# -*- coding: utf-8 -*-
"""
Created on Mon Apr 25 15:28:21 2016
@author: Matteus
"""
from trainPlay import refsecSearch
import sqlite3
path = 'http://arxiv.org/pdf/1501.02262v1.pdf'
dref = refsecSearch(path)
dref.findrefSection()
Jname = []
conn = sqlite3.connect('bajs.db')
c = conn.cursor()
for clusts in dref.clust:
for itr in range(clusts[0],clusts[1]):
c.execute("SELECT * FROM JournalNames WHERE instr(?,Jname) > 0;",(dref.rows[itr]))
Jname.append(c.fetchall())
c.close()
conn.close()
from trainPlay import refsecSearch
import os
#str_inp = True
ref_dir = os.path.join('C:\\','Users','Matteus','various papers')
labeled_data ={"article":[], "ref_sec":[]}
str_inp = ['dftb3.pdf','ZernerPertubation1.pdf','ZernerPertubation2.pdf']
outfile = os.path.join(ref_dir,'labeled_data')
all_labels = []
for istr in str_inp:
# while str_inp:
# str_inp = raw_input('input dir for paper: ')
ref_section = []
dref = refsecSearch(os.path.join(ref_dir,istr))
nrows = len(dref.rows)
print 'There are a total of '+str(nrows)+' rows'
dref.findrefSection()
print 'the clusters found are the following'
print dref.clust
while True:
start_val = int(raw_input('from what row do you want to start? '))
end_val = int(raw_input('until what row? '))
if start_val != 0:
try:
print dref.rows[start_val-1]
except:
print dref.rows[start_val-1].encode('utf-8')
print '_____________________________________'
for row in dref.rows[start_val:end_val]:
try:
print row
