def test_citation_tree():
output_filename = "citation-tree.json"
paper = list(search("^Casey, A", sort="citations", order="desc", rows=1))[0]
paper.build_citation_tree(depth=3)
network.export(paper, "citations", output_filename,
article_repr=lambda article: article.author[0],
new_branch_repr=lambda article, branch: {"name": article.author[0], "children": branch},
end_branch_repr=lambda article: {"name": article.author[0]},
indent=2, clobber=True)
if os.path.exists(output_filename):
os.unlink(output_filename)
if __name__ == "__main__":
# Let's do it for *last* month
current_time = localtime()
# Just grab the last month (there are more Pythonic ways to do this)
year = current_time.tm_year - 1 if current_time.tm_mon == 1 else current_time.tm_year
month = current_time.tm_mon - 1 if current_time.tm_mon > 1 else 12
# The affiliation string to search for
my_affiliation = '"Australian National University"'
# Get all the articles
articles = ads.search(
affiliation=my_affiliation,
filter="database:astronomy AND property:refereed",
dates="{year}/{month}".format(year=year, month=month))
# ads.search yields a generator, so let's list-ify the articles for multiple use
articles = list(articles)
print("There were {0} articles found. Sorting and downloading..".format(len(articles)))
# Let's do something interesting with the data first
# We'll sort all the articles by first-authors with our matched affiliation first.
sorted_articles = sorted(articles,
key=lambda article: [(my_affiliation.strip('"').lower() in affiliation.lower()) for affiliation in article.aff].index(True))
# Great! Now let's actually do something real with these articles
# At Mount Stromlo Observatory (the Research School of Astronomy & Astrophysics within
# the Australian National University), we have a "monthly papers" board that shows the
# first page of every paper published by someone at Stromlo within the last month.
# Standard library
import httplib
import json
import os
import urllib
from collections import Counter
# Module specific
import ads
# Couple of mutable variables for the reader
author_query = "^Casey, Andrew R."
records_filename = "citations.json"
papers = ads.search(author_query)
# How many citations did we have last time this ran?
if not os.path.exists(records_filename):
all_citations_last_time = {
"total": 0
}
else:
with open(records_filename, "r") as fp:
all_citations_last_time = json.load(fp)
# Build a dictionary with all of our citations
bibcodes, citations = zip(*[(paper.bibcode, paper.citation_count) for paper in papers])
all_citations = dict(zip(bibcodes, citations))
all_citations["total"] = sum(citations)
# Standard library
import httplib
import json
import os
import urllib
from collections import Counter
# Module specific
import ads
# Couple of mutable variables for the reader
author_query = "^Casey, Andrew R."
records_filename = "citations.json"
papers = ads.search(author_query)
# How many citations did we have last time this ran?
if not os.path.exists(records_filename):
all_citations_last_time = {
"total": 0
}
else:
with open(records_filename, "r") as fp:
all_citations_last_time = json.load(fp)
# Build a dictionary with all of our citations
bibcodes, citations = zip(*[(paper.bibcode, paper.citation_count) for paper in papers])
all_citations = dict(zip(bibcodes, citations))
all_citations["total"] = sum(citations)
# coding: utf-8
""" Who are the most cited astronomers? """
__author__ = "Andy Casey <*****@*****.**>"
import ads
# Let's assume the most cited people have the most cited papers, since we can only search for papers, not people
most_cited_papers = ads.search(sort="citations", filter="database:astronomy", rows=200)
# Who are these successful people, anyways?
successful_astronomers = [paper.author[0] for paper in most_cited_papers]
# Okay, let's get the top 200 most-cited papers for each person and see how many citations they have in total
total_citations = {}
for astronomer in successful_astronomers:
papers = ads.search("^{0}".format(astronomer.encode("utf-8")),
sort="citations", filter="database:astronomy", rows=200)
total_citations[astronomer] = sum([paper.citation_count for paper in papers])
