XHTML and the & in url's.

I've been trying to convert my habits over to XHTML.

One stumbling block, does one have to use & in URL's to separate
form variables? <a href="foo.cgi?q=1&name=jerry">foo</a>

Of course, &name is choked on, because it's an entity reference w/out
a semicolon.

One thing I've seen CGI.pm (from the perl world) do is accept semi-colons:

<a href="foo.cgi?q=1;name=jerry">foo</a>

This makes more sense to me, but is it an actual standard? if so, how would
I get PHP to use it instead of &? (without any editing of PHP's .ini files)

Going through and changing & to &amp; in my URL's would probably break lots
of things and be harder to maintain. Short of parsing the query string myself,
there doesn't seem to be a lot of alternatives.

Ideas? Any other solutions?

Jamie

--
http://www.geniegate.com Custom web programming
guhzo_42@lnubb.pbz (rot13) User Management Solutions

nospam@geniegate.com wrote:

> I've been trying to convert my habits over to XHTML.
>
> Of course, &name is choked on, because it's an entity reference w/out
> a semicolon.
>
> One thing I've seen CGI.pm (from the perl world) do is accept semi-colons:
>
> <a href="foo.cgi?q=1;name=jerry">foo</a>
>
> This makes more sense to me, but is it an actual standard? if so, how
> would I get PHP to use it instead of &? (without any editing of PHP's .ini
> files)

Have a look at the arg_separator.input config directive

>
> Going through and changing & to &amp; in my URL's would probably break
> lots of things and be harder to maintain. Short of parsing the query
> string myself, there doesn't seem to be a lot of alternatives.

Why would changing & to &amp; in your URLs in the HTML documents break
anything? The browser won't actually send &amp;, it will send just &.


--
Tommy

According to the XHTML specs, you should never use an ampsersand as a
literal, you should always convert it to an entity reference (i.e.
&amp when using it in XHTML documents.

See http://www.w3.org/TR/xhtml1/#C_12.

Oli

Tommy Gildseth wrote:
> nospam@geniegate.com wrote:
>
>
>>I've been trying to convert my habits over to XHTML.
>>
>>Of course, &name is choked on, because it's an entity reference w/out
>>a semicolon.
>>
>>One thing I've seen CGI.pm (from the perl world) do is accept semi-colons:
>>
>><a href="foo.cgi?q=1;name=jerry">foo</a>
>>
>>This makes more sense to me, but is it an actual standard? if so, how
>>would I get PHP to use it instead of &? (without any editing of PHP's .ini
>>files)
>
>
> Have a look at the arg_separator.input config directive
>
>
>>Going through and changing & to &amp; in my URL's would probably break
>>lots of things and be harder to maintain. Short of parsing the query
>>string myself, there doesn't seem to be a lot of alternatives.
>
>
> Why would changing & to &amp; in your URLs in the HTML documents break
> anything? The browser won't actually send &amp;, it will send just &.
>
>

nospam@geniegate.com wrote:
> I've been trying to convert my habits over to XHTML.
>
> One stumbling block, does one have to use &amp; in URL's to separate
> form variables? <a href="foo.cgi?q=1&name=jerry">foo</a>
>
> Of course, &name is choked on, because it's an entity reference w/out
> a semicolon.
>
> One thing I've seen CGI.pm (from the perl world) do is accept semi-colons:
>
> <a href="foo.cgi?q=1;name=jerry">foo</a>
>
> This makes more sense to me, but is it an actual standard? if so, how would
> I get PHP to use it instead of &? (without any editing of PHP's .ini files)
>
> Going through and changing & to &amp; in my URL's would probably break lots
> of things and be harder to maintain. Short of parsing the query string myself,
> there doesn't seem to be a lot of alternatives.
>
> Ideas? Any other solutions?
>
> Jamie
>


<a href="foo.cgi?q=1&amp;name=jerry">foo</a>

--


x theSpaceGirl (miranda)

# lead designer @ http://www.dhnewmedia.com #
# remove NO SPAM to email, or use form on website #

According to the W3C specs on XHTML, if you want an ampersand literal
within an XHTML document, you should always use its entity reference &amp;

See http://www.w3.org/TR/xhtml1/#C_12.

