Extracting text from a Webpage using BeautifulSoup

[Magnus.Moraberg]

Hi,

I wish to extract all the words on a set of webpages and store them in
a large dictionary. I then wish to procuce a list with the most common
words for the language under consideration. So, my code below reads
the page -

http://news.bbc.co.uk/welsh/hi/newsid_7420000/newsid_7420900/7420967.stm

a welsh language page. I hope to then establish the 1000 most commonly
used words in Welsh. The problem I'm having is that
soup.findAll(text=True) is returning the likes of -

u'doctype html public "-//W3C//DTD HTML 4.0 Transitional//EN" "http://
www.w3.org/TR/REC-html40/loose.dtd"'

and -

<a href=" \'+url+\'?rss=\'+rssURI+\'" class="sel"

Any suggestions how I might overcome this problem?

Thanks,

Barry.


Here's my code -

import urllib
import urllib2
from BeautifulSoup import BeautifulSoup

# proxy_support = urllib2.ProxyHandler({"http":"http://
999.999.999.999:8080"})
# opener = urllib2.build_opener(proxy_support)
# urllib2.install_opener(opener)

page = urllib2.urlopen('http://news.bbc.co.uk/welsh/hi/newsid_7420000/
newsid_7420900/7420967.stm')
soup = BeautifulSoup(page)

pageText = soup.findAll(text=True)
print pageText

 

[Marc 'BlackJack' Rintsch]

I wish to extract all the words on a set of webpages and store them in
a large dictionary. I then wish to procuce a list with the most common
words for the language under consideration. So, my code below reads
the page -

http://news.bbc.co.uk/welsh/hi/newsid_7420000/newsid_7420900/7420967.stm

a welsh language page. I hope to then establish the 1000 most commonly
used words in Welsh. The problem I'm having is that
soup.findAll(text=True) is returning the likes of -

u'doctype html public "-//W3C//DTD HTML 4.0 Transitional//EN" "http://
www.w3.org/TR/REC-html40/loose.dtd"'

Just extract the text from the body of the document.

body_texts = soup.body(text=True)

and -

<a href=" \'+url+\'?rss=\'+rssURI+\'" class="sel"

Any suggestions how I might overcome this problem?

Ask the BBC to produce HTML that's less buggy. ;-)

http://validator.w3.org/ reports bugs like "'body' tag not allowed here"
or closing tags without opening ones and so on.

Ciao,
Marc 'BlackJack' Rintsch

 

[Magnus.Moraberg]

Just extract the text from the body of the document.

body_texts = soup.body(text=True)




Ask the BBC to produce HTML that's less buggy. ;-)

http://validator.w3.org/reports bugs like "'body' tag not allowed here"
or closing tags without opening ones and so on.

Ciao,
Marc 'BlackJack' Rintsch

Great, thanks!

 

[Paul McGuire]

Hi,

I wish to extract all the words on a set of webpages and store them in
a large dictionary. I then wish to procuce a list with the most common
words for the language under consideration. So, my code below reads
the page -

http://news.bbc.co.uk/welsh/hi/newsid_7420000/newsid_7420900/7420967.stm

a welsh language page. I hope to then establish the 1000 most commonly
used words in Welsh. The problem I'm having is that
soup.findAll(text=True) is returning the likes of -

u'doctype html public "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd"'

and -

<a href=" \'+url+\'?rss=\'+rssURI+\'" class="sel"

Any suggestions how I might overcome this problem?

Thanks,

Barry.

Here's my code -

import urllib
import urllib2
from BeautifulSoup import BeautifulSoup

# proxy_support = urllib2.ProxyHandler({"http":"http://
999.999.999.999:8080"})
# opener = urllib2.build_opener(proxy_support)
# urllib2.install_opener(opener)

page = urllib2.urlopen('http://news.bbc.co.uk/welsh/hi/newsid_7420000/
newsid_7420900/7420967.stm')
soup = BeautifulSoup(page)

pageText = soup.findAll(text=True)
print pageText

As an alternative datapoint, you can try out the htmlStripper example
on the pyparsing wiki: http://pyparsing.wikispaces.com/space/showimage/htmlStripper.py

-- Paul