def generate_booktitle(record):
booktitle = ''
pubinfo = ''
if 'publication_info' in record:
pubinfo = record['publication_info']
for field in pubinfo:
if 'reportnumber' in field:
rn = field['reportnumber']
if rn:
acronym = field['acronym']
if acronym:
booktitle = "%s: %s" % (rn, acronym, )
else:
records = perform_es_search(
"reportnumber:%s" % (rn,),
"records-hep"
)
if records:
rec = records.hits[0]
for title in rec['titles']:
booktitle = title.get('title', "")
if title.get('subtitle'):
booktitle += ': ' + title.get('subtitle')
if not booktitle:
result = []
for field in pubinfo:
if 'pubinfo_freetext' in field:
result.append(field['pubinfo_freetext'])
if result:
if any(isinstance(i, list) for i in result):
nested_list = list(traverse(result))
booktitle = ', '.join(str(title) for title in nested_list)
else:
booktitle = ', '.join(str(title) for title in result)
return booktitle
def proceedings_link(record):
cnum = record.get('cnum', '')
out = ''
if not cnum:
return out
records = perform_es_search(
'cnum:%s and 980__a:proceedings' % cnum, 'records-hep')
if len(records):
if len(records) > 1:
proceedings = []
for i, record in enumerate(records.hits, start=1):
try:
dois = record['dois']
proceedings.append(
'<a href="/record/{recid}">#{i}</a> (DOI: <a '
'href="http://dx.doi.org/{doi}">{doi}</a>'.format(
recid=record['control_number'],
doi=dois[0]['value'], i=i))
except KeyError:
# Guards both against records not having a "dois" field
# and doi values not having a "value" field.
proceedings.append(
'<a href="/record/{recid}">#{i}</a>'.format(
recid=record['control_number'], i=i))
out = 'Proceedings: '
out += ', '.join(proceedings)
else:
out += '<a href="/record/{recid}">Proceedings</a>'.format(
recid=records[0]['control_number'])
return out
def experiment_people_from_es(experiment_name):
"""Query ES for conferences in the same series."""
query = 'experiments.name:"{}"'.format(experiment_name)
return perform_es_search(
query,
'records-authors'
).hits
def institutions():
"""View for institutions collection landing page."""
institutions_list = perform_es_search('', 1, 250, 'institutions')
return render_template(
'inspirehep_theme/search/collection_institutions.html',
ctx={'results': institutions_list},
collection='institutions')
def _get_upcoming_conferences():
today = date.today()
in_six_months = today + relativedelta(months=+6)
upcoming_conferences = perform_es_search(
'opening_date:{0}->{1}'.format(str(today), str(in_six_months)),
'records-conferences', 1, 100, {'opening_date': 'asc'})
return [hit['_source'] for hit in upcoming_conferences.to_dict()['hits']['hits']]
def institutions():
"""View for institutions collection landing page."""
institutions_list = perform_es_search('', 'records-institutions', 0, 250)
results = [hit['_source'] for hit in institutions_list.to_dict()['hits']['hits']]
return render_template(
'inspirehep_theme/search/collection_institutions.html',
ctx={'results': results}, collection='institutions')
def conferences_in_the_same_series_from_es(seriesname):
"""Query ES for conferences in the same series."""
query = 'series:"{}"'.format(seriesname)
return perform_es_search(query, 'records-conferences',
sort="-opening_date",
fields=[
'control_number',
'titles',
'address',
'opening_date',
'closing_date'
]
).hits
def link_to_hep_affiliation(record):
try:
icn = record['ICN']
except KeyError:
return ''
records = perform_es_search('affiliation:%s' % icn, 'records-institutions')
results = records.hits.total
if results:
if results == 1:
return str(results) + ' Paper from ' + \
str(record['ICN'])
else:
return str(results) + ' Papers from ' + \
str(record['ICN'])
else:
return ''
def experiment_contributions_from_es(experiment_name):
"""Query ES for conferences in the same series."""
query = 'accelerator_experiments.experiment:"{}"'.format(experiment_name)
return perform_es_search(
query,
'records-hep',
sort='-citation_count',
size=100,
fields=[
'control_number',
'earliest_date',
'titles',
'authors',
'publication_info',
'citation_count',
'collaboration'
]
).hits
def conferences():
"""View for conferences collection landing page."""
