scripturelookup 0.1.1__tar.gz → 0.1.2__tar.gz

{scripturelookup-0.1.1/src/scripturelookup.egg-info → scripturelookup-0.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scripturelookup
-Version: 0.1.1
+Version: 0.1.2
 Summary: Python and command-line utility for converting scripture references between formats.
 Author-email: Samuel Bradshaw <samuel.h.bradshaw@gmail.com>
 License: MIT License

{scripturelookup-0.1.1 → scripturelookup-0.1.2}/src/scripturelookup/data.py

@@ -144,11 +144,15 @@ highest_book_number = len(roman_numeral_words)
 ordinal_forms_cache = {}
 def get_ordinal_forms(lang):
   if lang not in ordinal_forms_cache:
-    ordinal_forms_cache[lang] = {
-      form.lower(): index + 1
-      for index, group in enumerate(ordinal_words.get(lang, []))
-      for form in group
-    }
+    forms = {}
+    for index, group in enumerate(ordinal_words.get(lang, [])):
+      for form in group:
+        forms[form.lower()] = index + 1
+        # The ordinal indicators are hard to type on many keyboards, so "1º" is also accepted as "1o" and
+        # "1ª" as "1a". They're derived rather than listed, so a language only spells out what's linguistic.
+        plain_form = form.lower().replace('º', 'o').replace('ª', 'a')
+        forms.setdefault(plain_form, index + 1)
+    ordinal_forms_cache[lang] = forms
   return ordinal_forms_cache[lang]
 
 # Get a map of number form to the number it stands for, for the prefix on a numbered book name. It's the
@@ -163,6 +167,22 @@ def get_book_number_forms(lang):
     book_number_forms_cache[lang] = forms
   return book_number_forms_cache[lang]
 
+# Get the maps above keyed for comparison, so the number a matched form stands for can be looked up however
+# the form was spelled. The pattern built from these forms matches text the map has no key for: a character
+# class matches "Decimo" where the map holds "décimo", and whitespace is matched loosely, so "décimo
+# primero" can arrive with a non-breaking space between the words.
+normalized_ordinal_forms_cache = {}
+def get_normalized_ordinal_forms(lang):
+  if lang not in normalized_ordinal_forms_cache:
+    normalized_ordinal_forms_cache[lang] = {normalize_for_compare(form): number for form, number in get_ordinal_forms(lang).items()}
+  return normalized_ordinal_forms_cache[lang]
+
+normalized_book_number_forms_cache = {}
+def get_normalized_book_number_forms(lang):
+  if lang not in normalized_book_number_forms_cache:
+    normalized_book_number_forms_cache[lang] = {normalize_for_compare(form): number for form, number in get_book_number_forms(lang).items()}
+  return normalized_book_number_forms_cache[lang]
+
 # List conjunctions from every language, split into the ones written as a word ("and", "y", "et") and the ones written as a symbol ("&")
 all_list_conjunctions = sorted({c for words in reference_words.values() for c in words['list_conjunctions']})
 list_conjunction_words = [c for c in all_list_conjunctions if any(char.isalpha() for char in c)]

{scripturelookup-0.1.1 → scripturelookup-0.1.2}/src/scripturelookup/lookup.py

@@ -201,7 +201,8 @@ def get_numbered_book_patterns(lang):
       numbered_book_pattern_cache[lang] = {
         'fragment': r'|'.join(rf'(?:{form_pattern})\s+(?:{stem_pattern})' for form_pattern, stem_pattern in branches),
         'rewrite': re.compile(rf'{patterns.reference_start_boundary_pattern}({rewrite_alternation})\s+', flags=re.IGNORECASE),
-        'number_by_form': number_by_form,
+        # Keyed for comparison, since the patterns above match spellings the plain forms don't cover
+        'number_by_form': data.get_normalized_book_number_forms(lang),
       }
   return numbered_book_pattern_cache[lang]
 
