feat: add support for roman numeral page numbers in TOC

pageindex/utils.py

@@ -691,14 +691,43 @@ def convert_physical_index_to_int(data):
     return data
 
 
+_ROMAN_REGEX = re.compile(r"^m{0,4}(cm|cd|d?c{0,3})(xc|xl|l?x{0,3})(ix|iv|v?i{0,3})$", re.IGNORECASE)
+_ROMAN_VALUES = {"i": 1, "v": 5, "x": 10, "l": 50, "c": 100, "d": 500, "m": 1000}
+
+
+def roman_to_int(s: str) -> int | None:
+    """Convert a Roman numeral string (e.g. 'iv', 'xii', 'IV') to an integer.
+    Returns None if s is not a valid Roman numeral."""
+    if not isinstance(s, str):
+        return None
+    cleaned = s.strip()
+    if not cleaned or not _ROMAN_REGEX.match(cleaned):
+        return None
+    total = 0
+    prev = 0
+    for ch in reversed(cleaned.lower()):
+        val = _ROMAN_VALUES.get(ch, 0)
+        if val < prev:
+            total -= val
+        else:
+            total += val
+        prev = val
+    return total if total > 0 else None
+
+
 def convert_page_to_int(data):
     for item in data:
-        if 'page' in item and isinstance(item['page'], str):
-            try:
-                item['page'] = int(item['page'])
-            except ValueError:
-                # Keep original value if conversion fails
-                pass
+        if 'page' in item and item['page'] is not None:
+            if isinstance(item['page'], int):
+                continue
+            if isinstance(item['page'], str):
+                cleaned = item['page'].strip()
+                try:
+                    item['page'] = int(cleaned)
+                except ValueError:
+                    roman_val = roman_to_int(cleaned)
+                    if roman_val is not None:
+                        item['page'] = roman_val
     return data
 
 

tests/test_roman_numerals.py

@@ -0,0 +1,87 @@
+import pytest
+from pageindex.utils import roman_to_int, convert_page_to_int
+
+
+@pytest.mark.parametrize(
+    "roman_str,expected",
+    [
+        ("i", 1),
+        ("ii", 2),
+        ("iii", 3),
+        ("iv", 4),
+        ("v", 5),
+        ("vi", 6),
+        ("vii", 7),
+        ("viii", 8),
+        ("ix", 9),
+        ("x", 10),
+        ("xi", 11),
+        ("xii", 12),
+        ("xiv", 14),
+        ("xv", 15),
+        ("xix", 19),
+        ("xx", 20),
+        ("xl", 40),
+        ("l", 50),
+        ("xc", 90),
+        ("c", 100),
+        ("cd", 400),
+        ("d", 500),
+        ("cm", 900),
+        ("m", 1000),
+        ("MCMLIV", 1954),
+        ("mmxxvi", 2026),
+        # Mixed casing and whitespace
+        ("  IV  ", 4),
+        ("Xii", 12),
+        ("vIi", 7),
+    ],
+)
+def test_roman_to_int_valid(roman_str, expected):
+    assert roman_to_int(roman_str) == expected
+
+
+@pytest.mark.parametrize(
+    "invalid_str",
+    [
+        "",
+        "   ",
+        "abc",
+        "123",
+        "iv2",
+        "iiii",  # invalid roman syntax
+        "vx",    # invalid subtraction
+        "ll",    # invalid repeated 50
+        None,
+        123,
+        [],
+    ],
+)
+def test_roman_to_int_invalid(invalid_str):
+    assert roman_to_int(invalid_str) is None
+
+
+def test_convert_page_to_int_handles_roman_and_arabic_pages():
+    toc = [
+        {"title": "Title Page", "page": "i"},
+        {"title": "Dedication", "page": "ii"},
+        {"title": "Table of Contents", "page": " iv "},
+        {"title": "Preface", "page": "VII"},
+        {"title": "Introduction", "page": "1"},
+        {"title": "Chapter 1", "page": 5},
+        {"title": "Chapter 2", "page": " 12 "},
+        {"title": "Appendix", "page": None},
+        {"title": "Unnumbered Note", "page": "not_a_page"},
+    ]
+
+    converted = convert_page_to_int(toc)
+
+    assert converted[0]["page"] == 1
+    assert converted[1]["page"] == 2
+    assert converted[2]["page"] == 4
+    assert converted[3]["page"] == 7
+    assert converted[4]["page"] == 1
+    assert converted[5]["page"] == 5
+    assert converted[6]["page"] == 12
+    assert converted[7]["page"] is None
+    assert converted[8]["page"] == "not_a_page"