1 files changed, 80 insertions, 0 deletions
diff --git a/app/tests/test_extractors.py b/app/tests/test_extractors.py
index 64666ba..619fe5a 100644
--- a/app/tests/test_extractors.py
+++ b/app/tests/test_extractors.py
@@ -53,6 +53,86 @@ class TxtExtractionTests(unittest.TestCase):
             with self.assertRaises(ValueError):
                 extract_text(path)
 
+    def test_utf16_without_bom_detected(self):
+        self.assertEqual(self._extract("chapter one".encode("utf-16-le")),
+                         "chapter one")
+
+    def test_lone_nul_does_not_flip_to_utf16(self):
+        # A single stray NUL byte in an otherwise-ASCII UTF-8 file must not
+        # switch the whole book to a UTF-16 decode (mojibake): the text
+        # comes back readable instead.
+        self.assertEqual(self._extract(b"hello world\x00rest"),
+                         "hello world\x00rest")
+
+    def test_standalone_page_numbers_removed_but_years_kept(self):
+        from converter.extractors import clean_text
+
+        cleaned = clean_text("Chapter 1\n\n42\n\nIt was 1984.")
+        self.assertNotIn("42", cleaned)
+        self.assertIn("1984", cleaned)
+        kept = clean_text("It was the year\n\n1984\n\nwhen it began.")
+        self.assertIn("1984", kept)
+
+
+class EpubZipfileFallbackTests(unittest.TestCase):
+    """The no-ebooklib EPUB fallback: spine order, no TOC narration."""
+
+    @staticmethod
+    def _write_epub(path: Path):
+        import zipfile
+
+        container = ("<?xml version=\"1.0\"?>"
+                     "<container><rootfiles>"
+                     "<rootfile full-path=\"OEBPS/content.opf\"/>"
+                     "</rootfiles></container>")
+        opf = ("<?xml version=\"1.0\"?>"
+               "<package xmlns=\"http://www.idpf.org/2007/opf\">"
+               "<manifest>"
+               "<item id=\"nav\" href=\"nav.xhtml\" properties=\"nav\"/>"
+               "<item id=\"c2\" href=\"text/chapterB.xhtml\"/>"
+               "<item id=\"c1\" href=\"text/chapterA.xhtml\"/>"
+               "</manifest>"
+               "<spine><itemref idref=\"nav\"/>"
+               "<itemref idref=\"c2\"/><itemref idref=\"c1\"/></spine>"
+               "</package>")
+        with zipfile.ZipFile(path, "w") as zf:
+            zf.writestr("mimetype", "application/epub+zip")
+            zf.writestr("META-INF/container.xml", container)
+            zf.writestr("OEBPS/content.opf", opf)
+            zf.writestr("OEBPS/nav.xhtml",
+                        "<html><body><p>Contents</p></body></html>")
+            zf.writestr("OEBPS/text/chapterA.xhtml",
+                        "<html><body><p>Alpha text.</p></body></html>")
+            zf.writestr("OEBPS/text/chapterB.xhtml",
+                        "<html><body><p>Beta text.</p></body></html>")
+
+    def test_spine_order_and_no_nav(self):
+        from converter.extractors import _read_epub_zipfile
+
+        with tempfile.TemporaryDirectory() as tmp:
+            path = Path(tmp) / "book.epub"
+            self._write_epub(path)
+            items = _read_epub_zipfile(path)
+
+        titles = [title for title, _ in items]
+        self.assertNotIn("nav", titles)
+        # Spine order (B before A) beats filename sort (A before B).
+        self.assertEqual(titles, ["chapterB", "chapterA"])
+
+    def test_unparsable_opf_falls_back_to_filename_order(self):
+        import zipfile
+
+        from converter.extractors import _read_epub_zipfile
+
+        with tempfile.TemporaryDirectory() as tmp:
+            path = Path(tmp) / "book.epub"
+            with zipfile.ZipFile(path, "w") as zf:
+                zf.writestr("a.xhtml", "<html><body><p>A</p></body></html>")
+                zf.writestr("b.xhtml", "<html><body><p>B</p></body></html>")
+            items = _read_epub_zipfile(path)
+
+        self.assertEqual([title for title, _ in items], ["a", "b"])
+
 
 def _build_test_epub(path: Path, chapters=(("One", "First chapter text."),
                                             ("Two", "Second chapter text."))) -> None: