blob: 2d29200ec66275c90ee9d2fd3e7b4b9a43c0f0b2 (
plain)
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
|
# URL extractor
# Copyright 2004, Paul McGuire
from pyparsing import makeHTMLTags, pyparsing_common as ppc
from urllib.request import urlopen
import pprint
linkOpenTag, linkCloseTag = makeHTMLTags("a")
linkBody = linkOpenTag.tag_body
linkBody.setParseAction(ppc.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 urlopen("https://www.cnn.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({toks.body: toks.href for toks, strt, end in link.scanString(htmlText)})
|
