We Keep Coding

Website Name extract in Python

I want to extract website names from the url. For e.g. https://plus.google.com/in/test.html
should give the output as - "plus google"
Some more testcases are -
WWW.OH.MADISON.STORES.ADVANCEAUTOPARTS.COM/AUTO_PARTS_MADISON_OH_7402.HTML
Output:- OH MADISON STORES ADVANCEAUTOPARTS
WWW.LQ.COM/LQ/PROPERTIES/PROPERTYPROFILE.DO?PROPID=6054
Output:- LQ
WWW.LOCATIONS.DENNYS.COM
Output:- LOCATIONS DENNYS
WV.WESTON.STORES.ADVANCEAUTOPARTS.COM
Output:- WV WESTON STORES ADVANCEAUTOPARTS
WOODYANDERSONFORDFAYETTEVILLE.NET/
Output:- WOODYANDERSONFORFAYETTEVILLE
WILMINGTONMAYFAIRETOWNCENTER.HGI.COM
Output:- WILMINGTONMAYFAIRETOWNCENTER HGI
WHITEHOUSEBLACKMARKET.COM/
Output:- WHITEHOUSEBLACKMARKET
WINGATEHOTELS.COM
Output:- WINGATEHOTELS
string = str(input("Enter the url "))
new_list = list(string)
count=0
flag=0
if 'w' in new_list:
index1 = new_list.index('w')
new_list.pop(index1)
count += 1
if 'w' in new_list:
index2 = new_list.index('w')
if index2 != -1 and index2 == index1:
new_list.pop(index2)
count += 1
if 'w' in new_list:
index3= new_list.index('w')
if index3!= -1 and index3== index2 and new_list[index3+1]=='.':
new_list.pop(index3)
count+=1
flag = 1
if flag == 0:
start = string.find('/')
start += 2
end = string.rfind('.')
new_string=string[start:end]
print(new_string)
elif flag == 1:
start = string.find('.')
start = start + 1
end = string.rfind('.')
new_string=string[start:end]
print(new_string)
The above works for some testcases but not all. Please help me with it.
Thanks

this is something you could build upon; using urllib.parse.urlparse:
from urllib.parse import urlparse
tests = ('https://plus.google.com/in/test.html',
('WWW.OH.MADISON.STORES.ADVANCEAUTOPARTS.COM/'
'AUTO_PARTS_MADISON_OH_7402.HTML'),
'WWW.LQ.COM/LQ/PROPERTIES/PROPERTYPROFILE.DO?PROPID=6054')
def extract(url):
# urlparse will not work without a 'scheme'
if not url.startswith('http'):
url = 'http://' + url
parsed = urlparse(url).netloc
split = parsed.split('.')[:-1] # get rid of TLD
if split[0].lower() == 'www':
split = split[1:]
ret = ' '.join(split)
return ret
for url in tests:
print(extract(url))

The function strips the url from the double slash to the single slash:
the rest is 'clean up'
def stripURL( url, TwoSlashes, OneSlash ):
try:
start = url.index(TwoSlashes) + len(TwoSlashes)
end = url.index( OneSlash, start )
return url[start:end]
except ValueError:
return ""
url= raw_input("URL : ")
if "www." in url:url=url.replace("www.","")
Strip = stripURL( url, "//", "/" )
# Strips anything after the last period found
Stripped = Strip[:Strip.rfind(".")]
# get rid of the any periods used in the name
Stripped = Stripped.replace("."," ")
print Stripped

python scripts showing different result( with one error ) in two similar input files

