def get_paper_info(doi=None, url=None):
# Resolve DOI or URL through PyPub pub_resolve methods
publisher_base_url, full_url = pub_resolve.get_publisher_urls(doi=doi, url=url)
pub_dict = pub_resolve.get_publisher_site_info(publisher_base_url)
# Create a PaperInfo object to hold all information and call appropriate scraper
paper_info = PaperInfo(doi=doi, scraper_obj=pub_dict['object'], url=full_url)
paper_info.populate_info()
return paper_info
def science_direct(self):
# Sample journal article
sd_link = 'http://www.sciencedirect.com/science/article/pii/S0006899313013048'
# Make a PaperInfo object from the live site information
pi = PaperInfo(url=sd_link, scraper_obj='sciencedirect_selenium')
pi.populate_info()
pi.publisher_interface = None
# Write saved version of the PaperInfo object
file_path_name = os.path.join(self.dirname, 'sd_info.txt')
with open(file_path_name, 'wb') as file:
pickle.dump(pi, file)
def springer(self):
# Sample journal article
sp_link = 'http://link.springer.com/article/10.1007/s10237-015-0706-9'
sp_doi = '10.1007/s10237-015-0706-9'
# Make a PaperInfo object from the live site information
pi = PaperInfo(url=sp_link, doi=sp_doi, scraper_obj='springer')
pi.populate_info()
pi.publisher_interface = None
# Write saved version of the PaperInfo object
file_path_name = os.path.join(self.dirname, 'sp_info.txt')
with open(file_path_name, 'wb') as file:
pickle.dump(pi, file)
def nature_nrg(self):
# Sample journal article
nrg_link = 'http://www.nature.com/nrg/journal/v15/n5/full/nrg3686.html'
nrg_doi = '10.1038/nrg3686'
# Make a PaperInfo object from the live site information
pi = PaperInfo(url=nrg_link, doi=nrg_doi, scraper_obj='nature_nrg')
pi.populate_info()
pi.publisher_interface = None
# Write saved version of the PaperInfo object
file_path_name = os.path.join(self.dirname, 'nrg_info.txt')
with open(file_path_name, 'wb') as file:
pickle.dump(pi, file)
def _construct_object(self, json):
if json is None:
return None
entry = models.ScopusEntry(json=json)
# Get references from the API response
ref_list = BibliographyRetrieval._refs_from_json(json=json)
references = []
if ref_list is not None:
for ref_json in ref_list:
references.append(models.ScopusRef(ref_json))
paper_info = PaperInfo()
#paper_info.entry = utils.convert_to_dict(entry)
#paper_info.references = utils.refs_to_list(references)
paper_info.entry = entry
paper_info.references = references
paper_info.doi = getattr(entry, 'doi', None)
paper_info.pdf_link = None
paper_info.publisher_interface = None
paper_info.scraper_obj = None
return paper_info
def doi_to_info(doi=None, url=None):
"""
Gets entry and references information for an article DOI.
Uses saved dicts matching DOI prefixes to publishers and web scrapers
to retrieve information. Will fail if DOI prefix hasn't been saved
with a publisher link or if a scraper for a specific publisher
site hasn't been built.
Parameters
----------
doi : str
Unique ID assigned to a journal article.
url : str
The CrossRef URL to the article page.
I.e. http://dx.doi.org/10.######
Returns
-------
paper_info : PaperInfo
Class containing parameters including the following:
entry_dict : dict
Contains information about the paper referenced by the DOI.
Includes title, authors, affiliations, publish date, journal
title, volume, and pages, and keywords. Some values are other
dicts (for example, the author info with affiliation values).
Formatted to be JSON serializable.
refs_dicts : list of dicts
Each list item is a dict corresponding to an individual reference
from the article's reference list. Includes title, authors,
publishing date, journal title, volume, and pages (if listed),
and any external URL links available (i.e. to where it is hosted
on other sites, or pdf links).
full_url : str
URL to the journal article page on publisher's website.
