I am using the following Expression to select all hyperlinks
//a[@href]
How can I write an expression to select all hyperlinks which match this format
here http://abc.com/articles/ is constant and the article number increases
+1
A:
That expression looks like XPath, not a regex. A regex for that particular URL would look like
^http://abc.com/articles/\d+$
But I guess you'll have to use your xpath query to find the hyperlinks, then filter them based on the HREF attribute using that regex.
Mark
2010-04-21 05:13:20
I am trying to reference it like this HtmlNodeCollection hrefs = _doc.DocumentNode.SelectNodes(@"^http://abc.com/articles/\d+$"); but it gives me an error. Any advice?
Veejay
2010-04-21 05:28:28
Ack!!! What did I just say? You're mixing different kinds of expressions! You *cannot* put a regular expression in there, you have to use the *xpath query* you came up with and *then* iterate over the nodes and throw away the ones you don't want using the *regular expression*.
Mark
2010-04-21 18:09:15
Actually, Pavel's solution is pretty good. It doesn't use a "regular expression" like you asked for, but you don't really need one in this case ;)
Mark
2010-04-21 18:10:20
+1
A:
<a\s.*?href=(?:["'](http://abc.com/articles/([0-9])+)["']).*?>(.*?)</a>
UPDATE:
If you need the xpath expression here it is:
a[starts-with(@href,'http://abc.com/articles/')]
this would return all the links which has href attribute which starts with 'http://abc.com/articles/' I hope this answers your qiestion.
Pavel Nikolov
2010-04-21 05:15:57
A:
It's a bit overkill, but this is the regex I use in my apps to find URLs in plain text:
(\b(?:(?:https?|ftp|file)://|www\.|ftp\.) (?:\([-A-Z0-9+&@#/%=~_|\$\?!:,\.]*\) |[-A-Z0-9+&@#/%=~_|\$\?!:,\.])* (?:\([-A-Z0-9+&@#/%=~_|\$\?!:,\.]*\) |[A-Z0-9+&@#/%=~_|\$]))
Brian Roach
2010-04-21 05:16:30