Parsing HTML document to get at plain text between the <P> tags

[Rob]

I'm writing an application without a user interface and I have a requirement
to find the plain ASCII text between <Ptags in an HTML document which
happens to have been obtained via POP3 and parsed out of a MIME message
body.

If I had a user interface, I could drop and IE web control onto a form, load
the HTML into that and then use the document parser. I've always hated that
route as it's so clunky and anyway, this app doesn't have a user interface
(it's processing emails in a service).

Any suggestions how to go about this?

I've come across SgmlReader from here:

http://www.gotdotnet.com/Community/U...4-C3BD760564BC

This could convert the HTML document into a clean XML document which I could
then parse with the dotnet XML objects. Does this sound a sensible route?

I thought implementing a POP3 reader (easy) and internet email MIME parser
(mindblowing) was difficult enough :-)

Thanks, Rob.

[Andrew Brook]

Could you just use a regular expression to pick out the pieces of text you
want?

Andrew

"Rob" <rob_nicholson@nospam_unforgettable.comwrote in message
news:ze******************@newsfe5-win.ntli.net...

I'm writing an application without a user interface and I have a
requirement to find the plain ASCII text between <Ptags in an HTML
document which happens to have been obtained via POP3 and parsed out of a
MIME message body.

If I had a user interface, I could drop and IE web control onto a form,
load the HTML into that and then use the document parser. I've always
hated that route as it's so clunky and anyway, this app doesn't have a
user interface (it's processing emails in a service).

Any suggestions how to go about this?

I've come across SgmlReader from here:

http://www.gotdotnet.com/Community/U...4-C3BD760564BC

This could convert the HTML document into a clean XML document which I
could then parse with the dotnet XML objects. Does this sound a sensible
route?

I thought implementing a POP3 reader (easy) and internet email MIME parser
(mindblowing) was difficult enough :-)

Thanks, Rob.

[Martin Honnen]

Rob wrote:

I've come across SgmlReader from here:

http://www.gotdotnet.com/Community/U...4-C3BD760564BC

This could convert the HTML document into a clean XML document which I could
then parse with the dotnet XML objects. Does this sound a sensible route?

Yes, use that route to parse the HTML in a way that you can chain it to
existing XML/XPath APIs in .NET. You can either use SgmlReader or HTML
agility pack
<http://www.codeplex.com/Wiki/View.aspx?ProjectName=htmlagilitypack>
--

Martin Honnen --- MVP XML
http://JavaScript.FAQTs.com/

[Peter Flynn]

Rob wrote:

I'm writing an application without a user interface and I have a requirement
to find the plain ASCII text between <Ptags in an HTML document which
happens to have been obtained via POP3 and parsed out of a MIME message
body.

Use Tidy to make it XHTML and then run XSLT to extract the paragraph text.

///Peter
--
XML FAQ: http://xml.silmaril.ie/

[Rob]

Could you just use a regular expression to pick out the pieces of text you

want?

Maybe... I'm not an expert on regular expressions - any references for this
particular use of regular expressions? I could just about formulate an
expression to get the text between <Pand </Pbut I also want to strip out
other tags like:

<p>Some <b>text</bstring</p>

I'm just interested in the "Some text string" bit.

Thanks, Rob.

[Rob]

<http://www.codeplex.com/Wiki/View.aspx?ProjectName=htmlagilitypack>

Thanks for the reference.

Cheers, Rob.

[Rob]

Thanks for the reference.

Later... the HTMLAgilityPack looks, at first glance, just what I need.

Cheers, Rob.

[Andrew Brook]

hehe ok - maybe thinking about this a little more would mean a fairly
complicated regular expression, especially if there are tags nested inside
:)

Andrew

"Rob" <rob_nicholson@nospam_unforgettable.comwrote in message
news:gr*******************@newsfe4-win.ntli.net...

>Could you just use a regular expression to pick out the pieces of text
you want?

Maybe... I'm not an expert on regular expressions - any references for
this particular use of regular expressions? I could just about formulate
an expression to get the text between <Pand </Pbut I also want to
strip out other tags like:

<p>Some <b>text</bstring</p>

I'm just interested in the "Some text string" bit.

Thanks, Rob.

[JohnAD]

kind of raw but you can remove anything that starts with < and ends with >
and everything in between. You might have to take care of any HTML tags that
are escaped e.g. &gt; for etc.

"Rob" <rob_nicholson@nospam_unforgettable.comwrote in message
news:ze******************@newsfe5-win.ntli.net...

I'm writing an application without a user interface and I have a
requirement to find the plain ASCII text between <Ptags in an HTML
document which happens to have been obtained via POP3 and parsed out of a
MIME message body.

If I had a user interface, I could drop and IE web control onto a form,
load the HTML into that and then use the document parser. I've always
hated that route as it's so clunky and anyway, this app doesn't have a
user interface (it's processing emails in a service).

Any suggestions how to go about this?

I've come across SgmlReader from here:

http://www.gotdotnet.com/Community/U...4-C3BD760564BC

This could convert the HTML document into a clean XML document which I
could then parse with the dotnet XML objects. Does this sound a sensible
route?

I thought implementing a POP3 reader (easy) and internet email MIME parser
(mindblowing) was difficult enough :-)

Thanks, Rob.

[Rob]

hehe ok - maybe thinking about this a little more would mean a fairly

complicated regular expression, especially if there are tags nested inside

Actually, maybe not. I can just start to envisage an expression that finds
any <??>text</??tag and then keep calling recursively. Might work.

Then again, there are various HTML -XML libraries which might work even
better.

Cheers, Rob.