@@ -239,7 +240,8 @@ def get_ordinal_patterns(lang):
       ordinal_run = rf'(?:{ordinals})(?:(?:{joiner_alternation})(?:{ordinals}))*'
 
       ordinal_pattern_cache[lang] = {
-        'number_by_form': number_by_form,
+        # Keyed for comparison, since the patterns above match spellings the plain forms don't cover
+        'number_by_form': data.get_normalized_ordinal_forms(lang),
         # Spliced into the detection pattern as an anchor that needs no number after it – the ordinal is the number
         'fragment': rf'(?:{ordinal_run})\s+(?:{books})',
         'rewrite': re.compile(rf'{patterns.reference_start_boundary_pattern}({ordinal_run})\s+({books}){patterns.not_followed_by_letter_pattern}', flags=re.IGNORECASE),
@@ -263,7 +265,7 @@ def replace_ordinal_references(input_string, lang):
 
   def replace(match):
     ordinal_run, book_name = match.group(1), match.group(2)
-    ordinal_numbers = [ordinal_patterns['number_by_form'][ordinal.group(0).lower()] for ordinal in ordinal_patterns['single_ordinal'].finditer(ordinal_run)]
+    ordinal_numbers = [ordinal_patterns['number_by_form'][data.normalize_for_compare(ordinal.group(0))] for ordinal in ordinal_patterns['single_ordinal'].finditer(ordinal_run)]
     is_range = len(ordinal_numbers) > 1 and ordinal_patterns['range_conjunction'] and ordinal_patterns['range_conjunction'].search(ordinal_run)
     numbers_string = f'{ordinal_numbers[0]}-{ordinal_numbers[-1]}' if is_range else ','.join(str(number) for number in ordinal_numbers)
     return f'{book_name} {numbers_string}'
@@ -278,7 +280,7 @@ def replace_book_number_prefixes(input_string, lang):
     return input_string
 
   def replace(match):
-    return str(numbered_book_patterns['number_by_form'][match.group(1).lower()]) + ' '
+    return str(numbered_book_patterns['number_by_form'][data.normalize_for_compare(match.group(1))]) + ' '
   return numbered_book_patterns['rewrite'].sub(replace, input_string)
 
 
@@ -507,6 +509,8 @@ def parse_verses_string(verses_string, lang = 'en', range_split_limit = 1):
   verses_string = (verses_string or '').replace('p', '').strip()
   if not verses_string:
     return None
+  # A footnote letter marks a study note, not part of the verse, so it's dropped. Example: "7a" –> "7"
+  verses_string = patterns.verse_footnote_letter.sub('', verses_string)
   
   unique_verses = set()
   all_verses_are_integers = True
@@ -732,7 +736,9 @@ def parse_references_string(input_string, lang = 'en', sort_by = None, skip_clea
         unparsed = input_string
         verses_string = ''
       verses_string, remaining_context_verses_string = (patterns.opening_parenthesis.split(patterns.closing_parenthesis.sub('', verses_string)) + [''])[:2]
-      if context_verses_string is None:
+      # Only numbers count as context verses, so stray words after an unclosed parenthesis are dropped. Example: "Matthew 7:1–2 (see" –> "Matthew 7:1–2"
+      remaining_context_verses_string = remaining_context_verses_string.strip()
+      if context_verses_string is None and patterns.numbers_and_separators.match(remaining_context_verses_string):
         context_verses_string = remaining_context_verses_string
 
       # Get chapter string and book string

{scripturelookup-0.1.1 → scripturelookup-0.1.2}/src/scripturelookup/patterns.py

@@ -56,6 +56,11 @@ chapter_range = re.compile(rf'\d+(?:{verse_range_separators_pattern})\d+')
 number_continuation_pattern = rf'\d|\s|{chapter_verse_separators_pattern}|{verse_range_separators_pattern}|{verse_group_separators_pattern}'
 # A string made up entirely of numbers and separators. Example: "3, 5–7"
 numbers_and_separators = re.compile(rf'^\d(?:{number_continuation_pattern})*$')
+# A footnote letter after a verse number. Example: "6:7a" –> "6:7"
+# Only parsing strips it – the detection patterns stop at the digit, so a detected span covers "Alma 32:21"
+# rather than "Alma 32:21a". Letting detection take a trailing letter would swallow the Spanish and
+# Portuguese range conjunction, as in "Alma 32:21 a 23".
+verse_footnote_letter = re.compile(rf'(?<=\d)[a-zA-Z]{not_followed_by_letter_pattern}')
 # The chapter and verses at the end of a reference, so the book name can be split off the front
 trailing_chapter = re.compile(rf'^.*?(\d(?:{number_continuation_pattern})*)$')
 # Text after the end of a reference. Example: "1 John 3:2 2" –> " 2"

