URLDecode

[googletired]

Recently i have engaged a company to track my articles across the net. at this point we have submitted them a few RSS feeds to load our content into their system for intial testing and what have you, nothing wrong with my RSS feeds, they are in standard .XML format UTF-8 encoding applied (yes i made i sure they were saved as such too.) Anyways they can not reach my content on the site because of "&"

now after some brief research i have come across URLDecode is the best way to handle this problem, OR URL Rewrite through the IIS.

So i guess my question is which is best, AND if URLDecode is the chosen method to handle this request what is the string that i have to pass to fix it.

Thanks in advance.

[jhardman]

check out this article
Jared

[googletired]

check out this article
Jared

handy little article thank you, but i guess i know less about urldecode than i thought :)


Below i have pasted the code from my localvars.asp which handles all the pages for my site, Do i need to add the url decode string here? or is it something that has to be added to everypage?

(sample link of how my site proceeds, notice all the "&" in the RSS feed these are all read as "&")

http://test.testsite.com/default.asp...ontent=summary
&type=print&lang=en&vol=XXXXXX&cat=X

http://test.testsite.com/default.asp...ent=getcontent
&type=print&lang=en&vol=XXXXXX&cat=X&articleID=XXX XXX&mode=view&id=

<% 
DIM strContent, strFileID, strType, strLang, strVol, strCat, strArticleID, strKeyword,_ 
        strThisRedirect, strDownload, strMode, strPageName, strBaseURL, strBaseQueryString, _
        strhttp_host, strSection
 
strSection = request("section")
strhttp_host = "http://" & request("http_host")
strContent = lcase(cstr(Request.QueryString("content")))
strFileID = lcase(cstr(Request.QueryString("id")))
strType = lcase(cstr(Request.QueryString("type")))
if strType = "" and strFileID <> "" Then strType = "digital"
strLang = lcase(Request.QueryString("lang"))
if strLang = "" Then strLang = "en"
strVol = lcase(cstr(Request.QueryString("vol")))
'if strVol = "" Then
'     If strType = "digital" then 
'             strVol = "current"
'     Else
'             strVol = ""
'     End If     
'End If
strCat = lcase(Request.QueryString("cat"))
strArticleID = Request.QueryString("articleid")
If strFileID <> "" and strArticleID = "" Then strArticleID = strFileID 
'Bridge code to support the absense of the content field in the url
if strContent = "" Then
     IF strArticleID <> "" Then 
             strContent = "getcontent"
     else
        strContent = "defcontent"
     End if
End if
strKeyword = Request.Form("keyword")
strThisRedirect = lcase(Request("QUERY_STRING"))
strDownload = lcase(Request.QueryString("download"))
strMode = lcase(Request.QueryString("mode"))
strPageName = "/scripts/" & lcase(Request.QueryString("pagename")) & ".asp"
if strPageName = "/scripts/.asp" THEN strPageName = "/scripts/default_content.asp"
strBaseURL = "default.asp?pagename=media"
strBaseQueryString = "&amp;type=" & strType & "&amp;lang=" & strLang
%>

Hope you can help with this.

[jhardman]

Notice that your line 40 includes "&". This is the problem (as far as I can tell). The thing is, when we include complex querystrings in URLs in links in a website, we are supposed to use "&" rather than "&" (according to the w3c). This is supposed to protect against browsers misinterpreting some querystring as a complex character code. (imagine if someone had a querystring variable named "quot" and you can imagine how some browsers might interpret "&quot"). So there is an inherent conflict in how you need to write your pages for use with a standard browser and how they should be written for this service.

In my experience, I haven't seen a browser slip up on this problem in years so I would suggest erring in favor of the search service. I don't think any current browsers will interpret a character code if it doesn't include a final semi-colon, but I have seen this problem in the past and some fossils do use older browsers. Just to be sure, make sure none of your querystring variables are inadvertently HTML character code names, and then change all of your "&"s to "&". This will not pass validation, it will definitely fail if you send it to w3c, but I would be very surprised if a browser ever got it wrong.

Jared

[googletired]

Yes i agree, but the problem lies within the company that crawls the net in search of the content, we submit an RSS feed to them and in the links possibly the text at times & is replaced with & to make the feed vailid, but when they load these feeds into their webcrawler/sniffer they cannot reach the content because the website (I think) does not handle full url Decoding.

and on line 40 i guess it's the website parsing those & but i should say & blah & srtblah

I am in process of trying a simple fix, i'll advise if it works.