Getting value string containing InnerHTML regarding differentes rules

Question

I have a string containing the inner value of a DIV (with content editable) coming from client side.
Inside this div there is some SPAN, some P, some Table,... everything you can have.

I'd like to get from this string only the value inside the P element, and sometimes inside the TD of the tables (sometimes a P is inside the TD, sometimes not), and to get the value if it is inside a DIV element.

The string can be :

string text = @"
<P>
    tset if it work 
    <SPAN onresizestart="return false" ondrag=javascript:dragActif(); contentEditable=false style="BACKGROUND-COLOR: #c0d4e6" edth_type="var" edth_var_pob="n" edth_var_pgm="RBLZVALO" edth_var_def="B" edth_var_casse="car" edth_var_lg="050" edth_var_type="c" edth_var_nom="Adr_Leg_Lig1" edth_var_lib="Ligne 1 adresse légale" edth_var_libc="Adr_Leg_Lig1" edth_var_num="1" edth_var_posFich="0">
       Adr_Leg_Lig1
    </SPAN> 
    test
</P>
<P>This <FONT size=2 edth_sizeUTIA4="8">should</FONT> Work
</P>"

I have try to parse it to XML with

XmlDocument xd = new XmlDocument();
xd.LoadXML(text);

but it failed, I try to parse it to HTML with this ParseHTML but it failed too.

I tried filtering every possibility with regex, but some times we have to take what is inside <FONT> like in this example, sometimes we don't.

Is there a way to convert this to HTML on server side with ASP.NET or to convert it to some sort of XML that I could use to manipulate it with its tag and attribute inside the tag ?

EDIT: ASP.NET 2.0, IE5, No Jquery (well IE5) are my configuration, I can't use external libraries.

Answer 1

It is unlikely, but using the code you provided, if you made a few minor changes you could use

string text = "<ME><P>tset if it work <SPAN onresizestart='return false' ondrag='javascript:dragActif();' contentEditable='false' style='BACKGROUND-COLOR: #c0d4e6' edth_type='var' edth_var_pob='n' edth_var_pgm='RBLZVALO' edth_var_def='B' edth_var_casse='car' edth_var_lg='050' edth_var_type='c' edth_var_nom='Adr_Leg_Lig1' edth_var_lib='Ligne 1 adresse légale' edth_var_libc='Adr_Leg_Lig1' edth_var_num='1' edth_var_posFich='0'>Adr_Leg_Lig1</SPAN> test</P><P>This <FONT size='2' edth_sizeUTIA4='8'>should</FONT> Work</P></ME>";

XDocument xd = XDocument.Parse(text);

I had to wrap it with a nonsense tag (<ME>) or else it will have multiple roots

I've also had to make sure where you have = you have 2 ' ' after! eg your original had:

This <FONT size=2 

Where I used

This <FONT size='2' 

Screenshot using the XML visualizer

Answer 2

Parsing HTML can is generally difficult and there are many edge cases to think off, so I would recommend the use of an external library like HTMLAgilityPack. If your client does not allow external libraries, you can just download the source code for this and include the relevant projects in your solution.

Using HTMLAgilityPack and the code snippet below I get the following output:

test if it work test
this should Work

You may need to filter for additional elements and to tweak the XPath expression to be more specific.

using System;
using System.Linq;
using HtmlAgilityPack;

namespace MongoDB
{

    public class Program
    {

        public static void Main()
        {
            string text =
                "<p>tset if it work <span onresizestart=\"return false\" ondrag=javascript:dragActif(); contenteditable=false style=\"BACKGROUND-COLOR: #c0d4e6\" edth_type=\"var\" edth_var_pob=\"n\" edth_var_pgm=\"RBLZVALO\" edth_var_def=\"B\" edth_var_casse=\"car\" edth_var_lg=\"050\" edth_var_type=\"c\" edth_var_nom=\"Adr_Leg_Lig1\" edth_var_lib=\"Ligne 1 adresse légale\" edth_var_libc=\"Adr_Leg_Lig1\" edth_var_num=\"1\" edth_var_posfich=\"0\">Adr_Leg_Lig1</span> test</p><p>This <font size=2 edth_sizeutia4=\"8\">should</font> Work</p>";

            HtmlDocument html = new HtmlDocument();
            html.LoadHtml(text);

            var nodes = html.DocumentNode.SelectNodes("//p");

            foreach (
                var line in
                    nodes.Select(node => node.ChildNodes.Where(childNode => childNode.Name!="span"))
                        .Select(
                            textNodes => textNodes.Aggregate(String.Empty, (current, node) => current + node.InnerText))
                )
            {
                Console.WriteLine(line);
            }
        }
    }
}

Answer 3

HTML is not usually valid XML. You need to use a parser that can parse HTML from strings from which you can extract content.

I do a lot of web-scraping and found that CSQuery does the trick nicely. It converts HTML into an in-memory DOM that can be queried using functions/selectors just like the ones jQuery provide.

Answer 4

This thread will tell you how to parse a html file with a regex or with the agility pack: How do I remove all HTML tags from a string without knowing which tags are in it?