def queryAltmetric(pmid):
"""
Check the altmetric journal percentile score of the publication
"""
a = Altmetric()
try:
resp = a.pmid(pmid)
if resp is None:
logger.debug("PMID %s. Not found" % pmid)
return -1
else:
if 'context' in resp:
metric = resp['context']['journal'][
'pct'] # Percentage attention for this journal
logger.debug("PMID %s. Metric %s" % (pmid, metric))
return metric
logger.debug("PMID %s. Percentage attention not found" % pmid)
return -2
except AltmetricHTTPException as e:
if e.status_code == 403:
logger.error(
"You aren't authorized for this call: {}".format(pmid))
elif e.status_code == 420:
logger.error(
'You are being rate limited, currently {}'.format(pmid))
elif e.status_code == 502:
logger.error(
'The API version you are using is currently down for maintenance.'
)
elif e.status_code == 404:
logger.error('Invalid API function')
logger.error(e.msg)
logger.warn("PMID %s. Exception %s" % (pmid, e.msg))
return -3
def bst_openaire_altmetric():
"""
"""
recids = search_pattern(p="0->Z", f="0247_a")
a = Altmetric()
for recid in recids:
try:
# Check if we already have an Altmetric id
sysno_inst = get_fieldvalues(recid, "035__9")
if ['Altmetric'] in sysno_inst:
continue
doi_val = get_fieldvalues(recid, "0247_a")[0]
json_res = a.doi(doi_val)
rec = {}
record_add_field(rec, "001", controlfield_value=str(recid))
if json_res:
record_add_field(rec,
'035',
subfields=[('a',
str(json_res['altmetric_id'])),
('9', 'Altmetric')])
bibupload(rec, opt_mode='correct')
except AltmetricHTTPException, e:
register_exception(prefix='Altmetric error (status code %s): %s' %
(e.status_code, str(e)),
alert_admin=False)
def bst_openaire_altmetric():
"""
"""
recids = search_pattern(p="0->Z", f="0247_a")
a = Altmetric()
for recid in recids:
try:
# Check if we already have an Altmetric id
sysno_inst = get_fieldvalues(recid, "035__9")
if ['Altmetric'] in sysno_inst:
continue
doi_val = get_fieldvalues(recid, "0247_a")[0]
json_res = a.doi(doi_val)
rec = {}
record_add_field(rec, "001", controlfield_value=str(recid))
if json_res:
record_add_field(rec, '035', subfields=[('a',
str(json_res['altmetric_id'])), ('9', 'Altmetric')])
bibupload(rec, opt_mode='correct')
except AltmetricHTTPException, e:
register_exception(prefix='Altmetric error (status code %s): %s' %
(e.status_code, str(e)), alert_admin=False)
def openaire_altmetric_update(recids, upload=True):
"""
Retrieve Altmetric information for a record.
"""
logger.debug("Checking Altmetric for recids %s" % recids)
a = Altmetric()
records = []
for recid in recids:
logger.debug("Checking Altmetric for recid %s" % recid)
try:
# Check if we already have an Altmetric id
sysno_inst = get_fieldvalues(recid, "035__9")
if ['Altmetric'] in sysno_inst:
continue
doi_val = get_fieldvalues(recid, "0247_a")[0]
logger.debug("Found DOI %s" % doi_val)
json_res = a.doi(doi_val)
logger.debug("Altmetric response: %s" % json_res)
rec = {}
record_add_field(rec, "001", controlfield_value=str(recid))
if json_res:
record_add_field(rec,
'035',
subfields=[('a',
str(json_res['altmetric_id'])),
('9', 'Altmetric')])
records.append(rec)
except AltmetricHTTPException as e:
logger.warning(
'Altmetric error for recid %s with DOI %s (status code %s): %s'
% (recid, doi_val, e.status_code, str(e)))
register_exception(prefix='Altmetric error (status code %s): %s' %
(e.status_code, str(e)),
alert_admin=False)
except IndexError:
logger.debug("No DOI found")
pass
if upload and records:
if len(records) == 1:
bibupload(record=records[0], file_prefix="altmetric")
else:
bibupload(collection=records, file_prefix="altmetric")
return records
def getAltScore(inputDF):
"""This function reads in an already existing dataframe, typically the one
obtained from the searchPubMed function. It returns said dataframe with an
additional column, giving the Altmetric score for a given article DOI in the
table according to Altmetrics.
Parameters:
inputDF(Pandas.DataFrame): this is a Pandas dataframe with PubMed ID as indices,
and a column labelled "DOI" with all the DOIs for the associated PubMed ID.
Returns:
Pandas.DataFrame, identical to input with addition of column "Altmetric Score,"
giving the Altmetric score, calculated by Altmetrics, for a given article.
