def get_data_from_doi(self, doi):
doi_doc = FullDoc(doi=doi)
if doi_doc.read(self.client):
# print("doi_doc.title: ", doi_doc.title)
doi_doc.write()
else:
print("Read document failed.")
return doi
id = None
if not 'pubmed-id' in doi_doc._data.keys():
print("no pubmed-id, trying with title")
# try with title
Entrez.email = '*****@*****.**'
query = doi_doc.title
handle = Entrez.esearch(db='pubmed', retmode='xml', term=query)
results = Entrez.read(handle)
if int(results['Count']) > 0:
id = results['IdList']
else:
id = doi_doc._data['pubmed-id']
if id != None:
return self.fetch_data_from_pubmed(id)
else:
print("no pubmed id")
return doi
def get_paper(doi: str):
## ScienceDirect (full-text) document example using DOI
doi_doc = FullDoc(doi=doi)
if doi_doc.read(client):
return doi_doc.data
else:
return False
def ElsevierScraper(client, target_DOI):
"""
Uses the Elsevier API with a valid key and a DOI to download the plain text article, including the title, abstract,
pub date, and the references as an unstructured string.
:param client: Elsevier client containing the API key
:param target_DOI: DOI of the article being scraped
:return: list containing the DOI, title, abstract, publication date, full text and unstructured string of references
"""
print(target_DOI)
doi_doc = FullDoc(doi=target_DOI)
if doi_doc.read(client):
data = doi_doc.data
coreData = data['coredata']
abstract = coreData['dc:description']
text = str(data['originalText']).split(abstract)[-1]
try:
references = text.split("References")[1]
except IndexError:
references = "NA"
text = text.split("References")[0]
date = coreData['prism:coverDisplayDate']
return [target_DOI, doi_doc.title, abstract, date, text, references]
else:
# Could save this to a separate file but it's easier to search for NA and make a new sublist after the fact
print("Error: couldn't read {}.".format(target_DOI))
return [target_DOI, "NA", "NA", "NA", "NA", "NA"]
def readFullDocWithDOI(self, doiID='10.1016/S1525-1578(10)60571-5'):
## ScienceDirect (full-text) document example using DOI
doi_doc = FullDoc(doi=doiID)
if doi_doc.read(self.client):
print("doi_doc.title: ", doi_doc.title)
doi_doc.write()
else:
print("Read document failed.")
def readFullDocWithPII(self, sd_piiID='S1270963817323015'):
## ScienceDirect (full-text) document example using PII
pii_doc = FullDoc(sd_pii=sd_piiID)
if pii_doc.read(self.client):
print(pii_doc)
print("pii_doc.title: ", pii_doc.title)
pii_doc.write()
else:
print("Read document failed.")
def __init__(self, *args):
"""
"""
print("PII : ", args[0])
self._sd_article = FullDoc(sd_pii=args[0])
print("init SD 1")
if not self._sd_article.read(els_client=args[1]):
print("raise HTTPError")
raise HTTPError
def find_abstract(doi):
client = elsevier_auth()
## ScienceDirect (full-text) document example using DOI
doi_doc = FullDoc(doi=doi)
if doi_doc.read(client):
print("doi_doc.title: ", doi_doc.title)
print("doi_doc.abstract: ", doi_doc.data['coredata']['dc:description'])
doi_doc.write()
else:
print("Read document failed.")
class TestFullDoc:
"""Test ScienceDirect article functionality"""
## Test data
full_pii_uri = "https://api.elsevier.com/content/article/pii/S1674927814000082"
sd_pii = 'S1674927814000082'
full_doi_uri = "https://api.elsevier.com/content/article/doi/10.1016/S1525-1578(10)60571-5"
doi = '10.1016/S1525-1578(10)60571-5'
## Test initialization
def test_init_uri(self):
""" Test case: uri is set correctly during initialization with uri"""
myFullDoc = FullDoc(uri = self.full_pii_uri)
assert myFullDoc.uri == self.full_pii_uri
def test_init_sd_pii(self):
""" Test case: uri is set correctly during initialization with ScienceDirect PII"""
myFullDoc = FullDoc(sd_pii = self.sd_pii)
assert myFullDoc.uri == self.full_pii_uri
def test_init_doi(self):
""" Test case: uri is set correctly during initialization with DOI"""
myFullDoc = FullDoc(doi = self.doi)
assert myFullDoc.uri == self.full_doi_uri
## Test reading/writing author profile data
bad_client = ElsClient("dummy")
good_client = ElsClient(config['apikey'], inst_token = config['insttoken'])
good_client.local_dir = str(test_path)
myFullDoc = FullDoc(uri = full_pii_uri)
def test_read_good_bad_client(self):
"""Test case: using a well-configured client leads to successful read
and using a badly-configured client does not."""
