Support multiple names for gnr_resolve(), Add POST. Fixes #12. Resolution from TNRS Fixes #8.

pytaxize/__init__.py

@@ -21,6 +21,7 @@
     6  6971717    Apis nigrocincta  Species]
 '''
 
+from .tnrs import tnrs_resolve
 from .gnr import gnr_datasources, gnr_resolve
 from .gni import gni_parse, gni_search, gni_details
 from .col import col_children, col_downstream, col_search

pytaxize/gnr.py

@@ -52,10 +52,9 @@ def gnr_datasources(todf=True):
     return df
 
 def gnr_resolve(names='Homo sapiens', source=None, format='json', resolve_once='false',
-    with_context='false', best_match_only='false', header_only='false', preferred_data_sources='false'):
+    with_context='false', best_match_only='false', header_only='false', preferred_data_sources='false', http='get'):
     '''
     Uses the Global Names Resolver to resolve scientific names
-
     :param names: List of taxonomic names
     :param source: Source to pull from, one of x, y, z
     :param format: One of json or xml
@@ -64,31 +63,54 @@ def gnr_resolve(names='Homo sapiens', source=None, format='json', resolve_once='
     :param best_match_only: Logical, if true (default) return the best match only
     :param header_only: Return header only, logical
     :param preferred_data_sources: Return only preferred data sources.
-
+    :param http: The HTTP method to use, one of "get" or "post". Default="get"
     Usage:
     >>> import pytaxize
     >>> pytaxize.gnr_resolve('Helianthus annus')
     [{u'classification_path': u'', u'data_source_title': u'EOL', u'match_type': 1, u'score': 0.988, u'url': u'http://eol.org/pages/468106/names/synonyms', u'classification_path_ranks': u'', u'name_string': u'Helianthus annus', u'prescore': u'3|0|0', u'canonical_form': u'Helianthus annus', u'classification_path_ids': u'', u'local_id': u'468106', u'data_source_id': 12, u'taxon_id': u's_5106367', u'gni_uuid': u'f5674e32-00cc-57e3-b632-6a0b89fa4df4'}, {u'classification_path': u'|Helianthus annus', u'data_source_title': u'uBio NameBank', u'match_type': 1, u'score': 0.988, u'url': u'http://www.ubio.org/browser/details.php?namebankID=10130157', u'classification_path_ranks': u'kingdom|', u'name_string': u'Helianthus annus', u'global_id': u'urn:lsid:ubio.org:namebank:10130157', u'prescore': u'3|0|0', u'canonical_form': u'Helianthus annus', u'classification_path_ids': u'', u'local_id': u'urn:lsid:ubio.org:namebank:10130157', u'data_source_id': 169, u'taxon_id': u'102910884', u'gni_uuid': u'f5674e32-00cc-57e3-b632-6a0b89fa4df4'}, {u'classification_path': u'', u'data_source_title': u'EOL', u'match_type': 2, u'score': 0.988, u'url': u'http://eol.org/pages/468106', u'classification_path_ranks': u'', u'name_string': u'Helianthus annus L.', u'prescore': u'3|0|0', u'canonical_form': u'Helianthus annus', u'classification_path_ids': u'', u'local_id': u'468106', u'data_source_id': 12, u'taxon_id': u'20584982', u'gni_uuid': u'e757b3c1-421f-5bb9-a27f-d56259baaf3d'}]
     >>> pytaxize.gnr_resolve(['Helianthus annus','Poa annua'])
     '''
     url = 'http://resolver.globalnames.org/name_resolvers'
-    if(names.__class__.__name__ == 'list'):
-        names = "|".join(names)
-    else:
-        pass
-    payload = {'names': names, 'data_source_ids': source, 'format': format,
+    payload = {'data_source_ids': source, 'format': format,
                 'resolve_once': resolve_once, 'with_context': with_context,
                 'best_match_only': best_match_only, 'header_only': header_only,
                 'preferred_data_sources': preferred_data_sources}
-    out = requests.get(url, params = payload)
-    out.raise_for_status()
-    result_json = out.json()
-    # Return [] for each query with no returned result
+    if names.__class__.__name__ == 'list':
+        if len(names) > 300 and http == 'get':
+            http = 'post'
+        else:
+            names = "|".join(names)
+            payload['names'] = names
+    else:
+        payload['names'] = names
+    if http == 'get':
+        out = requests.get(url, params = payload)
+        out.raise_for_status()
+        result_json = out.json()
+    else:
+        if names.__class__.__name__ != 'list':
+            out = requests.post(url, params = payload)
+            out.raise_for_status()
+            result_json = out.json()
+        else:
+            with open('names_list.txt', 'wb') as f:
+                for name in names:
+                    f.write(name+"\n")
+            f.close()
+            out = requests.post(url, params = payload, files = {'file': open('names_list.txt', 'rb')} )
+            out.raise_for_status()
+            result_json = out.json()
+            while result_json['status'] == 'working':
+                result_url = result_json['url']
+                time.sleep(10)
+                out = requests.get(url=result_url)
+                result_json = out.json()
+
     data = []
     for each_result in result_json['data']:
         data.append( each_result['results'] if 'results' in each_result else [])
     return data
-    
+
 if __name__ == "__main__":
     import doctest
     doctest.testmod()

