Extract action attribute in a Form tag with Regex in C#?

Question

I wanna extract https://www.sth.com/yment/Paymentform.aspx from below string

<form id='paymentUTLfrm' action='https://www.sth.com/yment/Paymentform.aspx' method='post'>

How can I do it with Regex or somthing ?

Answer 1

Use Html Agility Pack. It will save you a lot of trouble in the long run.

using HtmlAgilityPack;     
var doc = new HtmlDocument();                                       
doc.LoadHtml("<form id='paymentUTLfrm' action='https://www.sth.com/yment/Paymentform.aspx' method='post'>");      
var form = doc.DocumentNode.SelectSingleNode("id('paymentUTLfrm')");
string action = form.Attributes["action"].Value;

It supports loading pages directly from the web, as well as XPath (used above). The HTML does not have to be valid.

EDIT: If you want to use the name:

doc.DocumentNode.SelectSingleNode("//*[@name='paymentUTLfrm']");

Answer 2

While I would agree that general html parsing is best done with html agility pack (etc) rather than with regex, this is a pretty simple requirement and a regex would be appropriate. I am no regex expert, but this one works:

action=["'](.*)["']

The (.*) will capture the url

maybe some expert can add a comnent to refine this...

Answer 3

While I don't encourage using regex to parse HTML, this is simple enough that a regex will suffice. For more complex operations, do use a proper (X)HTML parser like HtmlAgilityPack.

This regex should work:

<\s*form[^>]*\s+action=(["'])(.*?)\1

EDIT:

Updated regex so it will work with apostrophes in URLs. Note that the URL is now in the 2nd capture group.

See it on rubular