assert self.myFullDoc.read(self.bad_client) == False
assert self.myFullDoc.read(self.good_client) == True
def test_json_to_dict(self):
"""Test case: the JSON read by the full article object from the
API is parsed into a Python dictionary"""
assert type(self.myFullDoc.data) == dict
def test_title_getter(self):
"""Test case: the title attribute is returned as a non-empty string"""
assert (type(self.myFullDoc.title) == str and self.myFullDoc.title != '')
def test_write(self):
"""Test case: the full article object's data is written to a file with the ID in the filename"""
self.myFullDoc.write()
## TODO: replace following (strung-together replace) with regex
assert util.file_exist_with_id(
self.myFullDoc.data['coredata']['pii'].replace('-','').replace('(','').replace(')',''))
def get_data_from_doi(self, doi, title):
id = None
affil = None
pub_name = None
pub_type = None
# try:
try:
doi_doc = ScopusSearch(doi, subscriber=False)
if 'pubmed-id' in doi_doc._json[0].keys():
id = doi_doc._json[0]["pubmed-id"]
if 'affiliation' in doi_doc._json[0].keys():
affil = doi_doc._json[0]['affiliation']
pub_name = doi_doc._json[0]['prism:publicationName']
pub_type = doi_doc._json[0]['subtypeDescription']
except:
print("failed with scopus")
if id == None:
doi_doc = FullDoc(doi=doi)
if doi_doc.read(self.client):
# print("doi_doc.title: ", doi_doc.title)
doi_doc.write()
pub_name = doi_doc.data['coredata']['prism:publicationName']
if 'pubType' in doi_doc.data['coredata'].keys():
pub_type = str(doi_doc.data['coredata']['pubType']).strip()
else:
print(
"Read document failed. no id for doi {}. trying with title"
.format(doi))
doi_doc = None
# return doi, affil
id = None
if doi_doc == None or (not 'pubmed-id' in doi_doc._data.keys()):
print("trying with title")
# try with title
Entrez.email = '*****@*****.**'
if doi_doc == None:
query = title
else:
query = doi_doc.title
handle = Entrez.esearch(db='pubmed', retmode='xml', term=query)
results = Entrez.read(handle)
if int(results['Count']) > 0:
id = results['IdList']
else:
id = doi_doc._data['pubmed-id']
if id != None:
return self.fetch_data_from_pubmed(id), affil, pub_name, pub_type
else:
print("no pubmed id found for doi {}".format(doi))
return doi, affil, pub_name, pub_type
def get_doc(self, dtype, identity):
"""
This method retrieves a 'Doc' object from the Elsevier API. The doc object contains metadata and full-text information
about a publication associated with a given PII.
Parameters:
-----------
dtype(str,required): The type of identification string being used to access the document. (Almost always PII in our case.)
identity: The actual identification string/ PII that will be used to query.
"""
if dtype == 'pii':
doc = FullDoc(sd_pii = identity)
elif dtype == 'doi':
doc= FullDoc(doi = identity)
if doc.read(ElsClient(self.API_list[0])):
#print ("doc.title: ", doc.title)
doc.write()
else:
print ("Read document failed.")
return doc
def get_authors_data_by_doi(self, doi):
doi_doc = FullDoc(doi=doi)
if doi_doc.read(self.client):
print("doi_doc.title: ", doi_doc.title)
doi_doc.write()
else:
print("Read document failed.")
return doi
id = None
if not 'pubmed-id' in doi_doc._data.keys():
print("no pubmed-id, trying with title")
# try with title
Entrez.email = '*****@*****.**'
query = doi_doc.title
handle = Entrez.esearch(db='pubmed', retmode='xml', term=query)
results = Entrez.read(handle)
if int(results['Count']) > 0:
id = results['IdList']
else:
id = doi_doc._data['pubmed-id']
if id != None:
Entrez.email = '*****@*****.**'
handle = Entrez.efetch(db='pubmed', retmode='xml', id=id)
results = Entrez.read(handle)
print(results)
if len(results['PubmedArticle']) > 0 and (
'MedlineCitation' in results['PubmedArticle'][0].keys()
) and ('Article'
in results['PubmedArticle'][0]['MedlineCitation'].keys()):
if 'AuthorList' in results['PubmedArticle'][0][
'MedlineCitation']['Article'].keys():
authors_list = results['PubmedArticle'][0][
'MedlineCitation']['Article']['AuthorList']
dates = results['PubmedArticle'][0]['PubmedData'][
'History']
else:
print("no authors list {}".format(
results['PubmedArticle'][0]['MedlineCitation']
['Article']))
return doi
else:
print("missing keys")
return doi
else:
print("no pubmed id")
return doi
return authors_list
def search(self,query="A Lightweight Autoencoder"):
doc_srch = ElsSearch(query,'sciencedirect')
doc_srch.execute(self.client, get_all = False)
for _,doc in doc_srch.results_df.iterrows():
pii_doc = FullDoc(sd_pii = doc['pii'])
if pii_doc.read(self.client):
try:
abstract = " ".join(pii_doc.data['coredata']['dc:description'].split()[1:])
doc_id = str(hex(time.time().as_integer_ratio()[0]))
title = doc['dc:title']
pdf_link = doc['link']['scidir']
dates = doc['load-date'].split('-')[0]
self.data[doc_id] = {"title": title, "year": dates, "link": pdf_link, "Abstract":abstract}
except:
pass
else:
print("Doc Skipped!!")
fieldnames = ['doi', 'title', 'text']
writer = csv.DictWriter(file, fieldnames=fieldnames)
writer.writeheader()
doi_list = []
middle = [f"{i:02}" for i in range(13)]
last = [f"{i:03}" for i in range(100)]
# generate potential doi list
for m in middle:
for l in last:
doi_list.append(year + "." + m + "." + l)
# loop through doi list
for doi_str in doi_list:
full_doi = '10.1016/j.jesp.' + doi_str
doi_doc = FullDoc(doi=full_doi)
# check if has content
if doi_doc.read(client):
try:
doi = full_doi
title = doi_doc.title
text = doi_doc.data[
"originalText"] # ['scopus-eid', 'originalText', 'scopus-id', 'pubmed-id', 'coredata', 'objects', 'link']
writer.writerow({
'doi': doi,
'title': title,
'text': text
}) # , 'text': text
except:
continue
def test_init_doi(self):
""" Test case: uri is set correctly during initialization with DOI"""
myFullDoc = FullDoc(doi = self.doi)
assert myFullDoc.uri == self.full_doi_uri
def test_init_sd_pii(self):
""" Test case: uri is set correctly during initialization with ScienceDirect PII"""
myFullDoc = FullDoc(sd_pii = self.sd_pii)
assert myFullDoc.uri == self.full_pii_uri
def test_init_uri(self):
""" Test case: uri is set correctly during initialization with uri"""
myFullDoc = FullDoc(uri = self.full_pii_uri)
assert myFullDoc.uri == self.full_pii_uri
link = papers[i]['link'][0]['@href']
idx = link.find('pii/')
pii.append(link[idx + 4:])
i += 1
# remove duplicates
for pii_this in pii:
if pii_this in pii_total:
pii.remove(pii_this)
pii_total.extend(pii)
# get url, title, and abstract of all 100 articles
n = 0
j = 1
for p in pii:
pii_doc = FullDoc(sd_pii=p)
try:
if pii_doc.read(client):
# get title
title = pii_doc.title
title.strip()
title.replace('\n', ' ')
# get abstract
text = pii_doc.data["coredata"]["dc:description"]
if text is not None:
text = text.strip()
if text.startswith(
('ABSTRACT', 'Abstract', 'Summary')):
text = text[8:]
text = text.strip()
# remove extra whitespace
## Load configuration
con_file = open("config.json")
config = json.load(con_file)
con_file.close()
## Initialize client
client = ElsClient(config['apikey'])
client.inst_token = config['insttoken']
str='SRCTITLE(IEEE Transactions on Pattern Analysis and Machine Intelligence) AND PUBYEAR > 2018 '
myDocSearch=ElsSearch(str, 'scopus')
myDocSearch.execute(client, get_all=False)
myDocSearch.results
pii_doc = FullDoc(sd_pii='0181551220301406')
if pii_doc.read(client):
print("pii_doc.title: ", pii_doc.title)
pii_doc.write()
else:
print("Read document failed.")
