feat: smart chunking to avoid chunk boundaries mid-sentence

1 files changed, 18 insertions, 0 deletions

diff --git a/app/tests/test_converter.py b/app/tests/test_converter.py
index 85ae6ca..9084898 100644
--- a/app/tests/test_converter.py
+++ b/app/tests/test_converter.py
@@ -585,6 +585,24 @@ class ChunkClampPromptTests(unittest.TestCase):
         with patch.object(config, "CHUNK_SIZE", 80):
             self.assertFalse(chunk_clamp_needed(self._entry()))
 
+    def test_per_run_clamp_caps_the_chapter_chunks(self):
+        # The popup's clamp caps the chapter chunks themselves, so each
+        # progress unit and debug dump spans one generation request.
+        converter = AudiobookConverter.__new__(AudiobookConverter)
+        converter.chunk_size = 4
+        with patch.object(config, "CHUNK_SIZE", 250):
+            chunks = converter._chapter_chunks(" ".join(f"Sw{i} ." for i in range(12)))
+        self.assertGreater(len(chunks), 1)
+        self.assertTrue(all(len(c.split()) <= 4 for c in chunks))
+
+    def test_chapter_chunks_follow_chunk_size_without_a_clamp(self):
+        converter = AudiobookConverter.__new__(AudiobookConverter)
+        self.assertIsNone(converter.chunk_size)
+        with patch.object(config, "CHUNK_SIZE", 8):
+            chunks = converter._chapter_chunks(" ".join(f"Sw{i} ." for i in range(20)))
+        self.assertGreater(len(chunks), 1)
+        self.assertTrue(all(len(c.split()) <= 8 for c in chunks))
+
     def test_prompt_answers(self):
         entry = self._entry()
         with patch("builtins.input", return_value=""):