fix(extract): keep AST progress denominator consistent to the end (#1693)

Intermediate progress lines count against len(uncached_work) ("X/Y uncached
files"), but the final line switched to total_files ("Y/Y files"), which includes
cached hits and files with no extractor that never entered uncached_work. On a
large corpus with unsupported-language files, the total jumped upward right after
99% with no explanation. Both the parallel and sequential final lines now report
the same uncached_work denominator, so the count no longer appears to change
mid-run.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

graphify/extract.py

@@ -16254,8 +16254,13 @@ def _extract_parallel(
         )
         return False
     if total_files >= _PROGRESS_INTERVAL:
+        # Report the same denominator the intermediate lines used (uncached files
+        # actually processed this run), not total_files — switching to the full
+        # corpus made the count jump upward at the end (cached hits + files with no
+        # extractor never entered uncached_work), which read as inconsistent (#1693).
+        _done = len(uncached_work)
         print(
-            f"  AST extraction: {total_files}/{total_files} files (100%) [{max_workers} workers]",
+            f"  AST extraction: {_done}/{_done} uncached files (100%) [{max_workers} workers]",
             flush=True,
         )
     return True
@@ -16290,7 +16295,9 @@ def _extract_sequential(
             save_cached(path, result, effective_root)
         per_file[idx] = result
     if total_files >= _PROGRESS_INTERVAL:
-        print(f"  AST extraction: {total_files}/{total_files} files (100%)", flush=True)
+        # Consistent denominator with the intermediate lines (#1693).
+        _done = len(uncached_work)
+        print(f"  AST extraction: {_done}/{_done} uncached files (100%)", flush=True)
 
 
 _PARALLEL_THRESHOLD = 20

tests/test_extract.py

@@ -1795,3 +1795,21 @@ def test_extract_no_warning_when_all_code_has_extractors(tmp_path, capsys):
     extract([py], cache_root=tmp_path)
     err = capsys.readouterr().err
     assert "no AST extractor" not in err
+
+
+def test_extract_progress_final_line_uses_consistent_denominator(tmp_path, capsys):
+    # #1693: intermediate progress lines count against uncached_work; the final
+    # "100%" line must NOT switch to total_files (which includes cached hits and
+    # files with no extractor), or the count appears to jump upward at the end.
+    for i in range(100):
+        (tmp_path / f"m{i}.py").write_text(f"def f{i}():\n    return {i}\n")
+    for i in range(5):
+        (tmp_path / f"s{i}.r").write_text(f"g{i} <- function(x) x\n")  # no extractor
+    paths = sorted(tmp_path.glob("*.py")) + sorted(tmp_path.glob("*.r"))  # total 105
+
+    extract(paths, cache_root=tmp_path, parallel=False)
+    out = capsys.readouterr().out
+
+    # final progress line reports the uncached count (100), not the total (105)
+    assert "100/100 uncached files (100%)" in out
+    assert "105/105 files" not in out, "final line must not switch to total_files (#1693)"