def collect_all_species(fname="../data/tfidf/data_gbif.json"):
''' collect all species '''
last_read = 1
offset = 0
ts = time.time()
dct = {}
nb_species = 0
data = []
while last_read > 0:
tab_rep = species.name_suggest(offset=offset)
offset += 1
# if time.time()-ts>60:
# break
last_read = len(tab_rep)
for x in tab_rep:
nb_species += 1
data.append(x)
for k in x.keys():
if k not in dct:
dct[k] = 1
else:
dct[k] += 1
pp = pprint.PrettyPrinter(2)
print('nb species', nb_species)
print('Dict:', len(dct))
pp.pprint(dct)
with open(fname, "w", encoding='utf-8') as jsonfile:
json.dump(data, jsonfile, ensure_ascii=False)
print(time.time() - ts)
def test_species():
data = species.name_suggest(q='Puma concolor')
for x in data:
print(len(x))
data2 = species.name_suggest()
for x in data2:
print(len(x))
nb_species = 0
for i in range(0, 10):
tab_rep = species.name_suggest(q='vespa', offset=i)
print(len(tab_rep))
for doc in tab_rep:
if 'species' in doc:
nb_species += 1
print(doc['species'])
print(nb_species)
def gbif_name_suggest(**kwargs):
# Wrapper for pygbif name_suggest function
response = species.name_suggest(**kwargs)
if len(response) == 0:
return None
accepted_data = None
synonym_data = None
other_data = None
for result in response:
rank = result.get('rank', '')
if rank.lower() in RANK_KEYS:
rank_key = rank.lower() + 'Key'
else:
rank_key = 'key'
key_found = ('nubKey' in result or rank_key in result)
if key_found and 'status' in result:
if result['status'] == 'ACCEPTED':
if accepted_data:
if result['key'] < accepted_data['key']:
accepted_data = result
else:
accepted_data = result
if result['status'] == 'SYNONYM':
if synonym_data:
if result['key'] < synonym_data['key']:
synonym_data = result
else:
synonym_data = result
else:
if other_data:
if result['key'] < other_data['key']:
other_data = result
else:
other_data = result
if accepted_data:
return accepted_data
if synonym_data:
return synonym_data
return other_data
def taxon_info(taxon):
"""
use pygbif.species to get GBIF taxonKey info
"""
info = species.name_suggest(taxon)
return info
def search_species(entry):
# To be used when calling the method
name_input = entry
# User input for testing purposes
# name_input = input("Enter a common species name: ")
results = {}
# Uses the pygbif method to find results
suggest = species.name_suggest(q=name_input, rank='SPECIES', limit=25)
# Pulls the results from the dataset in json format
suggest_data = suggest['data']['results']
# print(suggest_data)
# Formats the data for pythonic purposes
data = suggest_data
# Reads data for each result
for o in data:
# Finds the gbif key
key = o['key']
# Uses pygbif to find number of occurrences of species key
occurs = occurrences.count(taxonKey=key)
# Searches occurrence data for the species key
occur_search = occurrences.search(taxonKey=key)
# print('occur search: ' + str(occur_search))
# for country in countries:
# print(country)
# print(occur_search)
# Runs if species has occurred more than zero times
if occurs > 0:
try:
# Tries to retrieve scientific name
canon_name = o['canonicalName']
except:
continue
# Vernacular name init
vern_name = ''
# Variable for list of vernacular names
names = o['vernacularNames']
# Summary init
summary = ''
try:
# If match found
print('Scientific name: ' + canon_name)
print('Vernacular names: ')
match_found = False
# Reads from results in matched name
for name in names:
# Variable for vernacular name
vern_name = name['vernacularName']
# Used if all languages want to be included
# print(name['vernacularName'])
language = (name['language'])
# Can be changed if user wants to select a specific language
if language == 'eng':
# Checks if vernacular name is matched with search input
if name_input in vern_name and match_found is False:
match_found = True
print(vern_name)
# If no exact match is found it reads the first vernacular name
if match_found is False:
name_store = []
for name in names:
vern_name = name['vernacularName']
language = (name['language'])
if vern_name not in name_store:
name_store.append(vern_name)
if language == 'eng':
print(vern_name)
# Adds scientific and vernacular name to results dictionary
results.setdefault(canon_name, []).append(vern_name)
# print(wikipedia.search(canon_name))
try:
# pulls the wiki page based on canonical name in url (usually works)
# desc = wikipedia.page(canon_name, auto_suggest=True)
# alternative wiki page including all sections
# page = wikipedia.WikipediaPage(canon_name)
# pulls the summary page for the species
summary = wikipedia.summary(canon_name, sentences=2)
# experimental for section pages
# sections = page.sections
# for section in sections:
# print(section)
# github push error, delete this
print(summary)
results.setdefault(canon_name, []).append(summary)
# print(desc.content)
except:
print("No description found")
print('GBIF Key: ' + str(key))
print('GBIF Species Page: ' + 'http://www.gbif.org/species/' + str(key))
print('Count: ' + str(occurs))
# This reads results for occurences and finds countries the species
# was observed in and how many occurrences there
occur_data = occur_search['results']
countries = {}
for occur in occur_data:
try:
country = occur['country']
if country not in countries:
countries[country] = 1
else:
countries[country] += 1
# print(occur['country'])
except:
continue
# for country in countries:
# print(country)
sorted_countries = sorted(countries.items(), key=lambda x: x[1], reverse=True)
if countries != {}:
print('Top 3 Countries Observed: ')
for country in sorted_countries[:3]:
print(str(country))
print('\n')
# return canon_name, vern_name, summary
except:
continue
print(results)
return results
def test_name_suggest():
"species.name_suggest - basic test"
res = species.name_suggest(q="Puma concolor")
assert list == res.__class__
assert True == all(
[bool(re.search("Puma concolor", z["canonicalName"])) for z in res])
def test_name_suggest_paging():
"species.name_suggest - paging"
res = species.name_suggest(q="Aso", limit=3)
assert list == res.__class__
assert 3 == len(res)
def search_taxa(q):
res = pd.DataFrame.from_dict(species.name_suggest(q))
return res