Add Italian lexemes query (#82)

* add italian lexemes query

* add it in default query

---------

Co-authored-by: Michael F. Schönitzer <[email protected]>

Author: Jerkiller

backend.py

@@ -233,6 +233,7 @@ def get_lexeme_values(self) -> Tuple[int, int, Optional[int]]:
     "queries/en.sparql",
     "queries/fr.sparql",
     "queries/he.sparql",
+    "queries/it.sparql",
     "queries/nb.sparql",
     "queries/sv.sparql",
     "queries/withoutdescriptions.sparql",

queries/default.sparql

@@ -12,7 +12,8 @@ SELECT * WHERE {
          ?lang != wd:Q25167 &&  #nb
          ?lang != wd:Q150 &&    #fr
          ?lang != wd:Q9035 &&   #da
-         ?lang != wd:Q9288      #he
+         ?lang != wd:Q9288 &&   #he
+         ?lang != wd:Q652       #it
         ).
   ?lexeme wikibase:lemma ?lemma.
   # Search for potentially matching items with descriptions

queries/it.sparql

@@ -0,0 +1,134 @@
+SELECT ?lexeme ?item ?lemma ?desc ?cat ?genus (wd:Q652 as ?lang) WITH {
+  SELECT *
+  WHERE {
+    ?lexeme wikibase:lemma ?lemma;
+      dct:language wd:Q652;
+      wikibase:lexicalCategory ?cat.
+    OPTIONAL { ?lexeme wdt:P5185 ?genus. } .
+    FILTER NOT EXISTS {
+    ?lexeme ontolex:sense ?sense.
+    ?sense wdt:P5137 ?item.
+  }.
+  FILTER NOT EXISTS { ?item a wikibase:Property}.
+  FILTER NOT EXISTS { ?lexeme wdt:P5402 ?homograph }.
+  ?item rdfs:label ?lemma.
+  ?item schema:description ?desc.
+  FILTER(LANG(?desc) = "it").
+  }} AS %i
+WHERE {
+  include %i
+  # Filter out common false-positives
+  # anything not having a P31 has to go too (this is the first thing to relax when few positives are left)
+  #filter not exists {?item wdt:P31 [].} # stuff without P31 which is usually false positives 
+  FILTER NOT EXISTS { ?item wdt:P131 [] }. # admin area items
+  FILTER NOT EXISTS {
+    VALUES ?not {
+    wd:Q101352   # surname
+    wd:Q10429085 # report
+    wd:Q105543609 # musical work
+    wd:Q11424    # movie
+    wd:Q226730   # silent movie
+    wd:Q202866   # animation movie
+    wd:Q11446    # ships
+    wd:Q1145276  # fictional country (star trek)
+    wd:Q11668273 # shogi pieces
+    wd:Q1229071  # motif
+    wd:Q13406463 # wikimedia list article
+    wd:Q13417114 # noble family
+    wd:Q1391494  # technical terminology
+    wd:Q147276   # names of any kind
+    wd:Q1505023  # interpellation
+    wd:Q15056993 # aircraft family
+    wd:Q15632617 # fictive person
+    wd:Q15642541 # human settlement definitions
+    wd:Q15831596 # class of fictional entities e.g. star wars sandcrawler
+    wd:Q17155032 # type of software
+    wd:Q1725664  # terminology of any kind
+    wd:Q17537576 # any creative work incl. music, video, text, etc.
+    wd:Q17638537 # routine - fix computer terminology
+    wd:Q1897960  # trotting horse
+    wd:Q19798642 # WD value
+    wd:Q19798644 # wikibase model terms
+    wd:Q19798645 # wikibase datatype
+    wd:Q20202269 # music terms
+    wd:Q208569   # album
+    wd:Q21191270 # tv series episode
+    wd:Q215380   # bands
+    wd:Q2235308  # types of ships
+    wd:Q23786    # eye color
+    wd:Q3305213  # painting
+    wd:Q3331189  # version, edition, or translation
+    wd:Q3491429  # military terms
+    wd:Q355304   # watercourse (this has a lot of false positives that are not relevant in a Swedish lexeme context)
+    wd:Q3744866  # common charge (heraldic)
+    wd:Q386724   # work
+    wd:Q395      # math
+    wd:Q40056    # software
+    wd:Q4167410  # disambigpage
+    wd:Q4167836  # wikimedia cat
+    wd:Q43229    # organization of any kind
+    wd:Q482994   # album
+    wd:Q494452   # heraldry
+    wd:Q5        # human
+    wd:Q5185279  # poem
+    wd:Q56876983 # symbol (heraldic)
+    wd:Q57814795 # domesticated animal
+    wd:Q58408484 # Wikimedia list of persons by surname
+    wd:Q7187     # gene
+    wd:Q7366     # song
+    wd:Q7397     # software
+    wd:Q8134     # economics
+    wd:Q819989   # lunar crater
+    wd:Q99045339 # written question
+    wd:Q99281788 # star trek location
+    wd:Q29168811 # animated film
+    wd:Q5398426  # tv series
+    wd:Q3744880  # nature figure (heraldry)
+    wd:Q6881511  # enterprise
+    wd:Q3658341  # literary character
+    wd:Q1475691  # mars crater
+    wd:Q1348589  # lunar crater
+    wd:Q20643955 # biblical human figure
+    wd:Q134556   # single (music)
+    wd:Q786820   # automobile manuf.
+    wd:Q11060274 # print
+    wd:Q18218093 # other type of art
+    wd:Q7725634  # literary work
+    wd:Q22808320 # disambig name type
+    wd:Q3404720  # tincture
+    wd:Q66050470 # conjugation table
+  }.
+    ?item wdt:P31 ?not  }.
+  # filter on properties:
+  FILTER NOT EXISTS {
+    VALUES ?not2 {
+      wdt:P688  # gene stuff
+      wdt:P195  # collection (of art)
+      wdt:P1482 # stack exchange
+      }
+      ?item ?not2 [] }.
+  # filter on subclass:
+  FILTER NOT EXISTS {
+    VALUES ?not3 {
+      wd:Q783794  # company
+      wd:Q486839  # member of parliament
+      }
+      ?item wdt:P279 ?not3 }.
+  # filter on aspect of:
+  FILTER NOT EXISTS {
+    VALUES ?not3 {
+      # 2021-07-29
+      wd:Q18336  # heraldry
+      wd:Q494452 # blazonry
+      }
+      ?item wdt:P1269 ?not3 }.
+  # filter on part of:
+  FILTER NOT EXISTS {
+    VALUES ?not3 {
+      wd:Q18336  # heraldry
+      wd:Q494452 # blazonry
+      }
+      ?item wdt:P361 ?not3 }.
+  # special terminology often match these descriptions
+  # FILTER(!contains(?desc, 'araldica')) .
+}