The script, originally taken and modified from (http://globplot.embl.de/):
#!/usr/bin/env python
# Copyright (C) 2003 Rune Linding - EMBL
# GlobPlot TM
# GlobPlot is licensed under the Academic Free license
from string import *
from sys import argv
from Bio import File
from Bio import SeqIO
import fpformat
import sys
import tempfile
import os
from os import system,popen3
import math
# Russell/Linding
RL = {'N':0.229885057471264,'P':0.552316012226663,'Q':-0.187676577424997,'A':-0.261538461538462,'R':-0.176592654077609, \
'S':0.142883029808825,'C':-0.0151515151515152,'T':0.00887797506611258,'D':0.227629796839729,'E':-0.204684629516228, \
'V':-0.386174834235195,'F':-0.225572305974316,'W':-0.243375458622095,'G':0.433225711769886,'H':-0.00121743364986608, \
'Y':-0.20750516775322,'I':-0.422234699606962,'K':-0.100092289621613,'L':-0.337933495925287,'M':-0.225903614457831}
def Sum(seq,par_dict):
sum = 0
results = []
raws = []
sums = []
p = 1
for residue in seq:
try:
parameter = par_dict[residue]
except:
parameter = 0
if p == 1:
sum = parameter
else:
sum = sum + parameter#*math.log10(p)
ssum = float(fpformat.fix(sum,10))
sums.append(ssum)
p +=1
return sums
def getSlices(dydx_data, DOM_join_frame, DOM_peak_frame, DIS_join_frame, DIS_peak_frame):
DOMslices = []
DISslices = []
in_DOMslice = 0
in_DISslice = 0
beginDOMslice = 0
endDOMslice = 0
beginDISslice = 0
endDISslice = 0
for i in range( len(dydx_data) ):
#close dom slice
if in_DOMslice and dydx_data[i] > 0:
DOMslices.append([beginDOMslice, endDOMslice])
in_DOMslice = 0
#close dis slice
elif in_DISslice and dydx_data[i] < 0:
DISslices.append([beginDISslice, endDISslice])
in_DISslice = 0
# elseif inSlice expandslice
elif in_DOMslice:
endDOMslice += 1
elif in_DISslice:
endDISslice += 1
# if not in slice and dydx !== 0 start slice
if dydx_data[i] > 0 and not in_DISslice:
beginDISslice = i
endDISslice = i
in_DISslice = 1
elif dydx_data[i] < 0 and not in_DOMslice:
beginDOMslice = i
endDOMslice = i
in_DOMslice = 1
#last slice
if in_DOMslice:
DOMslices.append([beginDOMslice, endDOMslice])
if in_DISslice:
DISslices.append([beginDISslice,endDISslice])
k = 0
l = 0
while k < len(DOMslices):
if k+1 < len(DOMslices) and DOMslices[k+1][0]-DOMslices[k][1] < DOM_join_frame:
DOMslices[k] = [ DOMslices[k][0], DOMslices[k+1][1] ]
del DOMslices[k+1]
elif DOMslices[k][1]-DOMslices[k][0]+1 < DOM_peak_frame:
del DOMslices[k]
else:
k += 1
while l < len(DISslices):
if l+1 < len(DISslices) and DISslices[l+1][0]-DISslices[l][1] < DIS_join_frame:
DISslices[l] = [ DISslices[l][0], DISslices[l+1][1] ]
del DISslices[l+1]
elif DISslices[l][1]-DISslices[l][0]+1 < DIS_peak_frame:
del DISslices[l]
else:
l += 1
return DOMslices, DISslices
def SavitzkyGolay(window,derivative,datalist):
SG_bin = 'sav_gol'
stdin, stdout, stderr = popen3(SG_bin + '-D' + str(derivative) + ' -n' + str(window)+','+str(window))
for data in datalist:
stdin.write(`data`+'\n')
try:
stdin.close()
except:
print stderr.readlines()
results = stdout.readlines()
stdout.close()
SG_results = []
for result in results:
SG_results.append(float(fpformat.fix(result,6)))
return SG_results
def reportSlicesTXT(slices, sequence, maskFlag):
if maskFlag == 'DOM':
coordstr = '|GlobDoms:'
elif maskFlag == 'DIS':
coordstr = '|Disorder:'
else:
raise SystemExit
if slices == []:
