Regex with HTML tags - html

Question

I have this regular expression:
(\S+)=[""']?((?:.(?![""']?\s+(?:\S+)=|[>""']))+.)[""']?
This regex expression will extract the name of the tag and the value from HTML string, everything is working fine, but, when I have a single char the regex will trap the left side quote and the character.
This is my string:
<select title="Campo" id="6:7" style="width: auto; cursor: pointer;" runat="server" controltype="DropDownList" column="Dummy_6"><option value="0">Value:0</option><option selected="selected" value='1'>Value:1Selected!</option></select>
I don't know how to modify this regex expression to capture the char correctly even there is only one character.

Answer 1

You should be using HTML parser for this task, regex cannot handle HTML properly.
To collect all tag names and there attribute names and values, I recommend the following HtmlAgilityPack-based solution:
var tags = new List<string>();
var result = new List<KeyValuePair<string, string>>();
HtmlAgilityPack.HtmlDocument hap;
Uri uriResult;
if (Uri.TryCreate(html, UriKind.Absolute, out uriResult) && uriResult.Scheme == Uri.UriSchemeHttp)
{ // html is a URL
var doc = new HtmlAgilityPack.HtmlWeb();
hap = doc.Load(uriResult.AbsoluteUri);
}
else
{ // html is a string
hap = new HtmlAgilityPack.HtmlDocument();
hap.LoadHtml(html);
}
var nodes = hap.DocumentNode.Descendants().Where(p => p.NodeType == HtmlAgilityPack.HtmlNodeType.Element);
if (nodes != null)
foreach (var node in nodes)
{
tags.Add(node.Name);
foreach (var attribute in node.Attributes)
result.Add(new KeyValuePair<string, string>(attribute.Name, attribute.Value));
}

Answer 2

I think you're trying something overly intricate and, ultimately, incorrect, with your regex.
If you want to naively parse an HTML attribute: this regex should do the trick:
(\S+)=(?:"([^"]+)"|'([^']+)')
Note that it parses single-quoted and double-quoted values in different legs of the regex. Your regex would find that in the following code:
<foo bar='fu"bar'>
the attribute's value is fu when it really is fu"bar.

Answer 3

There are better ways to parse HTML, but here's my take at your question anyway.
(?<attr>(?<=\s).+?(?==['"]))|(?<val>(?<=\s.+?=['"]).+?(?=['"]))
Without capture group names:
((?<=\s).+?(?==['"]))|((?<=\s.+?=['"]).+?(?=['"]))
quotes included:
((?<=\s).+?(?==['"]))|((?<=\s.+?=)['"].+?['"])
Update: For more in-depth usage, do give HTML Agility Pack a try.