Example:
queryDF = getAltScore(queryDF)"""
#First gather the DOI for each article in the dataframe.
doiList = inputDF["DOI"].tolist()
listAltScore = []
count = 1
print("Getting Altmetrics Score")
#Now for every DOI, use CrossRef to find the associated Altmetrics score.
for entry in doiList:
print(count)
try:
if entry != "":
a = Altmetric()
thisDOI = a.doi(entry)
extractScore = thisDOI["score"]
listAltScore.append(extractScore)
else:
listAltScore.append(None)
except:
listAltScore.append(None)
count = count + 1
#Now add a column of Altmetrics scores to the dataframe.
inputDF["Altmetric Score"] = listAltScore
print("Altmetrics Search Complete!")
return inputDF
def main():
try:
args = get_args()
twitter_api = setup_tweeting()
preprints = load_database()
shuffle(preprints)
altmetric_api = Altmetric(apikey=secrets.altmetric_key)
for preprint in preprints:
if preprint.status == 'tweeted':
continue
return_code, res = preprint.query_altmetric(altmetric_api)
if return_code:
my_logger.error(res)
else:
if res >= 90:
if not args.dry:
preprint.tweet(twitter_api)
else:
preprint.dry_print()
sleep(5)
clean_database(preprints)
my_logger.info('Finished.\n')
except Exception as e:
my_logger.error(e, exc_info=True)
print("Terminating...")
clean_database(preprints)
raise
def openaire_altmetric_update(recids, upload=True):
"""
Retrieve Altmetric information for a record.
"""
logger.debug("Checking Altmetric for recids %s" % recids)
a = Altmetric()
records = []
for recid in recids:
logger.debug("Checking Altmetric for recid %s" % recid)
try:
# Check if we already have an Altmetric id
sysno_inst = get_fieldvalues(recid, "035__9")
if ["Altmetric"] in sysno_inst:
continue
doi_val = get_fieldvalues(recid, "0247_a")[0]
logger.debug("Found DOI %s" % doi_val)
json_res = a.doi(doi_val)
logger.debug("Altmetric response: %s" % json_res)
rec = {}
record_add_field(rec, "001", controlfield_value=str(recid))
if json_res:
record_add_field(rec, "035", subfields=[("a", str(json_res["altmetric_id"])), ("9", "Altmetric")])
records.append(rec)
except AltmetricHTTPException as e:
logger.warning(
"Altmetric error for recid %s with DOI %s (status code %s): %s"
% (recid, doi_val, e.status_code, str(e))
)
register_exception(
prefix="Altmetric error (status code %s): %s" % (e.status_code, str(e)), alert_admin=False
)
except IndexError:
logger.debug("No DOI found")
pass
if upload and records:
if len(records) == 1:
bibupload(record=records[0], file_prefix="altmetric")
else:
bibupload(collection=records, file_prefix="altmetric")
return records
def openaire_altmetric_update(recids, upload=True):
"""
Retrieve Altmetric information for a record.
"""
logger.debug("Checking Altmetric for recids %s" % recids)
a = Altmetric()
records = []
for recid in recids:
logger.debug("Checking Altmetric for recid %s" % recid)
try:
# Check if we already have an Altmetric id
sysno_inst = get_fieldvalues(recid, "035__9")
if ['Altmetric'] in sysno_inst:
continue
doi_val = get_fieldvalues(recid, "0247_a")[0]
logger.debug("Found DOI %s" % doi_val)
json_res = a.doi(doi_val)
logger.debug("Altmetric response: %s" % json_res)
rec = {}
record_add_field(rec, "001", controlfield_value=str(recid))
if json_res:
record_add_field(rec, '035', subfields=[
('a', str(json_res['altmetric_id'])),
('9', 'Altmetric')
])
records.append(rec)
except AltmetricHTTPException, e:
logger.warning(
'Altmetric error for recid %s with DOI %s (status code %s): %s'
% (recid, doi_val, e.status_code, str(e))
)
register_exception(
prefix='Altmetric error (status code %s): %s' % (
e.status_code, str(e)),
alert_admin=False
)
except IndexError:
logger.debug("No DOI found")
pass
s = ',' # separator
if os.path.isfile(outputFileName):
outputFile_r = open(outputFileName, 'r')
for line in outputFile_r:
listpmid_alreadyDone.append(line.split(',')[0])
outputFile_r.close()
outputFile = open(outputFileName, 'a')
else:
outputFile = open(outputFileName, 'w')
outputFile.write('pmid, score, numReaders, cited (altmetric), scopus\n')
am_metadata = Altmetric()
c = 0
for line in setLinks:
if c > numLines2Skip:
parsedData = line.split('\n')[0].split(',')
pmid = parsedData[2]
if not (pmid in listpmid_alreadyDone):
paperMetadata = am_metadata.pmid(pmid)
if paperMetadata != None:
def update(self):
rsp = None
a = Altmetric(settings.ALTMETRIC_API_KEY)
