'@@ -(\d+),(\d+) \+(\d+),(\d+) @@'
+    SUBMODULE_MODE = '160000'
+    DIFF_BLOCK_REGEX = r'@@ -(\d+)(?:,\d+)? \+(\d+)(?:,\d+)? @@'
 
     def __init__(self, revision_since: str, revision_until: str) -> None:
         self.revision_since = revision_since
@@ -39,18 +40,8 @@ class LineShiftChecker:
                 for file_info in self.__get_changed_files()}
 
     def __get_changed_files(self) -> List[Dict]:
-        process_output = subprocess_readlines(['git', 'diff', '--name-status', '--diff-filter=MR',
-                                               self.revision_since, self.revision_until])
-
-        file_list = []
-        for line in process_output:
-            raw_file_info = line.split()
-            file_list.append({
-                'src': raw_file_info[1],
-                'dst': raw_file_info[2] if len(raw_file_info) > 2 else raw_file_info[1],
-            })
-
-        return file_list
+        return [file_info for file_info in get_changed_files(self.revision_since, self.revision_until, diff_filter='MR')
+                if file_info['mode_src'] != LineShiftChecker.SUBMODULE_MODE]
 
     def __get_shifted_lines_in_file(self, file_info: Dict[str, str]) -> ShiftedLines:
         process_output = subprocess_readlines(['git', 'diff',
@@ -64,8 +55,8 @@ class LineShiftChecker:
         for line in process_output:
             matches = re.search(LineShiftChecker.DIFF_BLOCK_REGEX, line)
             if matches:
-                diff_block_src_start = int(matches.group(1))
-                diff_block_dst_start = int(matches.group(3))
+                diff_block_src_start = max(int(matches.group(1)), 1)  # line number should never be 0
+                diff_block_dst_start = max(int(matches.group(2)), 1)  # line number should never be 0
 
                 # fill shifted lines between 2 diff blocks
                 for i in range(0, diff_block_src_start - src_line_index):