Hello

I have created a basic RSS news reader which is fine apart from one xml file - its actually the BBC News. They seem to have unicode characters within the xml file (upside down question marks and Pound signs) which causes a problem with the MSXML parser it basically dosen't create a document object model :( If I manually take out these charcters it works a treat.

Anybody help with this?

Thanks
Phil

Can you point to the RSS link? These problems are normally caused by encoding incorrectly declared or the process changing it at some stage. Can you show some of your code?

Joe (MVP - xml)

The url for the rss is http://news.bbc.co.uk/rss/newsonline...lth/rss091.xml

My code is

<code>
url = request.querystring("url")
call getXmlNews(url)

Sub getXmlNews(address)

Set oXMLHttp = Server.CreateObject("MSXML2.ServerXMLHTTP")

        dim URL
        dim i
        dim j
        dim root

    URL= address
          oXMLHttp.open "GET", URL, false
         oXMLHttp.send() ' Send the request.'

      If oXMLHttp.status = 200 Then

                sXML = oXMLHttp.responseText ' Retrieve from serverB.'
                removeDoctype(sXML)

                set oXML = Server.CreateObject("MSXML2.DomDocument")
                'response.write(sXML)'

                oXML.loadXML(sXML)

'*** create Title info'

    newstitle = oXML.getElementsByTagName("title").item(0).text
    response.write("<div id=""maintitle"">"&newstitle&"</div>")
    response.write("")



'*** Write news items'

                Set objLst = oXML.getElementsByTagName("item")
                intNoOfHeadlines = objLst.length -1
                'response.write(intNoOfHeadlines)'
                strContent = ""

             For i = 0 To (intNoOfHeadlines)

strContent = strContent &"<div class=""main"">"
                 Set objHdl = objLst.item(i) 'loop through each item'

                    for j = 0 to objHdl.childNodes.length -1


                        If objHdl.childNodes(j).nodeName = "title" Then
                            strContent = strContent &"<div id=""title"">"&objHdl.childNodes(j).text&"</div>"
                        End If

                        If objHdl.childNodes(j).nodeName = "description" Then
                            strContent = strContent &"<div id=""description"">"&objHdl.childNodes(j).text&"</div>"
                        End If


                        If objHdl.childNodes(j).nodeName = "link" Then
                            strContent = strContent &"<div id=""link""><a href="""&objHdl.childNodes(j).text&" "" target=""blank"">View Item</a></div>"
                        End If

                        If objHdl.childNodes(j).nodeName = "pubDate" Or objHdl.childNodes(j).nodeName = "dc:date" Then
                            strContent = strContent &"<div id=""pubdate"">Document Published: "&objHdl.childNodes(j).text&"</div>"
                        End If

                    Next

                strContent = strContent &"</div>"


            Next

strContent = strContent &"<br /><br />" 'make alittle space at bottom of page'

     Else

         strContent = "Sorry...the news source is temporarily unavailable - Please try again later. " '&oXMLHttp.status'

   End if


Response.Write(strContent) 'Write out the content'

     set oXML = Nothing
    set oXMLHttp = Nothing

End Sub
</code>

Thanks for your quick response
phil

I will have a closer look if you wish but at first glance the problem is this. The site is using an encoding of ISO-8859-1. When ever you write string in VB, via Response.Write in this case, you are writing a unicode string encoded as utf-16. You are also using responseText property, if the xml is well formed you should be using oXMLHttp.responseXML. This is breaking the encoding. The most elegant way to fix this would be to abandon the dom parsing and use XSLT to transform to HTML. You might get away with setting the Response.CharSet to "ISO-8859-1", you might need to set the Response.CodePage as well.
As a test try this after posting
As a secondary point you'll get more consistent performance by using the correct progids. If you have version four parser use:
MSXML2.ServerXMLHTTP.4.0 and
MSXML2.DomDocument.4.0
with version 3 change 4 to 3:)





--

Joe

Hey Joe - many thanks for this

I have been considering using XSLT after having a tinker with it.
When you say XSLT do you mean asp with XSLT or just XSLT on its own?

I will play around with your suggestions.

Thanks again
phil

XSLT with ASP, this, combined with using the dom directly and not moving between it and strings, should solve your problems.

--

Joe