ids = self.paper.get_ids()
# Fetch altmetric data based on paper supported id
try:
if 'doi' in ids:
rsp = a.doi(ids.get('doi', ''))
if not rsp and 'arx' in ids:
rsp = a.arxiv(ids.get('arx'))
if not rsp and 'pmi' in ids:
rsp = a.pmid(ids.get('pmi'))
except AltmetricHTTPException:
raise
# Parse json
if rsp:
self.score = rsp.get('score')
self.altmetric_id = rsp.get('altmetric_id')
self.altmetric_jid = rsp.get('altmetric_jid', '')
self.score_1d = rsp.get('history').get('1d')
self.score_2d = rsp.get('history').get('2d')
self.score_3d = rsp.get('history').get('3d')
self.score_4d = rsp.get('history').get('4d')
self.score_5d = rsp.get('history').get('5d')
self.score_6d = rsp.get('history').get('6d')
self.score_1w = rsp.get('history').get('1w')
self.score_1m = rsp.get('history').get('1m')
self.score_3m = rsp.get('history').get('3m')
self.score_6m = rsp.get('history').get('6m')
self.score_1y = rsp.get('history').get('1y')
self.cited_by_posts_count = rsp.get('cited_by_posts_count', 0)
self.cited_by_delicious_count = rsp.get('cited_by_delicious_count',
0)
self.cited_by_fbwalls_count = rsp.get('cited_by_fbwalss_count', 0)
self.cited_by_feeds_count = rsp.get('cited_by_feeds_count', 0)
self.cited_by_forum_count = rsp.get('cited_by_forum_count', 0)
self.cited_by_gplus_count = rsp.get('cited_by_gplus_count', 0)
self.cited_by_linkedin_count = rsp.get('cited_by_linkedin_count',
0)
self.cited_by_msm_count = rsp.get('cited_by_msm_count', 0)
self.cited_by_peer_review_sites_count = rsp.get(
'cited_by_peer_review_sites_count', 0)
self.cited_by_pinners_count = rsp.get('cited_by_pinners_count', 0)
self.cited_by_policies_count = rsp.get('cited_by_policies_count',
0)
self.cited_by_qs_count = rsp.get('cited_by_qs_count', 0)
self.cited_by_rdts_count = rsp.get('cited_by_rdts_count', 0)
self.cited_by_rh_count = rsp.get('cited_by_rh_count', 0)
self.cited_by_tweeters_count = rsp.get('cited_by_tweeters_count',
0)
self.cited_by_videos_count = rsp.get('cited_by_videos_count', 0)
self.cited_by_weibo_count = rsp.get('cited_by_weibo_count', 0)
self.cited_by_wikipedia_count = rsp.get('cited_by_wikipedia_count',
0)
if isinstance(rsp.get('readers'), dict):
self.readers_citeulike = rsp.get('readers').get('citeulike')
self.readers_mendeley = rsp.get('readers').get('mendeley')
self.image = rsp.get('images').get('small')
match = re.findall(r'types=(P?[\w]+)', self.image)
if match:
self.type = match[0]
# save
self.save()
metrics_columns = [
"pmid", "doi", "pub_year", "altmetric_id", "altmetric_queried",
"paperbuzz_queried"
]
altmetric_fields = config['altmetric_fields']
paperbuzz_fields = config['paperbuzz_fields']
am_metrics = sorted(altmetric_fields.keys())
paperbuzz_metrics = sorted(paperbuzz_fields.keys())
metrics_columns = metrics_columns + am_metrics + paperbuzz_metrics
# Initialize APIs
ALTMETRIC_KEY = config['keys']['altmetric']
altmetric = Altmetric(ALTMETRIC_KEY)
paperbuzz = PaperbuzzAPI(config['contact']['email'])
logging.info('Iterating over input files')
for name, input_file in config['input_files'].items():
logging.info('Processing {}: {}'.format(name, input_file))
# Load input file
input_df = pd.read_csv(data_dir / input_file)
output_dir = base_dir / name
if not os.path.exists(str(output_dir)):
os.makedirs(str(output_dir))
# # Query Paperbuzz API
# logging.info('Querying Paperbuzz API')
# Output columns for results
METRICS_COLUMNS = ["pmid", "doi", "resp", "err", "ts"]
# Init logging
logging.basicConfig(level=logging.INFO,
format="%(asctime)s - %(message)s")
# Init
logging.info('Initialise configuration')
data_dir = Path("../data/")
config = load_config("../config.yml")
output_dir = select_basedir(data_dir)
# Initialize APIs
ALTMETRIC_KEY = config['keys']['altmetric']
altmetric = Altmetric(ALTMETRIC_KEY)
paperbuzz = PaperbuzzAPI(config['contact']['email'])
logging.info('Iterating over each query')
for idx, query in enumerate(config['queries'].keys()):
input_df = pd.read_csv(output_dir / query / "articles.csv")
input_df = input_df[input_df.error.isna()]
# Query Paperbuzz API
logging.info('Collecting Paperbuzz metrics for {}'.format(query))
with open(str(output_dir / query / "paperbuzz.csv"), "w") as f:
csv_writer = csv.writer(f)
csv_writer.writerow(METRICS_COLUMNS)
for pmid, doi in tqdm(zip(input_df.pmid, input_df.doi), total=len(input_df)):
now = datetime.now()
