Merge pull request gitpython-developers#1943 from JonasScharpf/featur…

…e/add-change-type-to-commit-stats

Add type of change to files_dict of a commit

AUTHORS

@@ -54,5 +54,6 @@ Contributors are:
 -Wenhan Zhu <wzhu.cosmos _at_ gmail.com>
 -Eliah Kagan <eliah.kagan _at_ gmail.com>
 -Ethan Lin <et.repositories _at_ gmail.com>
+-Jonas Scharpf <jonas.scharpf _at_ checkmk.com>
 
 Portions derived from other open source works and are clearly marked.

git/objects/commit.py

@@ -377,15 +377,25 @@ def stats(self) -> Stats:
         :return:
             :class:`Stats`
         """
-        if not self.parents:
-            text = self.repo.git.diff_tree(self.hexsha, "--", numstat=True, no_renames=True, root=True)
-            text2 = ""
-            for line in text.splitlines()[1:]:
+
+        def process_lines(lines: List[str]) -> str:
+            text = ""
+            for file_info, line in zip(lines, lines[len(lines) // 2 :]):
+                change_type = file_info.split("\t")[0][-1]
                 (insertions, deletions, filename) = line.split("\t")
-                text2 += "%s\t%s\t%s\n" % (insertions, deletions, filename)
-            text = text2
+                text += "%s\t%s\t%s\t%s\n" % (change_type, insertions, deletions, filename)
+            return text
+
+        if not self.parents:
+            lines = self.repo.git.diff_tree(
+                self.hexsha, "--", numstat=True, no_renames=True, root=True, raw=True
+            ).splitlines()[1:]
+            text = process_lines(lines)
         else:
-            text = self.repo.git.diff(self.parents[0].hexsha, self.hexsha, "--", numstat=True, no_renames=True)
+            lines = self.repo.git.diff(
+                self.parents[0].hexsha, self.hexsha, "--", numstat=True, no_renames=True, raw=True
+            ).splitlines()
+            text = process_lines(lines)
         return Stats._list_from_string(self.repo, text)
 
     @property

git/types.py

@@ -248,6 +248,7 @@ class Files_TD(TypedDict):
     insertions: int
     deletions: int
     lines: int
+    change_type: str
 
 
 class Total_TD(TypedDict):

git/util.py

@@ -910,6 +910,7 @@ class Stats:
       deletions = number of deleted lines as int
       insertions = number of inserted lines as int
       lines = total number of lines changed as int, or deletions + insertions
+      change_type = type of change as str, A|C|D|M|R|T|U|X|B
 
     ``full-stat-dict``
 
@@ -938,7 +939,7 @@ def _list_from_string(cls, repo: "Repo", text: str) -> "Stats":
             "files": {},
         }
         for line in text.splitlines():
-            (raw_insertions, raw_deletions, filename) = line.split("\t")
+            (change_type, raw_insertions, raw_deletions, filename) = line.split("\t")
             insertions = raw_insertions != "-" and int(raw_insertions) or 0
             deletions = raw_deletions != "-" and int(raw_deletions) or 0
             hsh["total"]["insertions"] += insertions
@@ -949,6 +950,7 @@ def _list_from_string(cls, repo: "Repo", text: str) -> "Stats":
                 "insertions": insertions,
                 "deletions": deletions,
                 "lines": insertions + deletions,
+                "change_type": change_type,
             }
             hsh["files"][filename.strip()] = files_dict
         return Stats(hsh["total"], hsh["files"])

test/fixtures/diff_numstat

@@ -1,2 +1,3 @@
-29	18	a.txt
-0	5	b.txt
+M	29	18	a.txt
+M	0	5	b.txt
+A	7	0	c.txt

test/test_commit.py

@@ -135,9 +135,12 @@ def test_stats(self):
         commit = self.rorepo.commit("33ebe7acec14b25c5f84f35a664803fcab2f7781")
         stats = commit.stats
 
-        def check_entries(d):
+        def check_entries(d, has_change_type=False):
             assert isinstance(d, dict)
-            for key in ("insertions", "deletions", "lines"):
+            keys = ("insertions", "deletions", "lines")
+            if has_change_type:
+                keys += ("change_type",)
+            for key in keys:
                 assert key in d
 
         # END assertion helper
@@ -148,7 +151,7 @@ def check_entries(d):
         assert "files" in stats.total
 
         for _filepath, d in stats.files.items():
-            check_entries(d)
+            check_entries(d, True)
         # END for each stated file
 
         # Check that data is parsed properly.

test/test_stats.py

@@ -14,13 +14,19 @@ def test_list_from_string(self):
         output = fixture("diff_numstat").decode(defenc)
         stats = Stats._list_from_string(self.rorepo, output)
 
-        self.assertEqual(2, stats.total["files"])
-        self.assertEqual(52, stats.total["lines"])
-        self.assertEqual(29, stats.total["insertions"])
+        self.assertEqual(3, stats.total["files"])
+        self.assertEqual(59, stats.total["lines"])
+        self.assertEqual(36, stats.total["insertions"])
         self.assertEqual(23, stats.total["deletions"])
 
         self.assertEqual(29, stats.files["a.txt"]["insertions"])
         self.assertEqual(18, stats.files["a.txt"]["deletions"])
+        self.assertEqual("M", stats.files["a.txt"]["change_type"])
 
         self.assertEqual(0, stats.files["b.txt"]["insertions"])
         self.assertEqual(5, stats.files["b.txt"]["deletions"])
+        self.assertEqual("M", stats.files["b.txt"]["change_type"])
+
+        self.assertEqual(7, stats.files["c.txt"]["insertions"])
+        self.assertEqual(0, stats.files["c.txt"]["deletions"])
+        self.assertEqual("A", stats.files["c.txt"]["change_type"])