str='affil(Public Health and Infection Research Group, Faculty of Health Sciences)'
myDocSearch=ElsSearch(str, 'affiliation')
myDocSearch.execute(client, get_all=False)
myDocSearch.results
str='SRCTITLE(Artifical)'
myDocSearch=ElsSearch(str, 'source')
myDocSearch.execute(client, get_all=False)
myDocSearch.results
config = json.load(config_file)
GET_ALL = config[
'get_all'] # False gets one chunk (25) True gets all or max (5000)
FULL_TEXT = config['full_text'] # Save fulltext
OPEN_ACCESS = config[
'open_access'] # Search only openaccess documents (so we can get the full text)
# "public policy AND (impact OR result OR evaluation OR evidence) AND (climate OR environment)"
query = config['query']
if OPEN_ACCESS:
query = "openaccess(1) AND " + query
client = ElsClient(config['api_key'])
doc_srch = ElsSearch(query, 'sciencedirect')
doc_srch.execute(client, get_all=GET_ALL)
for doc in doc_srch.results:
doi = doc['dc:identifier']
print(doi)
if FULL_TEXT:
## ScienceDirect (full-text) document example using DOI
doi_doc = FullDoc(doi=doi)
if doi_doc.read(client):
doi_doc.write()
else:
print("Read full-text failed for DOI", doi)
print("# Found", len(doc_srch.results), "results.")
print ("my_aff.name: ", my_aff.name)
my_aff.write()
else:
print ("Read affiliation failed.")
## Scopus (Abtract) document example
# Initialize document with ID as integer
scp_doc = AbsDoc(scp_id = 84872135457)
if scp_doc.read(client):
print ("scp_doc.title: ", scp_doc.title)
scp_doc.write()
else:
print ("Read document failed.")
## ScienceDirect (full-text) document example using PII
pii_doc = FullDoc(sd_pii = 'S1674927814000082')
if pii_doc.read(client):
print ("pii_doc.title: ", pii_doc.title)
pii_doc.write()
else:
print ("Read document failed.")
## ScienceDirect (full-text) document example using DOI
doi_doc = FullDoc(doi = '10.1016/S1525-1578(10)60571-5')
if doi_doc.read(client):
print ("doi_doc.title: ", doi_doc.title)
doi_doc.write()
else:
print ("Read document failed.")
def pii_fulltext(pii=None):
## ScienceDirect (full-text) document example using PII
pii_doc = FullDoc(sd_pii=pii)
if pii_doc.read(client):
abstract = pii_doc.data['coredata']['dc:description']
return abstract
def doi_fulltext(doi=None):
"""ScienceDirect (full-text) document example using DOI"""
doi_doc = FullDoc(doi=doi)
if doi_doc.read(client):
abstract = doi_doc.data['coredata']['dc:description']
return abstract
@author: josephwy
"""
import os
import re
import pandas as pd
import pickle
import json
import requests as r
import sys
sys.path.append('R:\\JoePriceResearch\\Python\\Anaconda3\\Lib\\site-packages')
from elsapy.elsclient import ElsClient
from elsapy.elsprofile import ElsAuthor, ElsAffil
from elsapy.elsdoc import FullDoc, AbsDoc
from elsapy.elssearch import ElsSearch
with open("config.json") as con_file:
config = json.load(con_file)
## Initialize client
client = ElsClient(config['apikey'])
client.inst_token = config['insttoken']
doi_doc = FullDoc(doi="10.1016/S1525-1578(10)60571-5")
if doi_doc.read(client):
print("doi_doc.title: ", doi_doc.title)
doi_doc.write()
print(doi_doc.data['originalText'])
else:
print("Read document failed.")