# Now there's a problem because astronomers publish under "Aaronson, A" and "Aaronson, Aaron". Ugh!
duplicate_astronomers = []
for astronomer in total_citations.keys():
# Look out for "Groups" or "Teams"
if "," not in astronomer and total_citations[astronomer] == 0:
print("{0} looks like a group without any citations, so we're going to delete this item.".format(astronomer))
duplicate_astronomers.append(astronomer)
continue
try:
if __name__ == "__main__":
# Let's do it for *last* month
current_time = localtime()
# Just grab the last month (there are more Pythonic ways to do this)
year = current_time.tm_year - 1 if current_time.tm_mon == 1 else current_time.tm_year
month = current_time.tm_mon - 1 if current_time.tm_mon > 1 else 12
# The affiliation string to search for
my_affiliation = '"Australian National University"'
# Get all the articles
articles = ads.search(affiliation=my_affiliation,
filter="database:astronomy AND property:refereed",
dates="{year}/{month}".format(year=year,
month=month))
# ads.search yields a generator, so let's list-ify the articles for multiple use
articles = list(articles)
print("There were {0} articles found. Sorting and downloading..".format(
len(articles)))
# Let's do something interesting with the data first
# We'll sort all the articles by first-authors with our matched affiliation first.
sorted_articles = sorted(
articles,
key=lambda article: [(my_affiliation.strip('"').lower(
) in affiliation.lower()) for affiliation in article.aff].index(True))
# Great! Now let's actually do something real with these articles
# coding: utf-8
""" Who are the most cited astronomers? """
__author__ = "Andy Casey <*****@*****.**>"
import ads
# Let's assume the most cited people have the most cited papers, since we can only search for papers, not people
most_cited_papers = ads.search(sort="citations",
filter="database:astronomy",
rows=200)
# Who are these successful people, anyways?
successful_astronomers = [paper.author[0] for paper in most_cited_papers]
# Okay, let's get the top 200 most-cited papers for each person and see how many citations they have in total
total_citations = {}
for astronomer in successful_astronomers:
papers = ads.search("^{0}".format(astronomer.encode("utf-8")),
sort="citations",
filter="database:astronomy",
rows=200)
total_citations[astronomer] = sum(
[paper.citation_count for paper in papers])
# Now there's a problem because astronomers publish under "Aaronson, A" and "Aaronson, Aaron". Ugh!
duplicate_astronomers = []
for astronomer in total_citations.keys():
# Look out for "Groups" or "Teams"
if "," not in astronomer and total_citations[astronomer] == 0:
print(
# coding: utf-8
""" Export the data to make a citation tree visualisation with D3. """
__author__ = "Andy Casey <*****@*****.**>"
import ads
# Let's grab a paper to build a citation tree from
paper = ads.search("^Hubble, E", sort="citations", order="desc")[0]
# Build our citation tree to a depth of 2
paper.build_citation_tree(depth=2)
# Make a function that turns "Lastname, Firstname I." -> "Lastname, F"
pretty_author_name = lambda author: author.split(",")[0] + author.split(",")[
1].strip()[1] + "."
# Export the network to citation-tree.json
ads.network.export(
paper,
"citations",
"citation-tree.json",
article_repr=lambda article: pretty_author_name(article.author[0]),
new_branch_repr=lambda article, branch: {
"name": pretty_author_name(article.author[0]),
"children": branch
},
end_branch_repr=lambda article:
{"name": pretty_author_name(article.author[0])},
indent=2,
clobber=True)
# coding: utf-8
""" Export the data to make a citation tree visualisation with D3. """
__author__ = "Andy Casey <*****@*****.**>"
import ads
# Let's grab a paper to build a citation tree from
paper = ads.search("^Hubble, E", sort="citations", order="desc")[0]
# Build our citation tree to a depth of 2
paper.build_citation_tree(depth=2)
# Make a function that turns "Lastname, Firstname I." -> "Lastname, F"
pretty_author_name = lambda author: author.split(",")[0] + author.split(",")[1].strip()[1] + "."
# Export the network to citation-tree.json
ads.network.export(paper, "citations", "citation-tree.json",
article_repr=lambda article: pretty_author_name(article.author[0]),
new_branch_repr=lambda article, branch: {"name": pretty_author_name(article.author[0]), "children": branch},
end_branch_repr=lambda article: {"name": pretty_author_name(article.author[0])},
indent=2, clobber=True)
