def search(filename, filename2, newmdata):
import csv
import crawl
import remove
import numpy as np
counter = 0
companies = []
webs = []
words = []
url = []
url2 = []
urls = []
urls2 = []
#open csv files, get url into array
geo = open(filename, 'rU')
csv_geo = csv.reader(geo)
csv_geo.next()
for row in csv_geo:
company = row[0]
web = row[10]
companies.append(company)
webs.append(web)
geo.close()
#open a csv file with keywords
geo2 = open(filename2, 'rU')
csv_geo2 = csv.reader(geo2)
csv_geo2.next()
for row in csv_geo2:
word = row[0]
words.append(word)
geo2.close()
header = ["Company Name", "URLs"]
#write to new csv document
gmas = open(newmdata,'wb')
csv_gmas = csv.writer(gmas)
csv_gmas.writerow(header)
for i in range(len(webs)):
try:
url = np.array(crawl.scraping(webs[i]))
except AttributeError:
pass
url2 = np.array(remove.removeduplicates(url))
for j in range(len(words)):
for k in range(len(url2)):
if words[j] in url2[k]:
counter = counter + 1
if counter > 1:
continue
else:
urls.append(url2[k])
else:
continue
counter = 0
urls.insert(0, companies[i])
urls2 = remove.removeduplicates(urls)
#only take the first four entry (for prototyping)
urls2 = urls2[:4]
csv_gmas.writerow(urls2)
url = []
urls = []
urls2 = []
gmas.close()
def sortandcleansekeywords(filename, filename2):
import csv
import link
import remove
btypes = []
btypes2 = []
headers = []
keys = []
companies = []
rows = []
counter = 0
counter2 = 0
filename3 = "industry_type_sorted.csv"
filename4 = "industry_type_sorted_&_cleaned.csv"
#open csv files, get url into array
geo = open(filename2, 'rU')
csv_geo = csv.reader(geo)
layers = csv_geo.next()
for k in range(len(layers)):
header = "Business Type" + str(k)
headers.append(header)
headers.insert(0, "Company Name")
#part one
#create tags csv file
f = open(filename, 'wb')
csv_f = csv.writer(f)
csv_f.writerow(headers)
f.close()
for row in csv_geo:
tags = row
for i, col in enumerate(row):
if col.isdigit() == True:
btype = layers[i]
else:
continue
btypes.append(btype)
btypes.insert(0, row[0])
if len(tags) == len(btypes):
for j in range(len(tags)):
if tags[j] == "1":
btype = btypes[j]
else:
continue
btypes2.append(btype)
else:
btype = "N/A"
btypes2.append(btype)
btypes2.insert(0, row[0])
link.writelink(btypes2, filename)
btypes = []
btypes2 = []
geo.close()
f = open(filename3, 'wb')
csv_f = csv.writer(f)
csv_f.writerow(headers)
f.close()
#part two
#edit csv
geo2 = open(filename, 'rU')
csv_geo2 = csv.reader(geo2)
csv_geo2.next()
for row in csv_geo2:
if counter > 0:
counter = counter + 1
temp2 = row
if temp2[0] == temp[0]:
counter2 = counter2 + 1
if len(temp) == len(temp2):
for l in range(len(temp)):
if temp[l] in temp2:
continue
else:
temp2.append(temp[l])
if counter2 == 0:
link.writelink(temp2, filename3)
else:
pass
counter2 = 0
elif temp2[0] != temp[0]:
link.writelink(temp, filename3)
temp = temp2
else:
counter = counter + 1
temp = row
geo2.close()
tcounter = counter
counter = 0
#second editing
geo2 = open(filename, 'rU')
csv_geo2 = csv.reader(geo2)
csv_geo2.next()
for row in csv_geo2:
counter = counter + 1
if counter == (tcounter - 1):
temp2 = row
elif counter == tcounter:
temp = row
if temp2[0] == temp[0]:
counter2 = counter2 + 1
if len(temp) == len(temp2):
for l in range(len(temp)):
if temp[l] in temp2:
continue
else:
temp2.append(temp[l])
if counter2 == 0:
link.writelink(temp2, filename3)
else:
pass
counter2 = 0
elif temp2[0] != temp[0]:
link.writelink(temp, filename3)
geo2.close()
#part three
crm = "customer relationship management"
erp = "entreprise resource planning"
ba = "business analytics"
f = open(filename4, 'wb')
csv_f = csv.writer(f)
csv_f.writerow(headers)
f.close()
#third editing
geo2 = open(filename3, 'rU')
csv_geo2 = csv.reader(geo2)
csv_geo2.next()
for row in csv_geo2:
temp = row
for m in range(len(temp)):
if temp[m] == crm :
temp[m] = "crm"
elif temp[m] == erp :
temp[m] = "erp"
elif temp[m] == ba :
temp[m] = "analytics"
else:
pass
temp = remove.removeduplicates(temp)
link.writelink(temp, filename4)
geo2.close()
return filename4
