rdf+bcp47+hxl (#41): OBO Ontology (from HXL hashtags) draft

EticaAI · Jun 6, 2022 · 7323103 · 7323103
1 parent 8baab27
commit 7323103
Showing 1 changed file with 57 additions and 7 deletions.
diff --git a/officina/999999999/0/L999999999_0.py b/officina/999999999/0/L999999999_0.py
@@ -383,7 +383,9 @@
     'obo:ro1025': 'obo:RO_0001025',  # purl.obolibrary.org/obo/RO_0001025
     'obo:rolocatedin': 'obo:BFO_0000002',
 }
-# @TODO https://protegeproject.github.io/owl2lpg/
+# Note: prefixes that already are lower case do not be here
+RDF_NAMESPACES_PREFIX_EXTRAS = {
+}
 
 
 def bcp47_langtag(
@@ -995,7 +997,7 @@ def bcp47_rdf_extension(
         # 'csvw:separator': '', # Added only if necessary
         # 'prefix': [],  # Added only if necessary
     }
-    _predicates = []
+    # _predicates = []
     _subjects = []
     _objects = []
 
@@ -1102,8 +1104,17 @@ def bcp47_rdf_extension(
         # if len(_predicates) > 0:
         #     _predicates.sort()
         #     result['rdf:predicate'] = _predicates
-        if len(_predicates) > 1:
+        if len(result['rdf:predicate']) > 0:
             result['rdf:predicate'].sort()
+            # raise ValueError(result['rdf:predicate'])
+            # for index in range(len(result['rdf:predicate'])):
+            #     predicate_prefix, predicate, subject = \
+            #         result['rdf:predicate'][index].split(':')
+            #     raw_predicate = f'{predicate_prefix}:{predicate}'
+            #     if raw_predicate in RDF_NAMESPACES_PREFIX:
+            #         result['rdf:predicate'] = '{0}:{1}'.format(
+            #             RDF_NAMESPACES_PREFIX[raw_predicate], subject)
+            #     # pass
 
         if len(_objects) > 0:
             _objects.sort()
@@ -2759,11 +2770,50 @@ def hxl_hashtag_to_bcp47(hashtag: str) -> str:
                 ))
 
             elif item.startswith('p_'):
-                _predicate = item.replace('p_', '').replace('_', ':')
-                result['extension']['r']['rdf:predicate'].append(_predicate)
-                _predicate_key, _object, _subject = _predicate.split(':')
+                _item_parts = item.replace('p_', '').replace('_', ':')
+                result['extension']['r']['rdf:predicate'].append(_item_parts)
+                _index_p = result['extension']['r']['rdf:predicate'].index(
+                    _item_parts
+                )
+                _predicate_ns, _predicate_item, _subject = \
+                    _item_parts.split(':')
+
+                raw_predicate = f'{_predicate_ns}:{_predicate_item}'
+                normalized_predicate = None
+
+                if raw_predicate in RDF_NAMESPACES_PREFIX:
+                    normalized_predicate = RDF_NAMESPACES_PREFIX[raw_predicate]
+                elif raw_predicate in RDF_NAMESPACES_PREFIX_EXTRAS:
+                    normalized_predicate = \
+                        RDF_NAMESPACES_PREFIX_EXTRAS[raw_predicate]
+
+                if normalized_predicate is not None:
+                    if normalized_predicate.startswith('obo:'):
+                        normalized_predicate = normalized_predicate.lower()
+                        _predicate_ns = 'obo'
+                        _predicate_item_raw = \
+                            normalized_predicate.replace('obo:', '')
+                        _predicate_item_raw_digits = ''.join(
+                            filter(str.isdigit, _predicate_item_raw))
+                        _predicate_item_raw_alpha = \
+                            _predicate_item_raw.replace(\
+                                _predicate_item_raw_digits, '').replace(
+                                    '_', '')
+                        _predicate_item = '{0}{1}'.format(
+                            _predicate_item_raw_alpha,
+                            _predicate_item_raw_digits.lstrip('0')
+                        )
+                    else:
+                        _predicate_ns, _predicate_item = \
+                            normalized_predicate.split(':')
+                        # pass
+
+                    result['extension']['r']['rdf:predicate'][_index_p] = \
+                        '{0}:{1}'.format(
+                        RDF_NAMESPACES_PREFIX[raw_predicate], _subject)
+
                 _bpc47_g_parts.append('p{0}-p{1}-p{2}'.format(
-                    _predicate_key.upper(), _object, _subject
+                    _predicate_ns.upper(), _predicate_item, _subject
                 ))
 
             elif item.startswith('y_'):