"""
# Resolve DOI or URL through PyPub pub_resolve methods
publisher_base_url, full_url = pub_resolve.get_publisher_urls(doi=doi, url=url)
pub_dict = pub_resolve.get_publisher_site_info(publisher_base_url)
# Create a PaperInfo object to hold all information and call appropriate scraper
paper_info = PaperInfo(doi=doi, scraper_obj=pub_dict['object'], url=full_url)
paper_info.populate_info()
return paper_info
def taylor_francis(self):
# NOTE: The current version of the T&F scraper is for a deprecated version
# of the site. All of the HTML tags need to be changed.
# Sample journal article
tf_link = 'http://www.tandfonline.com/doi/full/10.1080/21624054.2016.1184390'
tf_doi = '10.1080/21624054.2016.1184390'
# Make a PaperInfo object from the live site information
pi = PaperInfo(url=tf_link, doi=tf_doi, scraper_obj='taylorfrancis')
pi.populate_info()
pi.publisher_interface = None
# Write saved version of the PaperInfo object
file_path_name = os.path.join(self.dirname, 'tf_info.txt')
with open(file_path_name, 'wb') as file:
pickle.dump(pi, file)
def wiley(self):
# Sample journal article
wy_link = 'http://onlinelibrary.wiley.com/doi/10.1002/biot.201400046/references'
wy_doi = '10.1002/biot.201400046'
# Make a PaperInfo object from the live site information
# pi.publisher_interface needs to be set to None or else
# the object could not be saved.
pi = PaperInfo(url=wy_link, doi=wy_doi, scraper_obj='wiley')
pi.populate_info()
pi.publisher_interface = None
# Write saved version of the PaperInfo object
file_path_name = os.path.join(self.dirname, 'wy_info.txt')
with open(file_path_name, 'wb') as file:
pickle.dump(pi, file)
def _construct_object(self, json):
if json is None:
return None
entry = models.ScopusEntry(json=json)
# Get references from the API response
ref_list = BibliographyRetrieval._refs_from_json(json=json)
references = []
if ref_list is not None:
for ref_json in ref_list:
references.append(models.ScopusRef(ref_json))
paper_info = PaperInfo()
#paper_info.entry = utils.convert_to_dict(entry)
#paper_info.references = utils.refs_to_list(references)
paper_info.entry = entry
paper_info.references = references
paper_info.doi = getattr(entry, 'doi', None)
paper_info.pdf_link = None
paper_info.publisher_interface = None
paper_info.scraper_obj = None
return paper_info
def __init__(self):
self.curpath = str(os.path.dirname(os.path.abspath(__file__)))
self.link = 'http://www.sciencedirect.com/science/article/pii/S0006899313013048'
self.doi = 'S0006899313013048'
# Make a PaperInfo object from the live site information
try:
pi = PaperInfo(url=self.link, doi=self.doi, scraper_obj='sciencedirect')
pi.populate_info()
except Exception:
self.pi = None
self.entry_dict = None
else:
self.pi = pi
self.entry_dict = self.pi.entry.__dict__
# Load saved version of the PaperInfo object
saved_dir = os.path.join(self.curpath, 'saved_info')
saved_file_path = os.path.join(saved_dir, 'sp_info.txt')
self.saved_pi = pickle.load(open(saved_file_path, 'rb'))
# Make the saved versions into dicts
self.saved_entry_dict = self.saved_pi.entry.__dict__
def __init__(self):
self.curpath = str(os.path.dirname(os.path.abspath(__file__)))
self.link = 'http://www.nature.com/nrg/journal/v15/n5/full/nrg3686.html'
self.doi = '10.1038/nrg3686'
# Make a PaperInfo object from the live site information
try:
pi = PaperInfo(url=self.link, doi=self.doi, scraper_obj='nature')
pi.populate_info()
except Exception:
self.pi = None
self.entry_dict = None
else:
self.pi = pi
self.entry_dict = self.pi.entry.__dict__
# Load saved version of the PaperInfo object
saved_dir = os.path.join(self.curpath, 'saved_info')
saved_file_path = os.path.join(saved_dir, 'sp_info.txt')
self.saved_pi = pickle.load(open(saved_file_path, 'rb'))
# Make the saved versions into dicts
self.saved_entry_dict = self.saved_pi.entry.__dict__
def __init__(self):
self.curpath = str(os.path.dirname(os.path.abspath(__file__)))