today = datetime.today()
today_str = today.strftime('%Y-%m-%d')
six_months_str = (today + relativedelta(months=+6)).strftime('%Y-%m-%d')
upcoming_conferences = perform_es_search(
'opening_date:{0}->{1}'.format(today_str, six_months_str), 1, 100,
'conferences', 'opening_date:asc')
return render_template(
'inspirehep_theme/search/collection_conferences.html',
ctx={
'conference_subject_areas': CONFERENCE_CATEGORIES_TO_SERIES,
'results': upcoming_conferences
},
collection='conferences')
def conferences():
"""View for conferences collection landing page."""
today = datetime.today()
today_str = today.strftime('%Y-%m-%d')
six_months_str = (today + relativedelta(months=+6)).strftime('%Y-%m-%d')
upcoming_conferences = perform_es_search(
'opening_date:{0}->{1}'.format(today_str, six_months_str),
'records-conferences', 1, 100, {'opening_date': 'asc'})
results = [hit['_source'] for hit in upcoming_conferences.to_dict()['hits']['hits']]
return render_template(
'inspirehep_theme/search/collection_conferences.html',
ctx={'conference_subject_areas': CONFERENCE_CATEGORIES_TO_SERIES,
'results': results},
collection='conferences')
def number_of_search_results(query, collection_name):
"""
Filter used to show total number of results out of filtered ones.
"""
session_key = 'last-query' + query + collection_name
if session.get(session_key):
query_timestamp = session[session_key]['timestamp']
seconds_since_query = (
datetime.datetime.utcnow() - query_timestamp).total_seconds()
if seconds_since_query < 300:
# Only use the session value if it is newer than 5 minutes
# This should allow for the common use case of navigating
# facets and avoid using an outdated value when using a direct
# link
return session[session_key][
"number_of_hits"
]
index = collection_to_index(collection_name)
return perform_es_search(query, index).hits.total
def get_institution_papers_from_es(recid):
"""
Get papers where some author is affiliated with institution.
:param recid: id of the institution.
:type recid: string
"""
query = "authors.affiliations.recid:{}".format(recid)
return perform_es_search(
query,
'records-hep',
sort='-earliest_date',
size=100,
fields=[
'control_number',
'earliest_date',
'titles',
'authors',
'publication_info',
'citation_count',
'collaboration'
]
).hits
def get_institution_people_datatables_rows(recid):
"""
Datatable rows to render people working in an institution.
:param recid: id of the institution.
:type recid: string
"""
query = {
"query": {
"term": {
"authors.affiliations.recid": recid
}
},
"aggs": {
"authors": {
"nested": {
"path": "authors"
},
"aggs": {
"affiliated": {
"filter": {
"term": {"authors.affiliations.recid": recid}
},
"aggs": {
"byrecid": {
"terms": {
"field": "authors.recid"
}
}
}
}
}
}
}
}
records_from_es = current_search_client.search(
index='records-hep',
doc_type='hep',
body=query,
search_type='count'
)
# Extract all the record ids from the aggregation
papers_per_author = records_from_es[
'aggregations'
]['authors']['affiliated']['byrecid']['buckets']
recids = [int(paper['key']) for paper in papers_per_author]
# Generate query to retrieve records from author index
query = ""
for i, recid in enumerate(recids):
query += "recid:{}".format(recid)
if i != len(recids) - 1:
query += " OR "
results = perform_es_search(
query, 'records-authors', size=9999, fields=['control_number', 'name']
)
recid_map = dict(
[(int(result.control_number), result.name) for result in results]
)
result = []
author_html_link = "<a href='/authors/{recid}'>{name}</a>"
for author in papers_per_author:
row = []
try:
row.append(
author_html_link.format(
recid=author['key'],
name=recid_map[author['key']].preferred_name
)
)
except:
# No preferred name, use value
row.append(
author_html_link.format(
recid=author['key'],
name=recid_map[author['key']].value
)
)
row.append(author['doc_count'])
result.append(row)
return result
def _get_some_institutions():
some_institutions = perform_es_search('', 'records-institutions', 0, 250)
return [hit['_source'] for hit in some_institutions.to_dict()['hits']['hits']]
def number_of_records(collection_name):
"""Returns number of records for the collection."""
index = collection_to_index(collection_name)
return perform_es_search('', index).hits.total
