Merge pull request #481 from OmarAI2003/check_SELECT_WHERE_labels_ORDER

Check select where labels order

src/scribe_data/check/check_query_forms.py

@@ -438,6 +438,61 @@ def check_defined_return_forms(query_text: str) -> str:
     return ""
 
 
+# MARK: Forms Order
+
+
+def check_forms_order(query_text: str) -> bool:
+    """
+    Checks that the order of variables in the SELECT statement (excluding lexeme and lexemeID)
+    matches the order of the same variables in the WHERE clause in the given SPARQL query file.
+
+    Parameters
+    ----------
+        query_file : str
+            The SPARQL query text as a string.
+
+    Returns
+    -------
+        bool
+            True if the order of the matches, False otherwise.
+    """
+    select_pattern = r"SELECT\s+(.*?)\s+WHERE"
+
+    # Extracting the variables from the SELECT statement.
+    if select_match := re.search(select_pattern, query_text, flags=re.DOTALL):
+        select_vars = re.findall(r"\?(\w+)", select_match[1])
+
+    else:
+        return False  # invalid query format if no SELECT match
+
+    # Exclude the first two variables from select_vars.
+    select_vars = select_vars[2:]
+    # Regex pattern to capture the variables in the WHERE clause.
+    dt_pattern = r"WHERE\s*\{[^}]*?wikibase:lemma\s*\?\s*(\w+)\s*[;.]\s*"
+    forms_pattern = r"ontolex:representation \?([^ ;]+)"
+    where_vars = []
+
+    # Extracting variables from the WHERE clause.
+    dt_match = re.findall(dt_pattern, query_text)
+    if dt_match == ["lemma"]:
+        where_vars.append("preposition")
+
+    elif dt_match:
+        where_vars.append(dt_match[0])
+
+    where_vars += re.findall(forms_pattern, query_text)
+
+    # Handling labels provided by the labeling service  like 'case' and 'gender' in the same order as in select_vars.
+    for var in ["case", "gender", "auxiliaryVerb"]:
+        if var in select_vars:
+            # Insert in the corresponding index of where_vars.
+            index = select_vars.index(var)
+            where_vars.insert(index, var)
+
+    # Check if the order of variables matches.
+    return select_vars == where_vars
+
+
 # MARK: Main Query Forms Validation
 def check_query_forms() -> None:
     """
@@ -471,6 +526,12 @@ def check_query_forms() -> None:
             error_output += f"\n{index}. {query_file_str}: {defined_unreturned_forms}\n"
             index += 1
 
+        # Check the order of variables in the WHERE and SELECT clauses.
+        select_where_labels_matching = check_forms_order(query_text)
+        if not select_where_labels_matching:
+            error_output += f"\n{index}. {query_file_str}:\n  - The order of variables in the SELECT statement does not match their order in the query.\n"
+            index += 1
+
         if extract_forms_from_sparql(query_file):
             query_form_check_dict = {}
             for form_text in extract_forms_from_sparql(query_file):

src/scribe_data/wikidata/language_data_extraction/danish/verbs/query_verbs.sparql

@@ -20,7 +20,7 @@ WHERE {
 
   ?lexeme dct:language wd:Q9035 ;
     wikibase:lexicalCategory wd:Q24905 ;
-    wikibase:lemma ?infinitive
+    wikibase:lemma ?infinitive .
 
   # MARK: Infinitive Active
 

src/scribe_data/wikidata/language_data_extraction/estonian/adjectives/query_adjectives_3.sparql

@@ -4,6 +4,7 @@
 
 SELECT
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
+  ?adjective
   ?adessiveSingular
   ?adessivePlural
   ?ablativeSingular

src/scribe_data/wikidata/language_data_extraction/estonian/adjectives/query_adjectives_4.sparql

@@ -4,6 +4,7 @@
 
 SELECT
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
+  ?adjective
   ?essiveSingular
   ?essivePlural
   ?abessiveSingular

src/scribe_data/wikidata/language_data_extraction/hebrew/verbs/query_verbs_2.sparql

@@ -4,7 +4,6 @@
 
 SELECT
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
-  ?infinitive
   ?feminineImperativeSecondPersonSingular
   ?masculineImperativeSecondPersonSingular
   ?feminineImperativeSecondPersonPlural

src/scribe_data/wikidata/language_data_extraction/hindustani/hindi/postpositions/query_postpositions.sparql

@@ -5,7 +5,6 @@
 # Note: We need to filter for "hi" to remove Urdu (ur) words.
 
 SELECT
-  ?lexeme
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
   ?postposition
 

src/scribe_data/wikidata/language_data_extraction/hindustani/hindi/prepositions/query_prepositions.sparql

@@ -5,7 +5,6 @@
 # Note: We need to filter for "hi" to remove Urdu (ur) words.
 
 SELECT
-  ?lexeme
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
   ?preposition
 

src/scribe_data/wikidata/language_data_extraction/hindustani/urdu/postpositions/query_postpositions.sparql

@@ -5,7 +5,6 @@
 # Note: We need to filter for "ur" to remove Hindi (hi) words.
 
 SELECT
-  ?lexeme
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
   ?postposition
 

src/scribe_data/wikidata/language_data_extraction/hindustani/urdu/prepositions/query_prepositions.sparql

@@ -5,7 +5,6 @@
 # Note: We need to filter for "ur" to remove Hindi (hi) words.
 
 SELECT
-  ?lexeme
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
   ?preposition
 

src/scribe_data/wikidata/language_data_extraction/italian/proper_nouns/query_proper_nouns.sparql

@@ -5,7 +5,6 @@
 SELECT
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
   ?singular
-  ?plural
   ?gender
 
 WHERE {

src/scribe_data/wikidata/language_data_extraction/punjabi/shahmukhi/nouns/query_nouns.sparql

@@ -5,7 +5,6 @@
 # Note: We need to filter for "pnb" to select Shahmukhi words.
 
 SELECT
-  ?lexeme
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
   ?singular
   ?plural

src/scribe_data/wikidata/language_data_extraction/punjabi/shahmukhi/proper_nouns/query_proper_nouns.sparql

@@ -5,7 +5,6 @@
 # Note: We need to filter for "pnb" to select Shahmukhi words.
 
 SELECT
-  ?lexeme
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
   ?singular
   ?plural

src/scribe_data/wikidata/language_data_extraction/ukrainian/adjectives/query_adjectives.sparql

@@ -4,7 +4,7 @@
 
 SELECT
   (REPLACE(STR(?lexeme), "http://www.wikidata.org/entity/", "") AS ?lexemeID)
-  ?lemma
+  ?adjective
   ?nominativeFeminineSingular
   ?nominativeMasculineSingular
   ?nominativeNeuterSingular
@@ -15,7 +15,7 @@ SELECT
 WHERE {
   ?lexeme dct:language wd:Q8798 ;
     wikibase:lexicalCategory wd:Q34698 ;
-    wikibase:lemma ?lemma .
+    wikibase:lemma ?adjective .
 
   OPTIONAL {
     ?lexeme ontolex:lexicalForm ?nominativeFeminineSingularForm .