#by default the sequence is in uppercase which is our search space
s = sequence
else:
# insert seq before first slide
if slices[0][0] > 0:
s = sequence[0:slices[0][0]]
else:
s = ''
for i in range(len(slices)):
#skip first slice
if i > 0:
coordstr = coordstr + ', '
coordstr = coordstr + str(slices[i][0]+1) + '-' + str(slices[i][1]+1)
#insert the actual slice
if maskFlag == 'DOM':
s = s + lower(sequence[slices[i][0]:(slices[i][1]+1)])
if i < len(slices)-1:
s = s + upper(sequence[(slices[i][1]+1):(slices[i+1][0])])
#last slice
elif slices[i][1] < len(sequence)-1:
s = s + lower(sequence[(slices[i][1]+1):(len(sequence))])
elif maskFlag == 'DIS':
s = s + upper(sequence[slices[i][0]:(slices[i][1]+1)])
#insert untouched seq between disorder segments, 2-run labelling
if i < len(slices)-1:
s = s + sequence[(slices[i][1]+1):(slices[i+1][0])]
#last slice
elif slices[i][1] < len(sequence)-1:
s = s + sequence[(slices[i][1]+1):(len(sequence))]
return s,coordstr
def runGlobPlot():
try:
smoothFrame = int(sys.argv[1])
DOM_joinFrame = int(sys.argv[2])
DOM_peakFrame = int(sys.argv[3])
DIS_joinFrame = int(sys.argv[4])
DIS_peakFrame = int(sys.argv[5])
file = str(sys.argv[6])
db = open(file,'r')
except:
print 'Usage:'
print ' ./GlobPipe.py SmoothFrame DOMjoinFrame DOMpeakFrame DISjoinFrame DISpeakFrame FASTAfile'
print ' Optimised for ELM: ./GlobPlot.py 10 8 75 8 8 sequence_file'
print ' Webserver settings: ./GlobPlot.py 10 15 74 4 5 sequence_file'
raise SystemExit
for cur_record in SeqIO.parse(db, "fasta"):
#uppercase is searchspace
seq = upper(str(cur_record.seq))
# sum function
sum_vector = Sum(seq,RL)
# Run Savitzky-Golay
smooth = SavitzkyGolay('smoothFrame',0, sum_vector)
dydx_vector = SavitzkyGolay('smoothFrame',1, sum_vector)
#test
sumHEAD = sum_vector[:smoothFrame]
sumTAIL = sum_vector[len(sum_vector)-smoothFrame:]
newHEAD = []
newTAIL = []
for i in range(len(sumHEAD)):
try:
dHEAD = (sumHEAD[i+1]-sumHEAD[i])/2
except:
dHEAD = (sumHEAD[i]-sumHEAD[i-1])/2
try:
dTAIL = (sumTAIL[i+1]-sumTAIL[i])/2
except:
dTAIL = (sumTAIL[i]-sumTAIL[i-1])/2
newHEAD.append(dHEAD)
newTAIL.append(dTAIL)
dydx_vector[:smoothFrame] = newHEAD
dydx_vector[len(dydx_vector)-smoothFrame:] = newTAIL
globdoms, globdis = getSlices(dydx_vector, DOM_joinFrame, DOM_peakFrame, DIS_joinFrame, DIS_peakFrame)
s_domMask, coordstrDOM = reportSlicesTXT(globdoms, seq, 'DOM')
s_final, coordstrDIS = reportSlicesTXT(globdis, s_domMask, 'DIS')
sys.stdout.write('>'+cur_record.id+coordstrDOM+coordstrDIS+'\n')
print s_final
print '\n'
return
runGlobPlot()
My input and output files are here: link
This script takes a input (input1.fa) and gives following output output1.txt
But when I try to run this script with similar type but larger input file (input2.fa) .. It shows following error:
Traceback (most recent call last):
File "final_script_globpipe.py", line 207, in <module>
runGlobPlot()
File "final_script_globpipe.py", line 179, in runGlobPlot
smooth = SavitzkyGolay('smoothFrame',0, sum_vector)
File "final_script_globpipe.py", line 105, in SavitzkyGolay
stdin.write(`data`+'\n')
IOError: [Errno 22] Invalid argument
I have no idea where the problem is. Any type of suggestion is appriciated.
I am using python 2.7 in windows 7 machine. I have also attached the Savitzky Golay module which is needed to run the script.
Thanks

