Fix bug: #12 case sensitivity causes problems with retrievers

dbxref/config.py

@@ -3,11 +3,14 @@ def get_providers_path():
     return pkg_resources.resource_filename('dbxref', 'providers.yaml')
 
 def load_providers():
+    return _load_providers(get_providers_path())
+
+def _load_providers(path):
     import yaml
     data = []
-    with open(get_providers_path()) as data_file:
+    with open(path) as data_file:
         data = yaml.load(data_file)
-    return index_providers(data)
+    return normalize_index(index_providers(data))
 
 def index_providers(providers):
     index = {}
@@ -15,3 +18,16 @@ def index_providers(providers):
         for db in p['prefixes']:
             index[db] = p
     return index
+
+def normalize_index(index):
+    'create a new index with lowercase keys'
+    return {k.lower():v for (k,v) in index.items()}
+
+def has_provider(provider):
+    return _has_provider(load_providers(), provider)
+
+def _has_provider(providers, provider):
+    return provider.lower() in providers
+
+def get_provider(provider):
+    return load_providers()[provider.lower()]

dbxref/providers.yaml

 - name: Enzyme
-  prefixes: ["EC", "ec"]
+  prefixes: ["EC"]
   resources:
       html: ["https://enzyme.expasy.org/EC/%i"]
       text: ["https://enzyme.expasy.org/EC/%i.txt"]
@@ -8,14 +8,14 @@
       type: 'internal'
       location: 'dbxref.retrieve.enzyme'
 - name: Gene Identifier
-  prefixes: ["GI", "gi"]
+  prefixes: ["GI"]
   resources:
       html: ["http://www.ncbi.nlm.nih.gov/protein/GI:%i"]
       xml: ["http://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?tool=portal&db=protein&dopt=xml&sort=&val=%i&retmode=file"]
   # does not work
   # check_existence: "http://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?tool=portal&db=protein&dopt=xml&sort=&val=%i&retmode=file"
 - name: Uniprot
-  prefixes: ["UniProtKB/TrEMBL", "UniProtKB/Swiss-Prot", "uniprotkb/trembl", "uniprotkb/swiss-prot"]
+  prefixes: ["UniProtKB/TrEMBL", "UniProtKB/Swiss-Prot"]
   resources:
       html: [ "http://www.uniprot.org/uniprot/%i"]
       xml: [ "http://www.uniprot.org/uniprot/%i.xml"]
@@ -24,7 +24,7 @@
       type: 'internal'
       location: 'dbxref.retrieve.uniprot'
 - name: Taxonomy
-  prefixes: ["Taxon", "taxon", "taxid"]
+  prefixes: ["Taxon", "taxid"]
   resources:
       html: ["http://www.uniprot.org/taxonomy/%i"]
       json: ["https://www.ebi.ac.uk/ena/data/taxonomy/v1/taxon/tax-id/%i"]
@@ -35,7 +35,7 @@
       type: 'internal'
       location: 'dbxref.retrieve.taxonomy'
 - name: SequenceOntology
-  prefixes: ["SO", "so"]
+  prefixes: ["SO"]
   resources:
       html: ["http://www.sequenceontology.org/browser/current_svn/term/SO:%i"]
       obo: ["http://www.sequenceontology.org/browser/current_svn/export/term_only/obo/SO:%i"]
@@ -46,19 +46,19 @@
       type: 'internal'
       location: 'dbxref.retrieve.sequence_ontology'
 - name: RFAM
-  prefixes: ["RFAM", "rfam"]
+  prefixes: ["RFAM"]
   resources:
       html: ["http://rfam.xfam.org/family/%i"]
       xml: ["http://rfam.xfam.org/family/%i?content-type=text%2Fxml"]
   # does not work
   # check_existence: "http://rfam.xfam.org/family/%i?content-type=text%2Fxml"
 - name: Pubmed
-  prefixes: ["pubmed", "Pubmed"]
+  prefixes: ["Pubmed"]
   resources:
       html: ["http://www.ncbi.nlm.nih.gov/pubmed/%i"]
   check_existence: "http://www.ncbi.nlm.nih.gov/pubmed/%i"
 - name: Protein Families
-  prefixes: ["PFAM", "Pfam", "pfam"]
+  prefixes: ["PFAM"]
   resources:
       html: ["http://pfam.xfam.org/family/%i"]
       xml: ["http://pfam.xfam.org/family/%i?output=xml"]
@@ -68,26 +68,26 @@
       type: 'internal'
       location: 'dbxref.retrieve.pfam'
 - name: PDB
-  prefixes: ["PDB", "pdb"]
+  prefixes: ["PDB"]
   resources:
       html: ["http://www.rcsb.org/pdb/explore/explore.do?structureId=%i"]
       xml: ["http://www.rcsb.org/pdb/download/downloadFile.do?fileFormat=xml&compression=NO&structureId=%i"]
   check_existence: "http://www.rcsb.org/pdb/download/downloadFile.do?fileFormat=xml&compression=NO&structureId=%i"
 - name: InterPro
-  prefixes: ["InterPro", "interpro"]
+  prefixes: ["InterPro"]
   resources:
       html: ["http://www.ebi.ac.uk/interpro/entry/%i"]
   # does not work
   # check_existence: "http://www.ebi.ac.uk/interpro/entry/%i"
 - name: GeneID
-  prefixes: ["GeneID", "geneid"]
+  prefixes: ["GeneID"]
   resources:
       html: ["http://www.ncbi.nlm.nih.gov/gene/%i"]
       xml: ["http://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?tool=portal&db=gene&dopt=xml&sort=&val=%i&retmode=file"]
   # does not work
   #  check_existence: "http://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?tool=portal&db=gene&dopt=xml&sort=&val=%i&retmode=file"
 - name: Gene Ontology
-  prefixes: ["GO", "go"]
+  prefixes: ["GO"]
   resources:
       html: ["http://www.ebi.ac.uk/QuickGO/GTerm?id=GO:%i"]
       xml: ["http://www.ebi.ac.uk/QuickGO/GTerm?id=GO:%i&format=oboxml"]

