from neo4jrestclient.client import GraphDatabase
import time
from Bio import Entrez
db = GraphDatabase("http://localhost:7474", username="neo4j", password="mypassword")



def search(query):
    Entrez.email = 'your.email@example.com'
    handle = Entrez.esearch(db='pubmed',
                            sort='relevance',
                            retmax='20000000',
                            retmode='xml',
                            term=query)
    results = Entrez.read(handle)
    return results

def fetch_details(id_list):
    ids = ','.join(id_list)
    Entrez.email = 'your.email@example.com'
    handle = Entrez.efetch(db='pubmed',
                           retmode='xml',
                           id=ids)
    results = Entrez.read(handle)
    return results

if __name__ == '__main__':
    results = search('1900[PDAT]:2020[PDAT],nature[JOURNAL]')
    id_list = results['IdList']
    print(len(id_list))

    if (len(id_list)>0):
        papers = fetch_details(id_list)
        print(papers)
        for i, paper in enumerate(papers['PubmedArticle']):
            #the api lets you import 10,000 record each time, so i made a sleep function for 3 minutes each iteration
            if(i%9995==0):
                time.sleep(3*60)

            print("%d) %s" % (i + 1, paper['MedlineCitation']['Article']['ArticleTitle']))
            paperDate = paper['MedlineCitation']['Article']['ArticleDate']
            if(len(paperDate)==0):
                try:
                    print(paper['MedlineCitation']['Article']['Journal']['JournalIssue']['PubDate']['Year'])
                except KeyError:
                    print("No Journal Year")
            else:
                for paperYear in enumerate(paperDate):
                    print("%d) %s" % (i + 1, paperYear[1]['Year']))

            try:
                authorList = paper['MedlineCitation']['Article']['AuthorList']
            except KeyError:
                print("No Artist List")
                continue

            for author in authorList:
                try:
                    print ("%d) %s" % (i + 1, "LastName: %s , Forename: %s , Initials: %s" % (
                    author['LastName'], author['ForeName'], author['Initials'])))
                    try:
                        print("Affiliation: %s" % (author['AffiliationInfo'][0]['Affiliation']))
                    except (IndexError,KeyError):
                        print("No Affiliation")

                except KeyError:
                    print("No Author name")
    else:
        print("No ID")

# Create some nodes
paper = db.node.create("Paper")
author = db.node.create("Author")
# Author-Wrote->Paper relationships
author.relationships.create("Wrote", paper)