UPDATE:
After trying to reproduce the error on linux it's showing a similar behavior, working fine with the first file but with the second is returning Errno32.
Traceback:
Traceback (most recent call last):
File "Glob.py", line 207, in <module>
runGlobPlot()
File "Glob.py", line 179, in runGlobPlot
smooth = SavitzkyGolay('smoothFrame',0, sum_vector)
File "Glob.py", line 105, in SavitzkyGolay
stdin.write(`data`+'\n')
IOError: [Errno 32] Broken pipe
Update:
Some calls of the SG_bin return that the -n parameter is the wrong type.
Wrong type of parameter for flag -n. Has to be unsigned,unsigned
This parameter comes from the window variable that is passed to the SavitzkyGolay function.
Surrounding the stdin.write with a trycatch block reveals that it breaks a hadnfull of times.
try:
for data in datalist:
stdin.write(repr(data)+'\n')
except:
print "It broke"

Extract data from web page

I have a script to extract data from here: http://espn.go.com/nba/statistics/player/_/stat/scoring-per-48-minutes/
Part of obtaining the data in the script looks like this:
pts_start = data.find('">',mpg_end) + 2
pts_end = data.find('<',pts_start)
store.append(data[pts_start:pts_end])
mf_start = data.find(' >',pts_end) + 2
mf_end = data.find('<',mf_start)
store.append(data[mf_start:mf_end])
fg_start = data.find(' >',mf_end) + 2
fg_end = data.find('<',fg_start)
store.append(data[fg_start:fg_end])
I see that the names like fg and pts correspond to the table headlines, but I don't understand why certain ones are abbreviated in the script.
I want to modify the script to obtain the headlines on this table: http://espn.go.com/nba/statistics/player/_/stat/rebounds. I tried doing this by just plugging in the names as they appear at the top of the table but the resulting CSV file had missing information.
Full code :
import os
import csv
import time
import urllib2
uri = 'http://espn.go.com/nba/statistics/player/_/stat/scoring-per-48-minutes'
def get_data():
try:
req = urllib2.Request(uri)
response = urllib2.urlopen(req, timeout=600)
content = response.read()
return content
except Exception, e:
print "\n[!] Error: " + str(e)
print ''
return False
def extract(data,rk):
print '\n[+] Extracting data.'
start = 0
while True:
store = [rk]
if data.find('nba/player/',start) == -1:
break
with open("data.csv", "ab") as fcsv:
main = data.find('nba/player/',start)
name_start = data.find('>',main) + 1
name_end = data.find('<',name_start)
store.append(data[name_start:name_end])
team_start = data.find('">',name_end) + 2
team_end = data.find('<',team_start)
store.append(data[team_start:team_end])
gp_start = data.find(' >',team_end) + 2
gp_end = data.find('<',gp_start)
store.append(data[gp_start:gp_end])
mpg_start = data.find(' >',gp_end) + 2
mpg_end = data.find('<',mpg_start)
store.append(data[mpg_start:mpg_end])
pts_start = data.find('">',mpg_end) + 2
pts_end = data.find('<',pts_start)
store.append(data[pts_start:pts_end])
mf_start = data.find(' >',pts_end) + 2
mf_end = data.find('<',mf_start)
store.append(data[mf_start:mf_end])
fg_start = data.find(' >',mf_end) + 2
fg_end = data.find('<',fg_start)
store.append(data[fg_start:fg_end])
m3_start = data.find(' >',fg_end) + 2
m3_end = data.find('<',m3_start)
store.append(data[m3_start:m3_end])
p3_start = data.find(' >',m3_end) + 2
p3_end = data.find('<',p3_start)
store.append(data[p3_start:p3_end])
ft_start = data.find(' >',p3_end) + 2
ft_end = data.find('<',ft_start)
store.append(data[ft_start:ft_end])
ftp_start = data.find(' >',ft_end) + 2
ftp_end = data.find('<',ftp_start)
store.append(data[ftp_start:ftp_end])
start = name_end
rk = rk + 1
csv.writer(fcsv).writerow(store)
fcsv.close()
def main():
print "\n[+] Initializing..."
if not os.path.exists("data.csv"):
with open("data.csv", "ab") as fcsv:
csv.writer(fcsv).writerow(["RK","PLAYER","TEAM","GP", "MPG","PTS","FGM-FGA","FG%","3PM-3PA","3P%","FTM-FTA","FT%"])
fcsv.close()
rk = 1
global uri
while True:
time.sleep(1)
start = 0
print "\n[+] Getting data, please wait."
data = get_data()
if not data:
break
extract(data,rk)
print "\n[+] Preparing for next page."
time.sleep(1.5)
rk = rk + 40
if rk > 300:
print "\n[+] All Done !\n"
break
uri = 'http://espn.go.com/nba/statistics/player/_/stat/scoring-per-48-minutes/sort/avg48Points/count/' + str(rk)
if __name__ == '__main__':
main()
I specifically want to know how to grab info based on the headlines. Like TEAM GP MPG PTS FGM-FGA FG% 3PM-3PA 3P% FTM-FTA FT%
So the script doesn't need to be changed besides things like pts or mpg in pts_start = data.find('">',mpg_end) + 2
I don't understand why I can't just input the name of the headline in the table has shown for certain ones. Like instead of FTM-FTA, the script puts ft.