dbxref/resolver.py

@@ -4,8 +4,7 @@ from cachecontrol.caches.file_cache import FileCache
 import logging
 logger = logging.getLogger(__name__)
 
-from dbxref.config import load_providers
-providers = load_providers()
+from dbxref import config
 
 cache = FileCache(".web_cache", forever=True)
 sess = CacheControl(requests.Session(), cache=cache)
@@ -24,8 +23,8 @@ def resolve(dbxrefs, check_existence=True):
         status = STATUS_NOT_CHECKED
         if check_existence:
            status = check_dbxref_exists(dbxref)
-        if dbxref['db'] in providers:
-            provider = providers[dbxref['db']]
+        if config.has_provider(dbxref['db']):
+            provider = config.get_provider(dbxref['db'])
             locations = {}
             for _type in provider['resources']:
                 urls = []
@@ -42,8 +41,8 @@ def convert_to_dbxrefs(strings):
   return list(map(convert_string_to_dbxref, strings))
 
 def check_dbxref_exists(dbxref):
-    if dbxref['db'] in providers:
-        provider = providers[dbxref['db']]
+    if config.has_provider(dbxref['db']):
+        provider = config.get_provider(dbxref['db'])
         urls = []
         exists = STATUS_NOT_CHECKED
         if 'check_existence' in provider:

dbxref/retriever.py

 import logging
 logger = logging.getLogger(__name__)
 
-from dbxref.config import load_providers
+from dbxref import config
 from itertools import groupby
 import json
 
-providers = load_providers()
-
 def retrieve(dbxrefs):
-    sorted(dbxrefs, key=lambda x: x['db'])
+    sorted(dbxrefs, key=lambda x: x['db'].lower()) # normalize db to lowercase to allow differently cased notations
     results = []
     for key, dbxrefs in groupby(dbxrefs, lambda x: x['db']):
-        if key.lower() in providers and 'retriever' in providers[key.lower()]:
-            provider = providers[key.lower()]
+        if config.has_provider(key):
+            provider = config.get_provider(key)
             logger.debug('{0} is supported'.format(key))
             if provider['retriever']['type'] == 'external':
                 results.extend( load_with_external_provider(provider, list(dbxrefs)))

tests/test_config.py

@@ -9,3 +9,12 @@ class TestConfig(unittest.TestCase):
     def test_index_providers(self):
         data = [{'name': 'test', 'prefixes':['a', 'b']}]
         self.assertEqual(config.index_providers(data), {'a': data[0], 'b': data[0]})
+
+    def test_normalize_index(self):
+        index = {'A': 'some value', 'b': 'some other value'}
+        self.assertEqual(config.normalize_index(index), {'a' : 'some value', 'b':'some other value'})
+
+    def test_has_provider(self):
+        index = config.normalize_index({'A': 'some value', 'b': 'some other value'})
+        self.assertTrue(config._has_provider(index, 'B'))
+        self.assertTrue(config._has_provider(index, 'a'))

tests/test_retriever.py

+import unittest
+from dbxref import retriever, resolver
+
+class TestDbxrefResolve(unittest.TestCase):
+
+    def test_different_case_database_prefix(self):
+        entries = resolver.convert_to_dbxrefs(['PFAM:PF00002','Pfam:PF00002','pfam:PF00002'])
+        documents = retriever.retrieve(entries)
+        for d in documents:
+          with self.subTest(d=d):
+            self.assertTrue('description' in d)