Re: [htdig-dev] "file name.html" -> "filename.html";(

[Joe R. J. <jj...@cl...>]

On Fri, 15 Mar 2002, Gilles Detillieux wrote:

> Date: Fri, 15 Mar 2002 16:34:36 -0600 (CST)
> From: Gilles Detillieux <gr...@sc...>
> To: jj...@cl...
> Cc: "ht://Dig developers list" <htd...@li...>
> Subject: Re: [htdig-dev] "file name.html" -> "filename.html";(
> 
> No, the code below does two things: 1) if allow_space_in_url is not
> set, the code works like the standard 3.1.x code does, i.e. in strips
> out all white space characters, and 2) if allow_space_in_url is set,
> the code strips out all white space characters other than the space
> itself - for the space character (ASCII 20 hex) it strips leading and
> trailing spaces and converts the spaces within the URL to %20.  The name
> allow_space_in_url is correct, because if the attribute is false,
> no spaces are allowed - they're stripped out, just as the currently
> released code does, in accordance with RFC 2396.  However, if you prefer
> encode_space_in_url we can go with that.  We're not going to start putting
> all sorts of wierd punctuation characters like "%" in attribute names.
> 
> > >     static int	allowspace = config.Boolean("allow_space_in_url", 0);
> > >     String	temp;
> > >     while (*ref)
> > >     {
> > > 	if (*ref == ' ' && temp.length() > 0 && allowspace)
> > > 	{
> > > 	    // Replace space character with %20 if there's more non-space
> > > 	    // characters to come...
> > > 	    char *s = ref+1;
> > > 	    while (*s && isspace(*s))
> > > 		s++;
> > > 	    if (*s)
> > > 		temp << "%20";
> > > 	}
> > > 	else if (!isspace(*ref))
> > > 	    temp << *ref;
> > > 	ref++;
> > >     }
> 
> Maybe my description of the code above helps you see the rationale more
> clearly.  The attribute selects both behaviours, not just the encoding.
> The reason to make it user-selectable option is that some users may
> actually prefer htdig to follow the standards rather than ignore them
> like MS/AOL do.
> 
> I'm not sure what you mean by integrating my option in the entire patch.
> The code above should be complete on its own, as a change to vanilla
> 3.1.6 URL.cc code.  You don't need to integrate it with earlier proposed
> changes - just put it in both URL methods you were changing before and
> make a patch out of it.

I misunderstood.  Here is the patch:
-------------------------------------8<-------------------------------------
*** htlib/URL.cc.031202	Thu Feb  7 17:15:38 2002
--- htlib/URL.cc	Fri Mar 15 15:25:27 2002
***************
*** 74,82 ****
  //
  URL::URL(char *ref, URL &parent)
  {
!     String	temp(ref);
!     temp.remove(" \r\n\t");
!     ref = temp;
  
      _host = parent._host;
      _port = parent._port;
--- 74,97 ----
  //
  URL::URL(char *ref, URL &parent)
  {
!     static int  allowspace = config.Boolean("allow_space_in_url", 0);
!     String      temp;
!     while (*ref)
!     {
!         if (*ref == ' ' && temp.length() > 0 && allowspace)
!         {
!             // Replace space character with %20 if there's more non-space
!             // characters to come...
!             char *s = ref+1;
!             while (*s && isspace(*s))
!                 s++;
!             if (*s)
!                 temp << "%20";
!         }
!         else if (!isspace(*ref))
!             temp << *ref;
!         ref++;
!     }
  
      _host = parent._host;
      _port = parent._port;
***************
*** 243,255 ****
  }
  
  //*****************************************************************************
! // void URL::parse(char *u)
  //   Given a URL string, extract the service, host, port, and path from it.
  //
! void URL::parse(char *u)
  {
!     String	temp(u);
!     temp.remove(" \t\r\n");
      char	*nurl = temp;
  
      //
--- 258,286 ----
  }
  
  //*****************************************************************************
! // void URL::parse(char *ref)
  //   Given a URL string, extract the service, host, port, and path from it.
  //
! void URL::parse(char *ref)
  {
!     static int  allowspace = config.Boolean("allow_space_in_url", 0);
!     String      temp;
!     while (*ref)
!     {
!         if (*ref == ' ' && temp.length() > 0 && allowspace)
!         {
!             // Replace space character with %20 if there's more non-space
!             // characters to come...
!             char *s = ref+1;
!             while (*s && isspace(*s))
!                 s++;
!             if (*s)
!                 temp << "%20";
!         }
!         else if (!isspace(*ref))
!             temp << *ref;
!         ref++;
!     }
      char	*nurl = temp;
  
      //
-------------------------------------8<-------------------------------------
But it failed to follow any link;(I must have misread your instructions;)
any ideas?

Regards,

Joe
-- 
     _/   _/_/_/       _/              ____________    __o
     _/   _/   _/      _/         ______________     _-\<,_
 _/  _/   _/_/_/   _/  _/                     ......(_)/ (_)
  _/_/ oe _/   _/.  _/_/ ah        jj...@cl...