Extracting html data rather easy with BeautifulSoup. Following example is you to get the idea but not a complete solution to your problem. However you can easily extend.
from bs4 import BeautifulSoup
import urllib2
def get_html_page_dom(url):
response = urllib2.urlopen(url)
html_doc = response.read()
return BeautifulSoup(html_doc, 'html5lib')
def extract_rows(dom):
table_rows = dom.select('.mod-content tbody tr')
for tr in table_rows:
# skip headers
klass = tr.get('class')
if klass is not None and 'colhead' in klass:
continue
tds = tr.select('td')
yield {'RK': tds[0].string,
'PLAYER': tds[1].select('a')[0].string,
'TEAM': tds[2].string,
'GP': tds[3].string
# you can fetch rest of the indexs for corresponding headers
}
if __name__ == '__main__':
dom = get_html_page_dom('http://espn.go.com/nba/statistics/player/_/stat/scoring-per-48-minutes/')
for data in extract_rows(dom):
print(data)
You can simply run and see the result ;).

Python Wiki Path Searching

On a personal whim I have written some code to search for the shortest series of links between any two Wikipedia articles. It turned out to be very brute force and takes a long long time to find the goal if it's more than a link or two deep, but it works! I will eventually keep track of and make use of the link paths and stuff, but I wanted to get the search working optimally first. Is there a faster way to do this or a good way to cut some major corners here?
import urllib2
from bs4 import BeautifulSoup
Start = 'http://en.wikipedia.org/wiki/Alan_Reid_%28politician%29'
End = 'http://en.wikipedia.org/wiki/Ayr'
#Using BeautifulSoup, this grabs the page
def soup_request(target):
request = urllib2.Request(target)
request.add_header("User-Agent", "Mozilla/5.0")
page = urllib2.urlopen(target)
soup = BeautifulSoup(page)
return soup
#This will grab all Wiki links off a given page
def get_links(Start):
soup = soup_request(Start)
Wiki_links = []
#Finds all links
for url in soup.findAll('a'):
result = url.get('href')
try:
if str(result)[:5] == '/wiki':
Wiki_links.append(result)
except:
pass
for q in range(len(Wiki_links)):
Wiki_links[q] = 'http://en.wikipedia.org'+str(Wiki_links[q])
print "Got new links from",Start
return Wiki_links
#This will check all the given links to see if the title matches the goal webpage
def check_links(Links,End):
goalsoup = soup_request(End)
goaltitle = goalsoup.html.title
Found = False
count = 0
for q in Links:
if Found:
break
length = len(Links)
#Runs through all the given links and checks their titles for correct one
if q is not None:
count += 1
soup = soup_request(q)
print "Checked",count,"links out of",length
try:
title = soup.html.head.title
if title == goaltitle:
Found = True
print "Found it!"
break
except:
print 'doh'
pass
return Found
#Top function to do all the stuff in the right order, applying a maximum depth of how deep into the links
def wiki_crawl(Start, End, depth):
Old_Links = [Start]
count = depth
while count > 0:
New_Links = []
for q in range(len(Old_Links)):
New_Links.extend(get_links(Old_Links[q]))
Found = check_links(New_Links,End)
if Found:
print "All done."
break
Old_Links = New_Links
count -= 1
print "_______________________________________________________________ROUND DONE"
if not Found:
print "Did not find the page, you must go deeper!"
wiki_crawl(Start, End, 2)

