1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
|
# URL extractor
# Copyright 2004, Paul McGuire
from pyparsing import makeHTMLTags, SkipTo, pyparsing_common
import urllib.request
from contextlib import closing
import pprint
linkOpenTag, linkCloseTag = makeHTMLTags('a')
linkBody = SkipTo(linkCloseTag)
linkBody.setParseAction(pyparsing_common.stripHTMLTags)
linkBody.addParseAction(lambda toks: ' '.join(toks[0].strip().split()))
link = linkOpenTag + linkBody("body") + linkCloseTag.suppress()
# Go get some HTML with some links in it.
with closing(urllib.request.urlopen("http://www.yahoo.com")) as serverListPage:
htmlText = serverListPage.read().decode("UTF-8")
# scanString is a generator that loops through the input htmlText, and for each
# match yields the tokens and start and end locations (for this application, we are
# not interested in the start and end values).
for toks,strt,end in link.scanString(htmlText):
print(toks.asList())
# Create dictionary from list comprehension, assembled from each pair of tokens returned
# from a matched URL.
pprint.pprint(
dict((toks.body, toks.href) for toks,strt,end in link.scanString(htmlText))
)
|
