demo of hacked-together script for creating a Mendeley collection from a set of PMIDs. uses Mendeley API

make_mendeley_collection_from_PMIDs.py

#! /usr/bin/python
 
# Requires Python >2.7.2 
 
# CAUTION: this script will create a new collection in your Mendeley library and add a bunch of documents to it.
 
import json
 
import urlparse
from mendeley_client import MendeleyClient
from Bio import Entrez, Medline
 
 
def update_uploaded_list(mendeley, group_id):
    """helper function to update the list of documents already in the group"""
    docs = mendeley.group_documents(group_id,items='5000')
    uploaded_PMIDs = []
    for doc_id in docs['document_ids']:
        doc_details = mendeley.group_doc_details(group_id,doc_id)
        if ('pmid' in doc_details['identifiers']):
            uploaded_PMIDs.append(doc_details['identifiers']['pmid'])
    return uploaded_PMIDs
 
consumer_key = 'XXXXXXXXXXXXXXXX'
consumer_secret = 'XXXXXXXXXXXXXXXX'
 
mendeley = MendeleyClient(consumer_key, consumer_secret)
 
try:
    mendeley.load_keys()
except IOError:
    mendeley.get_required_keys()
    mendeley.save_keys()
 
# check to see if there is already a group w/ 'ABCD' in the title
abcd_group_id = []
groups = mendeley.groups()
for gr in groups:
    if 'abcd' in gr['name'].lower():
        abcd_group_id = gr['id']
 
if not abcd_group_id:
    # create public invite-only group for ABCD
    response = mendeley.create_group(group=json.dumps({'name':'ABCD', 'type': 'invite'}))
    abcd_group_id = response['group_id']
 
# load ABCD dump
with open("abcd_connections.json","rb") as f:
    connections = json.load(f)
 
PMIDs = []
ASINs = []
# for each citation in the ABCD connection dump
for cnxn_id, cnxn_info in connections.iteritems():
    if 'citations' in cnxn_info:
        for cite in cnxn_info['citations']:
            if 'PMID' in cite.keys():
                if not cite['PMID'] in PMIDs:
                    PMIDs.append(cite['PMID'])
            elif 'ASIN' in cite.keys():
                if not cite['ASIN'] in ASINs:
                    ASINs.append(cite['ASIN'])
 
print "%s PMIDs referenced in the ABCD" % len(PMIDs)
 
print "updating the list of uploaded PMIDS..."
uploaded_PMIDs = update_uploaded_list(mendeley,abcd_group_id)
print "%s PMIDs already uploaded" % len(uploaded_PMIDs)
 
for identifier in PMIDs:
    #   if it exists in the mendeley group, then skip it
    if identifier in uploaded_PMIDs:
        print "skip %s: %s" % ('pmid', identifier)
    else:
        print "adding %s: %s" % ('pmid', identifier)
        document = mendeley.details(identifier, type='pmid')
        if 'error' in document:
            print "Error getting document: %s" % document['error']
            print "since PMID not found in Mendley, extracting basic info on %(identifier)s from pubmed"
            handle = Entrez.efetch(db="pubmed", id=identifier, rettype="medline", retmode="text")
            records = Medline.parse(handle)
            records = list(records)
             
            document = {'title': records[0]['TI'],
                        'authors': records[0]['FAU'],
                        'abstract': records[0]['AB'],
                        'identifiers': {'pmid': identifier},
                        'type': 'Journal Article',
                         
                        }
 
        # strip extra fields to prep to add to group
        fields_to_remove = ['error', 
                            'mendeley_url', 
                            'stats', 
                            'uuid', 
                            'publication_outlet', 
                            'categories',
                            'groups',]
        for field in fields_to_remove:
            if field in document:
                del document[field]
        document['group_id'] = abcd_group_id
 
        # add the citation to the group
        response = mendeley.create_document(document=json.dumps(document))
        if 'error' in response:
            print "Error adding document: %s" % response['error']
        else:
            uploaded_PMIDs.append(identifier)
        # #update the doc list
        # print "updating the list of uploaded PMIDS..."
        # uploaded_PMIDs = update_uploaded_list(mendeley,abcd_group_id)
        # print "%s PMIDs already uploaded" % len(uploaded_PMIDs)