Oli

Tommy Gildseth wrote:
> nospam@geniegate.com wrote:
>
>
>>I've been trying to convert my habits over to XHTML.
>>
>>Of course, &name is choked on, because it's an entity reference w/out
>>a semicolon.
>>
>>One thing I've seen CGI.pm (from the perl world) do is accept semi-colons:
>>
>><a href="foo.cgi?q=1;name=jerry">foo</a>
>>
>>This makes more sense to me, but is it an actual standard? if so, how
>>would I get PHP to use it instead of &? (without any editing of PHP's .ini
>>files)
>
>
> Have a look at the arg_separator.input config directive
>
>
>>Going through and changing & to &amp; in my URL's would probably break
>>lots of things and be harder to maintain. Short of parsing the query
>>string myself, there doesn't seem to be a lot of alternatives.
>
>
> Why would changing & to &amp; in your URLs in the HTML documents break
> anything? The browser won't actually send &amp;, it will send just &.
>
>

Tommy Gildseth wrote:
> nospam@geniegate.com wrote:
>
>
>>I've been trying to convert my habits over to XHTML.
>>
>>Of course, &name is choked on, because it's an entity reference w/out
>>a semicolon.
>>
>>One thing I've seen CGI.pm (from the perl world) do is accept semi-colons:
>>
>><a href="foo.cgi?q=1;name=jerry">foo</a>
>>
>>This makes more sense to me, but is it an actual standard? if so, how
>>would I get PHP to use it instead of &? (without any editing of PHP's .ini
>>files)
>
>
> Have a look at the arg_separator.input config directive
>
>
>>Going through and changing & to &amp; in my URL's would probably break
>>lots of things and be harder to maintain. Short of parsing the query
>>string myself, there doesn't seem to be a lot of alternatives.
>
>
> Why would changing & to &amp; in your URLs in the HTML documents break
> anything? The browser won't actually send &amp;, it will send just &.
>
>

Putting plain ampersand (&) signs in your anchors is a violation of the
W3C (X)HTML specification and pages with them wont validate, if you wish
to write valid (X)HTML you must always use &amp;.

~Cameron

On Sun, 26 Dec 2004, SpaceGirl wrote:

> nospam@geniegate.com wrote:
> > I've been trying to convert my habits over to XHTML.
> >
> > One stumbling block, does one have to use &amp; in URL's to separate
> > form variables? <a href="foo.cgi?q=1&name=jerry">foo</a>
> >
> > Of course, &name is choked on, because it's an entity reference w/out
> > a semicolon.
> >
> > One thing I've seen CGI.pm (from the perl world) do is accept semi-colons:
> >
> > <a href="foo.cgi?q=1;name=jerry">foo</a>
> >
> > This makes more sense to me, but is it an actual standard? if so, how would
> > I get PHP to use it instead of &? (without any editing of PHP's .ini files)
> >
> > Going through and changing & to &amp; in my URL's would probably break lots
> > of things and be harder to maintain. Short of parsing the query string myself,
> > there doesn't seem to be a lot of alternatives.
> >
> > Ideas? Any other solutions?
> >
> > Jamie
> >
>
>
> <a href="foo.cgi?q=1&amp;name=jerry">foo</a>

Indeed. See RFC 1866 - Hypertext Markup Language - 2.0, section 8.2.1 at
one of these locations for more on this:
http://www.faqs.org/rfcs/rfc1866.html
http://info.internet.isi.edu/in-note...es/rfc1866.txt
ftp://ftp.isi.edu/in-notes/rfc1866.txt

It not only requires '&' in URLs to be encoded as '&amp;' but
recommends the use of ';' instead of '&' as a parameter separator:

[snip]
: 8.2.1. The form-urlencoded Media Type
[snip]
: NOTE - The URI from a query form submission can be
: used in a normal anchor style hyperlink.
: Unfortunately, the use of the `&' character to
: separate form fields interacts with its use in SGML
: attribute values as an entity reference delimiter.
: For example, the URI `http://host/?x=1&y=2' must be
^^^^
: written `<a href="http://host/?x=1&y=2"' or `<a
: href="http://host/?x=1&amp;y=2">'.
:
: HTTP server implementors, and in particular, CGI
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
: implementors are encouraged to support the use of
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
: `;' in place of `&' to save users the trouble of
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
: escaping `&' characters this way.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[snip]


A parameter name in a URL that is preceded by an unescaped ampersand can
easily be mistaken by a standards-conforming browser as a character
entity specification (since the trailing ';' in a character entity if the
following character is not a letter or a digit) and completely foul up
any attempt to access the desired page.

I have already seen problems with:

"&image=" (to select an image) being interpreted as (imaginary 'i') + "=",
"&lang=" (language) being interpreted as (left angle bracket) + "=",<br>
"&prod=" (product -- as in commercial product such as toilet paper)
being interpreted as (mathematical product sign -- result of a
series of multiplications) + "=",
"&bull=" (bulletin) being interpreted as (bullet) + "=", and
"&part=" (part -- of a multi-part thread or set of web pages) being
interpreted as (mathematical partial differential) + "=".

Google uses "&num=" to specify the number of entries to display and lynx
(when the optional extended entity set is compiled in) recognises "&num;"
as a non-standard entity for '#' so the "&num=" is interpreted as "#=",
preventing lynx users from following the links that use this unless they
edit the URL.

--
Norman De Forest http://www.chebucto.ns.ca/~af380/Profile.html
af380@chebucto.ns.ca [=||=] (A Speech Friendly Site)
"O'Reilly is to a system administrator as a shoulder length latex glove
is to a veterinarian." -- Peter da Silva in the scary devil monastery