self.link = 'http://link.springer.com/article/10.1186/s12984-016-0150-9'
self.doi = '10.1186/s12984-016-0150-9'
# Make a PaperInfo object from the live site information
try:
pi = PaperInfo(url=self.link, doi=self.doi, scraper_obj='Springer')
pi.populate_info()
except Exception:
self.pi = None
self.entry_dict = None
else:
self.pi = pi
self.entry_dict = self.pi.entry.__dict__
# Load saved version of the PaperInfo object
saved_dir = os.path.join(self.curpath, 'saved_info')
saved_file_path = os.path.join(saved_dir, 'sp_info.txt')
self.saved_pi = pickle.load(open(saved_file_path, 'rb'))
# Make the saved versions into dicts
self.saved_entry_dict = self.saved_pi.entry.__dict__
def __init__(self):
self.curpath = str(os.path.dirname(os.path.abspath(__file__)))
self.link = 'http://onlinelibrary.wiley.com/doi/10.1002/biot.201400046/references'
self.doi = '10.1002/biot.201400046'
# Make a PaperInfo object from the live site information
try:
pi = PaperInfo(url=self.link, doi=self.doi, scraper_obj='wiley')
pi.populate_info()
except Exception:
self.pi = None
self.entry_dict = None
else:
self.pi = pi
self.entry_dict = self.pi.entry.__dict__
# Load saved version of the PaperInfo object
saved_dir = os.path.join(self.curpath, 'saved_info')
saved_file_path = os.path.join(saved_dir, 'wy_info.txt')
self.saved_pi = pickle.load(open(saved_file_path, 'rb'))
# Make the saved versions into dicts
self.saved_entry_dict = self.saved_pi.entry.__dict__
def resolve_doi(doi):
"""
Gets the paper and references information from an article DOI.
Parameters
----------
doi : str
DOI = digital object identifier.
Unique ID assigned to a journal article.
Example: 10.1002/biot.201400046
Returns
-------
paper_info : PaperInfo
See resolve_citation for description.
"""
doi_prefix = doi[0:7]
# Same steps as in resolve_citation
saved_info = get_saved_info(doi)
if saved_info is not None:
saved_paper_info = PaperInfo()
for k, v in saved_info.items():
setattr(saved_paper_info, k, v)
saved_paper_info.make_interface_object()
return saved_paper_info
paper_info = doi_to_info(doi)
idnum = assign_id()
paper_info.idnum = idnum
log_info(paper_info)
return paper_info
def get_paper_info(self):
paper_info = PaperInfo()
paper_info.entry = self.get_entry_info()
def resolve_citation(citation):
"""
Gets the paper and references information from
a plaintext citation.
Uses a search to CrossRef.org to retrive paper DOI.
Parameters
----------
citation : str
Full journal article citation.
Example: Senís, Elena, et al. "CRISPR/Cas9‐mediated genome
engineering: An adeno‐associated viral (AAV) vector
toolbox. Biotechnology journal 9.11 (2014): 1402-1412.
Returns
-------
paper_info : PaperInfo
Class containing relevant paper meta-information and
references list.
Information about the paper itself is in 'entry' value, and is a dict
(with str and dict values). References list is in 'references'
value and is a list of dicts (each with str and dict values).
Must call .__dict__ to be JSON-serializable
"""
# Encode raw citation
citation = urllib_quote(citation)
# Search for citation on CrossRef.org to try to get a DOI link
api_search_url = 'http://search.labs.crossref.org/dois?q=' + citation
response = requests.get(api_search_url).json()
doi = response[0]['doi']
print(doi)
# If crossref returns a http://dx.doi.org/ link, retrieve the doi from it
# and save the URL to pass to doi_to_info
url = None
if doi[0:18] == 'http://dx.doi.org/':
url = doi
doi = doi[18:]
doi_prefix = doi[0:7]
# Check if this DOI has been searched and saved before.
# If it has, return saved information
saved_info = get_saved_info(doi)
if saved_info is not None:
saved_paper_info = PaperInfo()
for k, v in saved_info.items():
setattr(saved_paper_info, k, v)
saved_paper_info.make_interface_object()
return saved_paper_info
paper_info = doi_to_info(doi, doi_prefix, url)
idnum = assign_id()
paper_info.idnum = idnum
log_info(paper_info)
return paper_info
