increased weight for symbol in api query search and added tests

biothings · Aug 2, 2024 · d91df91 · d91df91
1 parent 3116a7b
commit d91df91
Show file tree

Hide file tree

Showing 2 changed files with 47 additions and 69 deletions.
diff --git a/src/tests/data_tests/test_2_query.py b/src/tests/data_tests/test_2_query.py
@@ -181,6 +181,11 @@ def test_260_order(self):
         assert (10116, 367901) in genes  # rat
         assert (695, 9606) not in genes  # this field should not be sorted
 
+    def test_261_order(self):
+        res = self.request("query?q=CTNNA2&species=human").json()
+        assert res["hits"][0]["_id"] == "1496"
+        assert res["hits"][0]["symbol"] == "CTNNA2"
+
 
 class TestQueryPOST(BiothingsDataTest):
     host = "mygene.info"

diff --git a/src/web/pipeline/legacy.py b/src/web/pipeline/legacy.py
@@ -8,21 +8,17 @@
 def dismax(q):
 
     _query = {
-
         "tie_breaker": 0,
         "boost": 1,
         "queries": [
             {
                 "function_score": {
                     "query": {
                         "match": {
-                            "symbol": {
-                                "query": q,
-                                "analyzer": "whitespace_lowercase"
-                            }
+                            "symbol": {"query": q, "analyzer": "whitespace_lowercase"}
                         },
                     },
-                    "weight": 5
+                    "weight": 8,
                 }
             },
             {
@@ -32,8 +28,7 @@ def dismax(q):
                         # kinase 2" appears first
                         "match_phrase": {"name": q},
                     },
-                    "weight": 4
-
+                    "weight": 4,
                 }
             },
             {
@@ -43,24 +38,21 @@ def dismax(q):
                             "name": {
                                 "query": q,
                                 "operator": "and",
-                                "analyzer": "whitespace_lowercase"
+                                "analyzer": "whitespace_lowercase",
                             }
                         },
                     },
-                    "weight": 3
+                    "weight": 3,
                 }
             },
             {
                 "function_score": {
                     "query": {
                         "match": {
-                            "unigene": {
-                                "query": q,
-                                "analyzer": "string_lowercase"
-                            }
+                            "unigene": {"query": q, "analyzer": "string_lowercase"}
                         }
                     },
-                    "weight": 1.1
+                    "weight": 1.1,
                 }
             },
             {
@@ -69,65 +61,53 @@ def dismax(q):
                         "multi_match": {
                             "query": q,
                             "fields": [
-                                'refseq.rna',
-                                'refseq.protein',
-                                'accession.rna',
-                                'accession.protein'
+                                "refseq.rna",
+                                "refseq.protein",
+                                "accession.rna",
+                                "accession.protein",
                             ],
-                            "operator": "or"
+                            "operator": "or",
                         }
                     },
-                    "weight": 1.1
+                    "weight": 1.1,
                 }
             },
             {
                 "function_score": {
                     "query": {
-                        "match": {
-                            "go": {
-                                "query": q,
-                                "analyzer": "string_lowercase"
-                            }
-                        }
+                        "match": {"go": {"query": q, "analyzer": "string_lowercase"}}
                     },
-                    "weight": 1.1
+                    "weight": 1.1,
                 }
             },
             {
                 "function_score": {
                     "query": {
-                        "query_string": {
-                            "query": q,
-                            "default_operator": "AND"
-                        },
+                        "query_string": {"query": q, "default_operator": "AND"},
                     },
-                    "weight": 1
+                    "weight": 1,
                 }
-            }
-        ]
+            },
+        ],
     }
 
     if is_int(q):
-        _query['queries'] = [
+        _query["queries"] = [
             {
                 "function_score": {
                     "query": {
                         "term": {"entrezgene": int(q)},
                     },
-                    "weight": 8
+                    "weight": 8,
                 }
             }
         ]
 
-    return {
-        "query": {
-            "dis_max": _query
-        }
-    }
+    return {"query": {"dis_max": _query}}
 
 
 def wildcard(q):
-    '''q should contains either * or ?, but not the first character.'''
+    """q should contains either * or ?, but not the first character."""
 
     _query = {
         "tie_breaker": 0,
@@ -166,45 +146,41 @@ def wildcard(q):
                     }
                 }
             },
-        ]
+        ],
     }
 
-    return {
-        "query": {
-            "dis_max": _query
-        }
-    }
+    return {"query": {"dis_max": _query}}
 
 
 def safe_genome_pos(s):
-    '''
-        safe_genome_pos(1000) = 1000
-        safe_genome_pos('1000') = 1000
-        safe_genome_pos('10,000') = 100000
-     '''
+    """
+    safe_genome_pos(1000) = 1000
+    safe_genome_pos('1000') = 1000
+    safe_genome_pos('10,000') = 100000
+    """
     if isinstance(s, int):
         return s
     elif isinstance(s, str):
-        return int(s.replace(',', ''))
+        return int(s.replace(",", ""))
     else:
         raise ValueError('invalid type "%s" for "save_genome_pos"' % type(s))
 
 
 def interval(chrom, gstart, gend, assembly=None):
-    '''By default if assembly is None, the lastest assembly is used.
-       for some species (e.g. human) we support multiple assemblies,
-       exact assembly is passed as well.
-    '''
+    """By default if assembly is None, the lastest assembly is used.
+    for some species (e.g. human) we support multiple assemblies,
+    exact assembly is passed as well.
+    """
     gstart = safe_genome_pos(gstart)
     gend = safe_genome_pos(gend)
-    if chrom.lower().startswith('chr'):
+    if chrom.lower().startswith("chr"):
         chrom = chrom[3:]
 
     genomic_pos_field = "genomic_pos"
     if assembly:
-        if assembly == 'hg19':
+        if assembly == "hg19":
             genomic_pos_field = "genomic_pos_hg19"
-        if assembly == 'mm9':
+        if assembly == "mm9":
             genomic_pos_field = "genomic_pos_mm9"
 
     _query = {
@@ -213,15 +189,12 @@ def interval(chrom, gstart, gend, assembly=None):
             "query": {
                 "bool": {
                     "must": [
-                        {"term": {
-                            genomic_pos_field + ".chr": chrom.lower()}},
-                        {"range": {
-                            genomic_pos_field + ".start": {"lte": gend}}},
-                        {"range": {
-                            genomic_pos_field + ".end": {"gte": gstart}}}
+                        {"term": {genomic_pos_field + ".chr": chrom.lower()}},
+                        {"range": {genomic_pos_field + ".start": {"lte": gend}}},
+                        {"range": {genomic_pos_field + ".end": {"gte": gstart}}},
                     ]
                 }
-            }
+            },
         }
     }
     return dict(query=_query)