extract --continue: skip complete dirs, count skipped files in --progress, refs #6857

Directories: same_item() already considered an existing directory with the archived
mode and mtime as complete, but do_extract never passed continue_extraction when
restoring directory metadata, so all directories got their metadata restored again.

--progress: the total includes the files that --continue skips, but their size was
never added to the progress, so the percentage stayed far below the real one.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

src/borg/archive.py

@@ -1031,6 +1031,8 @@ Duration: {0.duration}
             st = os.stat(path, follow_symlinks=False)
             if continue_extraction and same_item(item, st):
                 # we already have fully extracted this file in a previous run.
+                if pi:
+                    pi.show(increase=item.get_size(), info=[remove_surrogates(item.path)])
                 if "hlid" not in item or not has_link:
                     return  # done!
                 # it is part of a group of hard links, keep the group together:

src/borg/archiver/extract_cmd.py

@@ -88,7 +88,7 @@ class ExtractMixIn:
                     while dirs and not item.path.startswith(dirs[-1].path):
                         dir_item = dirs.pop(-1)
                         try:
-                            archive.extract_item(dir_item, stdout=stdout)
+                            archive.extract_item(dir_item, stdout=stdout, continue_extraction=continue_extraction)
                         except BackupError as e:
                             self.print_warning_instance(BackupWarning(remove_surrogates(dir_item.path), e))
 
@@ -122,7 +122,7 @@ class ExtractMixIn:
                 pi.show()
                 dir_item = dirs.pop(-1)
                 try:
-                    archive.extract_item(dir_item, stdout=stdout)
+                    archive.extract_item(dir_item, stdout=stdout, continue_extraction=continue_extraction)
                 except BackupError as e:
                     self.print_warning_instance(BackupWarning(remove_surrogates(dir_item.path), e))
         for pattern in matcher.get_unmatched_include_patterns():
@@ -170,8 +170,10 @@ class ExtractMixIn:
         ``--continue`` extracts into a non-empty directory. It is made for continuing a previously
         interrupted extraction of the same archive into the same directory: an existing regular
         file that has the same type, permissions (mode), size and modification time as the
-        archived file is considered to be fully extracted already and is skipped. Everything else
-        is extracted, replacing existing files. Files that are in the directory, but not in the
+        archived file is considered to be fully extracted already and is skipped. Likewise, the
+        metadata of an existing directory that has the same permissions (mode) and modification
+        time as the archived directory is not restored again. Everything else is extracted,
+        replacing existing files. Files that are in the directory, but not in the
         archive, are left as they are. ``--continue`` is thus also needed to restore files into an
         existing directory tree. Note that a file that was damaged without a change of its size
         and modification time (e.g. by bit rot) is skipped, not replaced: remove it before

src/borg/testsuite/archiver/extract_cmd_test.py

@@ -1035,6 +1035,7 @@ def test_extract_continue(archivers, request):
         assert file3_st.st_mtime_ns == new_file3_st.st_mtime_ns  # file3 was extracted again
         # windows has a strange ctime behaviour when deleting and recreating a file
         if not is_win32:
+            assert dir1_st.st_ctime_ns == now_dir1_st.st_ctime_ns  # dir1 metadata not restored again
             assert file1_st.st_ctime_ns == now_file1_st.st_ctime_ns  # file not extracted again
             assert file2_st.st_ctime_ns != new_file2_st.st_ctime_ns  # file extracted again
             assert file3_st.st_ctime_ns != new_file3_st.st_ctime_ns  # file extracted again
@@ -1047,6 +1048,22 @@ def test_extract_continue(archivers, request):
             assert f.read() == CONTENTS3
 
 
+def test_extract_continue_progress(archivers, request, monkeypatch):
+    # --progress must count the files that --continue skips as extracted.
+    archiver = request.getfixturevalue(archivers)
+    monkeypatch.setenv("BORG_PROGRESS_FPS", "1000000")  # do not rate limit the progress output
+    cmd(archiver, "repo-create", RK_ENCRYPTION)
+    create_regular_file(archiver.input_path, "file1", size=900 * 1024)
+    create_regular_file(archiver.input_path, "file2", size=100 * 1024)
+    cmd(archiver, "create", "arch", "input/file1", "input/file2")  # extract file1 before file2
+    with changedir("output"):
+        cmd(archiver, "extract", "arch")
+        os.truncate("input/file2", 123)  # file1 is fully extracted, file2 is not
+        output = cmd(archiver, "extract", "arch", "--continue", "--progress")
+    # file2 has a single chunk, its extraction starts after the skipped 90% of the data in file1:
+    assert " 90.0% Extracting: input/file2" in output
+
+
 @requires_hardlinks
 @pytest.mark.parametrize("missing", ["first", "last"])
 def test_extract_continue_hardlinks(archivers, request, missing):