Here are some functions to take info from wiki. The only problems with it is that sometimes it takes out a space from the info on the webpage.
def take_out_parenthesis(st):
string = list(st)
for a in string:
if a == '(':
del string[st.find(a)]
if a == ')':
del string[st.find(a) - 1]
return ''.join(string)
def take_out_tags(string):
st = list(string)
odd = ['<', '>']
times = 0
for a in string:
if a in odd:
times += 1
times /= 2
for b in range(times):
start = string.find('<') - 1
end = string.find('>')
bet = end - start + 1
for a in range(bet):
del st[start]
string = ''.join(st)
return string
def take_out_brackets(string):
st = list(string)
odd = ['[', ']']
times = 0
for a in string:
if a in odd:
times += 1
times /= 2
for b in range(times):
start = string.find('[') - 1
end = string.find(']')
bet = end - start + 1
for a in range(bet):
del st[start]
string = ''.join(st)
return string
def take_from_web_page(text):
n = 0
url = text.replace(" ", "_")
search = "http://en.wikipedia.org/wiki/%s" % url
page = urllib2.urlopen(search).read()
start = page.find('<p><b>') + 6
end = page.find('</a>.', start) + 5
new_page = page[start:end]
for a in new_page:
if a == '<':
if new_page[n - 1] != ' ':
lst = list(new_page)
lst.insert(n, ' ')
new_page = ''.join(lst)
n += 1
n += 1
return take_out_parenthesis(take_out_brackets(take_out_tags(new_page)))

Python xlrd- Unicode error

I'm reading my data from the excel file and then writing it into the DB in Django. I'm using python xlrd module
I'm getting the following error:-
'ascii' codec can't encode character u'\xc1' in position 6: ordinal not in range(128)
I've tried all the solutions like
1) I was using str(variable) . Removed it. Now storing the value as it is in the DB.
2) Tried wb = open_workbook('static/'+filename, encoding_override="utf_16_le")
3) .encode(error=replace)
But nothing worked. How am I supposed to write-off this error?
Here is my code
def __init__(self, arm_id, dsp_name, DSP, hubcode, Pincode, pptl,state):
self.arm_id = arm_id
self.dsp_name = dsp_name
self.DSP = DSP.zfill(2)
self.hubcode = hubcode
self.Pincode = Pincode
self.pptl = pptl
self.state = state
wb = open_workbook('static/'+filename, encoding_override="utf_16_le")
for sheet in wb.sheets():
number_of_rows = sheet.nrows
number_of_columns = sheet.ncols
items = []
arm_list = []
pptl_list = []
pptlcode_list = []
count = 1
status = 0
for row in range(1, number_of_rows):
values = []
for col in range(number_of_columns):
value = (sheet.cell(row,col).value)
try: value = str(int(value))
except ValueError: pass
finally: values.append(value)
item = Excel(*values)
count +=1
arm_id = item.arm_id
if arm_id not in arm_list:
description = 'Arm'+arm_id
arm_obj = Arm(arm_id = arm_id, description = description)
arm_obj.save()
arm_list.append(arm_id)
pptl_id = (item.pptl)
if pptl_id not in pptl_list:
try :
pptl_obj = PPTLconfig.objects.get(pptl_id = pptl_id)
pptl_obj.arm_id = arm_obj
pptl_obj.hubcode = hubcode
except :
description = 'PPTL'+pptl_id
pptl_obj = PPTLconfig(pptl_id = pptl_id, description = description , arm_id = arm_obj, hubcode = (item.hubcode))
finally :
pptl_obj.save()
pptl_list.append(pptl_id)
code = []
for factors in SORTATION_FACTORS:
if factors == 'DSP': code.append((item.DSP))
elif factors == 'Pincode': code.append((item.Pincode))
elif factors == 'DG': code.append((item.state).zfill(4))
code = ','.join(code)
if code not in pptlcode_list :
try :
code_obj = PPTLcode.objects.get(code = code)
code_obj.pconf_id = pptl_obj
except : code_obj = PPTLcode(code=code, pconf_id=pptl_obj)
finally :
code_obj.save()
pptlcode_list.append(code)
else :
error = "Duplicate PPTLcode " + code + " at Row " + str(count)
status = 1
delete_data(1)
return (status,error)
###############Add ArmPrinterMapping ######################
arm_obj_list = Arm.objects.all()
for arm_obj in arm_obj_list:
printer_name = 'Arm'+str(arm_obj.arm_id)
ap_mapping = ArmPrinterMapping(arm_id = arm_obj, printer_name = printer_name)
ap_mapping.save()
return (0,0)

set default encoding to utf8, it should work then
reload(sys)
sys.setdefaultencoding('utf8')