{scripturelookup-0.1.1 → scripturelookup-0.1.2/src/scripturelookup.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scripturelookup
-Version: 0.1.1
+Version: 0.1.2
 Summary: Python and command-line utility for converting scripture references between formats.
 Author-email: Samuel Bradshaw <samuel.h.bradshaw@gmail.com>
 License: MIT License

scripturelookup-0.1.2/src/scripturelookup.egg-info/scm_version.json

@@ -0,0 +1,8 @@
+{
+  "tag": "0.1.2",
+  "distance": 0,
+  "node": "gae49c65f49fb9037317e89a50a7d60cc7ce68ca9",
+  "dirty": false,
+  "branch": "HEAD",
+  "node_date": "2026-09-20"
+}

{scripturelookup-0.1.1 → scripturelookup-0.1.2}/tests/test_detect.py

@@ -255,6 +255,31 @@ def test_spaced_out_verses_are_not_stripped_as_trailing_text():
   assert lookup.get_label('1 John 3:2 2') == '1\xa0John\xa03:2'
 
 
+# Words inside a parenthetical aren't context verses, even when a second reference splits the parenthetical apart
+def test_words_in_a_parenthetical_are_not_context_verses():
+  text = 'Matthew 7:1–2 (see JST Matthew 7:1–2)'
+  assert lookup.get_church_uri(text) == '/scriptures/nt/matt/7.1-2\n/scriptures/jst/jst-matt/7.1-2'
+  assert lookup.get_church_uri(text, use_query_parameters = True) == '/scriptures/nt/matt/7?id=p1-p2\n/scriptures/jst/jst-matt/7?id=p1-p2'
+  # ...while numbers still are
+  assert lookup.get_church_uri('Gen. 1:3 (3–4)') == '/scriptures/ot/gen/1.3(3-4)'
+
+
+# A footnote letter after a verse marks a study note, so it's dropped rather than taking the verse with it
+@pytest.mark.parametrize('text, expected_uri', [
+  ('Genesis 6:7a', '/scriptures/ot/gen/6.7'),
+  ('Genesis 6:7a, 13', '/scriptures/ot/gen/6.7,13'),
+  ('Alma 32:21a–23b', '/scriptures/bofm/alma/32.21-23'),
+  ('Gen. 1:3a (3–4)', '/scriptures/ot/gen/1.3(3-4)'),
+])
+def test_verse_footnote_letters_are_dropped(text, expected_uri):
+  assert lookup.get_church_uri(text) == expected_uri
+
+
+# A footnote letter is citation syntax rather than sloppy input, so it goes even when cleanup is skipped
+def test_verse_footnote_letters_are_dropped_even_when_cleanup_is_skipped():
+  assert lookup.get_church_uri('Genesis 6:7a', skip_cleanup = True) == '/scriptures/ot/gen/6.7'
+
+
 # A book with only one chapter takes a bare number as its verse – there is no Jude 3
 @pytest.mark.parametrize('text, expected_uris', [
   ('Jude 3', ['/scriptures/nt/jude/1.3']),
@@ -291,11 +316,20 @@ def test_single_chapter_rule_applies_even_when_cleanup_is_skipped():
   ('Premier Néphi 3:7', 'fr', '/scriptures/bofm/1-ne/3.7'),
   ('1er Néphi 3:7', 'fr', '/scriptures/bofm/1-ne/3.7'),
   ('Primer Nefi 3:7', 'es', '/scriptures/bofm/1-ne/3.7'),
+  ('1o Nefi 3:7', 'es', '/scriptures/bofm/1-ne/3.7'),
+  ('3a Nefi 11', 'es', '/scriptures/bofm/3-ne/11'),
+  ('2o Néfi 2:25', 'pt', '/scriptures/bofm/2-ne/2.25'),
 ])
 def test_book_numbers_may_be_written_out(text, lang, expected_uri):
   assert lookup.get_church_uri(text, lang = lang) == expected_uri
 
