Made the wordproject import more robust

openlp/plugins/bibles/lib/importers/wordproject.py

@@ -122,11 +122,16 @@ class WordProjectBible(BibleImport):
                 log.debug('Cannot find chapters, using all p instead')
                 chapters_p = soup.p
         log.debug(chapters_p)
+        chapter_number = 0
         for item in chapters_p.contents:
             if self.stop_import_flag:
                 break
             if isinstance(item, Tag) and item.name in ['a', 'span']:
-                chapter_number = int(item.string.strip())
+                old_chapter_number = chapter_number
+                chapter_number = self.get_number(item.string.strip())
+                if chapter_number is None:
+                    # attempt to fix a broken bible input by assuming the next chapter is just an increment
+                    chapter_number = old_chapter_number + 1
                 self.set_current_chapter(db_book.name, chapter_number)
                 self.process_verses(db_book, book_id, chapter_number)
 
@@ -150,7 +155,11 @@ class WordProjectBible(BibleImport):
             if isinstance(item, Tag) and 'verse' in item.get('class', []):
                 if verse_number > 0:
                     self.process_verse(db_book, chapter_number, verse_number, verse_text.strip())
-                verse_number = int(item.string.strip())
+                old_verse_number = verse_number
+                verse_number = self.get_number(item.string.strip())
+                if verse_number is None:
+                    # attempt to fix a broken bible input by assuming the next verse is just an increment
+                    verse_number = old_verse_number + 1
                 verse_text = ''
             elif isinstance(item, NavigableString):
                 verse_text += str(item)
@@ -186,3 +195,14 @@ class WordProjectBible(BibleImport):
             result = self.process_books()
         self._cleanup()
         return result
+
+    def get_number(self, input_: str) -> None | int:
+        """
+        Given a string extracts the integer value from the beginning of the string, stopping at first non-numeric char
+        :param str: input string
+        :return: integer if found, else None
+        """
+        if result := re.findall(r'\d+', input_):
+            return int(result[0])
+        else:
+            return None

tests/openlp_plugins/bibles/test_wordprojectimport.py

@@ -21,6 +21,7 @@
 """
 This module contains tests for the WordProject Bible importer.
 """
+import pytest
 from pathlib import Path
 from unittest.mock import MagicMock, call, patch
 
@@ -232,3 +233,14 @@ def test_do_import_no_language(settings):
     assert mocked_process_books.call_count == 0
     mocked_cleanup.assert_called_once_with()
     assert result is False
+
+
+@pytest.mark.parametrize('input_, output', [
+    ('1', 1),
+    ('02.', 2),
+    ('3-4', 3),
+    ('exodus', None)
+])
+def test_get_number(input_: str, output: int, settings):
+    importer = WordProjectBible(MagicMock(), path='.', name='.', file_path='dummy.zip')
+    assert importer.get_number(input_) == output

tests/resources/bibles/wordproject_chapter.htm

@@ -172,7 +172,7 @@
 
 <!--... the Word of God:--></a>
   <p><span class="verse" id="1">1</span> In the beginning God created the heaven and the earth. 
-<br /><span class="verse" id="2">2</span> And the earth was without form, and void; and darkness was upon the face of the deep. And the Spirit of God moved upon the face of the waters. 
+<br /><span class="verse" id="2">02.</span> And the earth was without form, and void; and darkness was upon the face of the deep. And the Spirit of God moved upon the face of the waters. 
 <br /><span class="verse" id="3">3</span> And God said, Let there be light: and there was light. 
 <br /><span class="verse" id="4">4</span> And God saw the light, that it was good: and God divided the light from the darkness. 
 <br /><span class="verse" id="5">5</span> And God called the light Day, and the darkness he called Night. And the evening and the morning were the first day.