httpd-cvs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dirk-Willem van Gulik <>
Subject Re: cvs commit: httpd-docs-1.3/htdocs/manual index.html.ja.jis
Date Sat, 30 Dec 2000 03:11:12 GMT


I am holding out on that change; and in fact almost object to it !

As both the HTTP/1.x spec (rfc2616) say that the string is case-sensitive:

	5.1.1 Method

	The Method token indicates the method to be performed on the resource
	identified by the Request-URI. The method is case-sensitive. 
i.e. on the wire it needs to be 'POST'. (And we certainly do a strcmp() in
the method <-> number translation function (though we do not nicely to
a reverse mapping to make it really safe). So any browser which passed

	<form action="poSt">

on as anything else but POST is arguably wrong. But we could be leisure in
what we accecpt; (though we do not currently do that too nicely).

Secondly the CGI/1.1 specification (The old NCSA one, Robinsons '95 one
and the recent one on Ken his web site) says that again the string is case


	The REQUEST_METHOD metavariable is set to the method with which the
	request was made, as described in section 5.1.1 of the HTTP/1.0
	specification [3] and section 5.1.1 of the
	HTTP/1.1 specification [8]. 

	    REQUEST_METHOD   = http-method
	    http-method      = "GET" | "HEAD" | "POST" | "PUT" | "DELETE"
                       | "OPTIONS" | "TRACE" | extension-method
	    extension-method = token
	The method is case sensitive. CGI/1.1 servers MAY choose to process some
	methods directly rather than passing them to scripts. 

So IMHO either the change below is bogus OR we are doing something badly
wrong inside apache. Or we need to chanmge the spec.

Plus because we do not do a carefull reverse mapping inside apache I worry
that making the script case-insensitive would have silly side effects with
things such as limit and so on.

Having said that; I will chekc in the searhc script later today into CVS
and fix it with the rearranging of the site so it properly indexes again.


On 30 Dec 2000 wrote:

> wrowe       00/12/29 18:52:27
>   Modified:    htdocs/manual index.html.ja.jis
>   Log:
>     The last two patches fixed the .en page, but not the .ja.jis encoded
>     one, as O is possibly a state transition.  Since there is a headache
>     here, I'm reverting to "post" and await a fixed search script.
>   Revision  Changes    Path
>   1.4       +1 -1      httpd-docs-1.3/htdocs/manual/index.html.ja.jis
>   Index: index.html.ja.jis
>   ===================================================================
>   RCS file: /home/cvs/httpd-docs-1.3/htdocs/manual/index.html.ja.jis,v
>   retrieving revision 1.3
>   retrieving revision 1.4
>   diff -u -r1.3 -r1.4
>   --- index.html.ja.jis	2000/12/29 23:37:45	1.3
>   +++ index.html.ja.jis	2000/12/30 02:52:27	1.4
>   @@ -48,7 +48,7 @@
>    <table cellspacing="0" cellpadding="0" border="0" width="600">
>    <tr><td align="center">
>   -   <form method="POST" action="">
>   +   <form method="post" action="">
>       <input type="hidden" name="what" value="ap">
>       <input type="hidden" name="results" value="20">
>       <input type="text" name="keyword" size="20">

View raw message