pytaxize/tnrs.py

@@ -0,0 +1,46 @@
+import sys
+import requests
+import json
+
+class NoResultException(Exception):
+    pass
+
+def tnrs_resolve(names='Homo sapiens', retrieve='all'):
+    '''
+    Uses the Global Names Resolver to resolve scientific names
+    :param names: List of taxonomic names
+    :param retrieve: all OR best
+    Usage:
+    >>> import pytaxize
+    >>> pytaxize.tnrs_resolve('Helianthus annus')
+    [{u'selected': True, u'infraspecific2Rank': u'', u'family': u'Asteraceae', u'infraspecific1EpithetScore': u'', u'infraspecific1Rank': u'', u'nameScientific': u'Helianthus annuus', u'speciesMatched': u'annuus', u'authorScore': u'', u'group': u'0', u'author': u'', u'acceptance': u'No opinion', u'authorAttributed': u'L.', u'unmatched': u"'", u'nameSubmitted': u"Helianthus annus'", u'genusScore': u'1', u'matchedFamilyScore': u'', u'infraspecific2EpithetScore': u'', u'infraspecific1Epithet': u'', u'infraspecific2Epithet': u'', u'familySubmitted': u'', u'epithet': u'annuus', u'acceptedName': u'Helianthus annuus', u'overall': u'0.8694502693699', u'speciesMatchedScore': u'0.83333333333333', u'matchedFamily': u'', u'acceptedNameUrl': u'http://www.tropicos.org/Name/2700851', u'epithetScore': u'0.83333333333333', u'annotation': u'', u'url': u'http://www.tropicos.org/Name/2700851', u'scientificScore': u'0.9694502693699', u'acceptedAuthor': u'L.', u'genus': u'Helianthus'}, {u'selected': False, u'infraspecific2Rank': u'', u'family': u'Violaceae', u'infraspecific1EpithetScore': u'', u'infraspecific1Rank': u'', u'nameScientific': u'Hybanthus nanus', u'speciesMatched': u'nanus', u'authorScore': u'', u'group': u'0', u'author': u'', u'acceptance': u'No opinion', u'authorAttributed': u'(A. St.-Hil.) Paula-Souza', u'unmatched': u"'", u'nameSubmitted': u"Helianthus annus'", u'genusScore': u'0.7', u'matchedFamilyScore': u'', u'infraspecific2EpithetScore': u'', u'infraspecific1Epithet': u'', u'infraspecific2Epithet': u'', u'familySubmitted': u'', u'epithet': u'nanus', u'acceptedName': u'Hybanthus nanus', u'overall': u'0.67149326622765', u'speciesMatchedScore': u'0.8', u'matchedFamily': u'', u'acceptedNameUrl': u'http://www.tropicos.org/Name/100000197', u'epithetScore': u'0.8', u'annotation': u'', u'url': u'http://www.tropicos.org/Name/100000197', u'scientificScore': u'0.77149326622765', u'acceptedAuthor': u'(A. St.-Hil.) Paula-Souza', u'genus': u'Hybanthus'}]
+    >>> pytaxize.tnrs_resolve(['Helianthus annus','Poa annua'])
+    '''
+    url = 'http://tnrs.iplantc.org/tnrsm-svc/matchNames'
+    if(names.__class__.__name__ == 'list'):
+        names = ",".join(names)
+    else:
+        pass
+    payload = {'retrieve': retrieve, 'names': names }
+    out = requests.get(url, params = payload)
+    out.raise_for_status()
+    result_json = out.json()
+
+    data = []
+    #List to accumulate results for each queried name
+    single_list=[]
+    index = result_json['items'][0]['group']
+    for each_result in result_json['items']:
+        if each_result['group'] == index:
+            single_list.append(each_result)
+        else:
+            data.append(single_list)
+            single_list = []
+            single_list.append(each_result)
+            index = each_result['group']
+    data.append(single_list)
+    return data
+
+if __name__ == "__main__":
+    import doctest
+    doctest.testmod()