Re: bug

From: <jhorman@lycos-inc.com> Date: Tue, 23 May 2000 14:26:14 -0400 To: "Fred Bone" <fred.bone@dial.pipex.com> cc: html-tidy@w3.org Message-ID: <852568E8.00654BDC.00@SMTPNotes1.ma.lycos.com> · This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:37:48 GMT

The application I am writing runs tidy and then programmatically extracts the
hrefs from the resulting tidied document and spiders those hrefs. The spider was
not replacing &amp; with & before sending the http request. I will do the
replacement inside the spider library. I just assumed that urls within hrefs
would be exactly the same after running jtidy.

-jason

|--------+-------------------------->
|        |          "Fred Bone"     |
|        |          <fred.bone@dial.|
|        |          pipex.com>      |
|        |                          |
|        |          05/23/2000 01:56|
|        |          PM              |
|        |                          |
|--------+-------------------------->
  >----------------------------------------------------------------------------|
  |                                                                            |
  |       To:     Jason Horman/Lycos@Lycos                                     |
  |       cc:     html-tidy@w3.org                                             |
  |       Subject:     Re: bug                                                 |
  >----------------------------------------------------------------------------|

On 23 May 2000, at 13:48, jhorman@lycos-inc.com wrote:

> No, the proper URL querystring should be & delimited not &amp; delimited.
> QueryString parsers such as asp would interpret the url as having a query
param
> called "amp;BV_EngineID", thereby breaking the cgi apps which these urls point
> to. Your right in that it is the proper xml encoding but not html.

The querystring sent by a browser will have had the "&amp;"
interpreted into "&".