Extracting text between tags

aware · #1 (**permalink**) December 17th, 2003, 03:51 AM

Hi

I've an application which is required to display 2 documents (html) from 2 sources on a single page, the sources are a database and static HTML files.

My question is how do I extract the text between the <BODY> tags of the html page?

-------UPDATE--------
I've gotten this far - The following code does what I want but it's not case sensitive on the tags, how can this be done, Also it was a bit of a fudge geting the numbers right - is this normal.

intStart = instr(strP, "<Body>") + 6
intEnd = instr(strP, "</Body>") - 7
strText = mid(strP, intStart, intEnd)
Response.Write strText

------FURTHER UPDATE--------------
Not very elagant - can anyone improve on this.
It works for now.

if instr(strP, "<body>") > 0 then '1
intStart = instr(strP, "<body>")
else
if instr(strP, "<Body>") > 0 then '2
intStart = instr(strP, "<Body>")
else
if instr(strP, "<BODY>") > 0 then '3
intStart = instr(strP, "<BODY>")
else
intstart = 0
End if '3
End if '2
End if '1

if instr(strP, "</body>") > 0 then '1
intEnd = instr(strP, "</body>")
else
if instr(strP, "</Body>") > 0 then '2
intEnd = instr(strP, "</Body>")
else
if instr(strP, "</BODY>") > 0 then '3
intEnd = instr(strP, "</BODY>")
else
intEnd = 1000
End if '3
End if '2
End if '1

intStart = intStart + 6
intEnd = IntEnd - 7

strText = mid(strP, intStart, intEnd)
Response.Write strText

Thanks in advance

Andy Green

pgtips · #2 (**permalink**) December 18th, 2003, 05:40 AM

Andy, do you know anything about regular expressions? they are tailor-made for things like this.

aware · #3 (**permalink**) December 19th, 2003, 06:29 AM

Hi

No, but now I know what to look for I'll see what I can find out.

Thanks

pgtips · #4 (**permalink**) December 23rd, 2003, 06:53 AM

Its a bit of a tricky regular expression to match a body tag which may or may not have attributes, and also which may be spread over multiple lines. Try this one:

Code:

' the text between the <BODY></BODY> tags will be stored in variable sBodyText
Set re = New RegExp
re.Pattern = "<body(?:.|\n)*?>((?:.|\n)*)</body>"
re.IgnoreCase = True
re.Global = False
re.Multiline = True
Set oMatches = re.Execute(strP)
If Not oMatches Is Nothing Then
    If oMatches.Count > 0 Then
        If Not oMatches(0).SubMatches Is Nothing Then
            sBodyText = oMatches(0).SubMatches(0)
        End If
    End If
End If

I'll try and explain the regexp:
<body - matches the opening body tag

(?:.|\n)*? - is for attributes, it matches any char, including new-line, zero or more times (the opening ?: prevents the characters here from being stored in the submatches)

((?:.|\n)*) - same again, except we want to capture the whole text between the tags in the submatches collection

</body> - matches the ending body tag

hth
Phil

aware · #5 (**permalink**) December 24th, 2003, 04:25 AM

Hi

Thanks for this.

I've done some research on Regular Expressions and played with RegExp.

I got my code to extract text but could not get the syntax correct for all cases of <body......>.

I'll give this a try. I'm now a convert to RE's

Thanks again

Andy G

Need to download code?

Navigation