Created
March 23, 2013 16:44
-
-
Save rahuldave/5228368 to your computer and use it in GitHub Desktop.
A script to get full text search results from ADS and try matching degree latitudes. Needs work with minutes and other stuff. Also ought to be made into a full fledged aribitrary search term handling script.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# -*- coding: utf-8 -*- | |
import requests | |
import json | |
import pprint | |
import time | |
PREFIX="http://adslabs.org/adsabs/api/search/" | |
APIKEY="INSERT CIA SECRET" | |
import re | |
re2='(°)' | |
re1='((?:\S+))' | |
re3=re1 | |
rg = re.compile(re1+re2+re3,re.IGNORECASE|re.DOTALL) | |
def myfunction(myword, start, numrows, filename, localmode=False): | |
theurlis=PREFIX+"?q="+myword+"&hl=full&hl=abstract&rows="+str(numrows)+"&start="+str(start)+"&dev_key="+APIKEY | |
matchresults={} | |
if not localmode: | |
print theurlis | |
result=requests.get(theurlis) | |
myjson=result.text | |
myresultdict=json.loads(myjson) | |
myresults=myresultdict['results']['docs'] | |
myshortformresults={e['bibcode']: e['highlights'] for e in myresults} | |
out=json.dumps(myshortformresults) | |
fd=open(filename, "w") | |
fd.write(out) | |
fd.close() | |
else: | |
stufftr=open(filename).read() | |
myshortformresults=json.loads(stufftr) | |
for bibcode in myshortformresults.keys(): | |
resultsdict=myshortformresults[bibcode] | |
for resulttype in resultsdict.keys(): | |
snippets=resultsdict[resulttype] | |
for snippet in snippets: | |
m=rg.search(snippet) | |
if m: | |
var1=m.group(1) | |
c1=m.group(2) | |
var3=m.group(3) | |
print bibcode, "("+var1+")"+"("+c1+")"+"("+var3+")"+"\n" | |
matchresults[bibcode]=(var1,var3) | |
return matchresults | |
if __name__=="__main__": | |
import sys | |
# snip="of 25°N and 40°N is used" | |
# m=rg.search(snip) | |
# if m: | |
# float1=m.group(1) | |
# c1=m.group(2) | |
# var1=m.group(3) | |
# print "("+float1+")"+"("+c1+")"+"("+var1+")"+"\n" | |
if len(sys.argv)<=1 or len(sys.argv) > 2: | |
print "No mode specified\n" | |
sys.exit(0) | |
mode=sys.argv[1] | |
print "MODE is", mode | |
if mode=="False": | |
mode=False | |
mode=bool(mode) | |
print "MODE IS", mode | |
for i in range(10): | |
out=myfunction("latitude", i*100+1, 100, "results"+str(i)+".txt", mode) | |
# fd=open("results"+str(i)+".txt", "w") | |
# fd.write(out) | |
# fd.close() | |
#time.sleep(1) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment