Regular expressions and HTML

[Kelsang Wangchuk]

Hi

I'm working on a method that will transform an HTML fragment into an
XHTML fragment. I thought I'd have a go at using regular expressions
in .NET, because I don't really understand the numerous regular
expression classes in .NET and I thought it might be a good
opportunity to learn.

I want to break down the HTML into element names, attribute names,
attribute values and element content. The regular expression I came up
with, that successfully matches fragments such as <div contenteditable
id="test">, looks like this:

^(?<precedingText>[^<>]*)((((<(?<elementName>[a-zA-Z_][a-zA-Z0-9_]*)\s*((?<=\s)(?<attributeName>[a-zA-Z_][a-zA-Z0-9_]*)((\s*=\s*""(?<attributeValue>[^""]*)""\s*)|(\s*=(?<attributeValue>[^\s"">]+)\s*)|(\s*(?!\s*=))))*\s*>))|(</(?<closingElementName>[a-zA-Z_][a-zA-Z0-9_]*)\s*>))(?<text>[^<>]*))*$

A bit complicated I know, but it works. However, now I don't know how
to extract the various elementName, attributeName etc. group values in
order to match them up. All seems a bit wild to me.

Can anyone tell me the best way to construct a regular expression in
order to produce the desired result? I suspect using constructs such
as (?<elementName>....)* doesn't work so well because of the repeat,
but how else could it be done?

Cheers, Wangchuk

 

[Jerry III]

There was an SGML parser on MSDN but I can't find the link anymore... But
that's what you could use, it's a lot easier and cleaner solution to your
problem...

Jerry

Hi

I'm working on a method that will transform an HTML fragment into an
XHTML fragment. I thought I'd have a go at using regular expressions
in .NET, because I don't really understand the numerous regular
expression classes in .NET and I thought it might be a good
opportunity to learn.

I want to break down the HTML into element names, attribute names,
attribute values and element content. The regular expression I came up
with, that successfully matches fragments such as <div contenteditable
id="test">, looks like this:

^(?<precedingText>[^<>]*)((((<(?<elementName>[a-zA-Z_][a-zA-Z0-9_]*)\s*((?<=
\s)(?<attributeName>[a-zA-Z_][a-zA-Z0-9_]*)((\s*=\s*""(?<attributeValue>[^""

A bit complicated I know, but it works. However, now I don't know how
to extract the various elementName, attributeName etc. group values in
order to match them up. All seems a bit wild to me.

Can anyone tell me the best way to construct a regular expression in
order to produce the desired result? I suspect using constructs such
as (?<elementName>....)* doesn't work so well because of the repeat,
but how else could it be done?

Cheers, Wangchuk

 

[Kelsang Wangchuk]

Ok, great, thanks very much. I'll have a look.

KW

Found the link... The whole article is at
http://msdn.microsoft.com/library/en-us/dnxmlnet/html/XMLToolsUpdate.asp and
the SGML reader can be downloaded at
http://www.gotdotnet.com/Community/...mpleGuid=B90FDDCE-E60D-43F8-A5C4-C3BD760564BC.
It will allow you to read an SGML document (such as HTML) into an
XmlDocument which is pretty much what you want, isn't it?

Jerry

There was an SGML parser on MSDN but I can't find the link anymore... But
that's what you could use, it's a lot easier and cleaner solution to your
problem...

Jerry

Hi

I'm working on a method that will transform an HTML fragment into an
XHTML fragment. I thought I'd have a go at using regular expressions
in .NET, because I don't really understand the numerous regular
expression classes in .NET and I thought it might be a good
opportunity to learn.

I want to break down the HTML into element names, attribute names,
attribute values and element content. The regular expression I came up
with, that successfully matches fragments such as <div contenteditable
id="test">, looks like this:

]*)""\s*)|(\s*=(?<attributeValue>[^\s"">]+)\s*)|(\s*(?!\s*=))))*\s*>))|(</

(?

A bit complicated I know, but it works. However, now I don't know how
to extract the various elementName, attributeName etc. group values in
order to match them up. All seems a bit wild to me.

Can anyone tell me the best way to construct a regular expression in
order to produce the desired result? I suspect using constructs such
as (?<elementName>....)* doesn't work so well because of the repeat,
but how else could it be done?

Cheers, Wangchuk