 
+# "1o" and "1a" stand in for "1º" and "1ª", but a verse with a footnote letter is never read as an ordinal
+def test_verse_letters_are_not_ordinals():
+  assert lookup.detect_references('Ver Génesis 6:7a Nefi 3:7.', lang = 'es') == [['Génesis 6:7', 4, 15]]
+  assert lookup.detect_references('Ver Alma 32:21a y 3a Nefi 11.', lang = 'es') == [['Alma 32:21', 4, 14], ['3a Nefi 11', 18, 28]]
+
+
 # An article of faith cited by position is that article, since the book is a single chapter
 @pytest.mark.parametrize('text, lang, expected_uris', [
   ('First Article of Faith', 'en', ['/scriptures/pgp/a-of-f/1.1']),
@@ -304,6 +338,8 @@ def test_book_numbers_may_be_written_out(text, lang, expected_uri):
   ('1er article de foi', 'fr', ['/scriptures/pgp/a-of-f/1.1']),
   ('Primer Artículo de Fe', 'es', ['/scriptures/pgp/a-of-f/1.1']),
   ('Primeira regra de fé', 'pt', ['/scriptures/pgp/a-of-f/1.1']),
+  ('1o artículo de fe', 'es', ['/scriptures/pgp/a-of-f/1.1']),
+  ('1a regra de fé', 'pt', ['/scriptures/pgp/a-of-f/1.1']),
   ('First and Third Articles of Faith', 'en', ['/scriptures/pgp/a-of-f/1.1', '/scriptures/pgp/a-of-f/1.3']),
   ('First through Third Articles of Faith', 'en', ['/scriptures/pgp/a-of-f/1.1-3']),
   ('Premier et troisième articles de foi', 'fr', ['/scriptures/pgp/a-of-f/1.1', '/scriptures/pgp/a-of-f/1.3']),
@@ -353,6 +389,18 @@ def test_abbreviated_ordinals_as_book_numbers(text, lang, expected_uri):
   assert lookup.get_church_uri(text, lang = lang) == expected_uri
 
 
+# An ordinal matched with its accent left out resolves to the same number, since the character classes in the
+# pattern accept spellings that aren't listed as forms
+@pytest.mark.parametrize('text, lang, expected_uri', [
+  ('Decimo articulo de fe', 'es', '/scriptures/pgp/a-of-f/1.10'),
+  ('decimo primer articulo de fe', 'es', '/scriptures/pgp/a-of-f/1.11'),
+  ('Deuxieme article de foi', 'fr', '/scriptures/pgp/a-of-f/1.2'),
+  ('Premier Nephi 3:7', 'fr', '/scriptures/bofm/1-ne/3.7'),
+])
+def test_ordinals_resolve_without_their_accents(text, lang, expected_uri):
+  assert lookup.get_church_uri(text, lang = lang) == expected_uri
+
+
 # Rewriting an ordinal or a written-out book number is cleanup, so skip_cleanup skips it
 def test_skip_cleanup_leaves_written_out_numbers_alone():
   assert lookup.get_church_uri('Premier article de foi', lang = 'fr', skip_cleanup = True) == ''

scripturelookup-0.1.1/src/scripturelookup.egg-info/scm_version.json

@@ -1,8 +0,0 @@
-{
-  "tag": "0.1.1",
-  "distance": 0,
-  "node": "g23aad54fd3c9ffe72276ccacf9b1a806ac7c05cb",
-  "dirty": false,
-  "branch": "HEAD",
-  "node_date": "2026-08-29"
-}