If there are no reasons speaking against this, it would be nice to follow the RFC.
EDIT: In this case it should be updated in feed_*.tpl too.
Code: Select all
Index: rss.php
===================================================================
--- rss.php (revision 860)
+++ rss.php (working copy)
@@ -184,6 +184,7 @@
break;
case 'atom1.0':
+ header('Content-Type: application/atom+xml; charset=utf8');
$namespace_hook = 'frontend_display:atom-1.0:namespace';
break;
}
I see, for the majority it'll be better to stick with text/xml. As far as I understand the RFC it's not a must.garvinhicking wrote:One problem of sending that mimetype still is that it will then no longer be rendered within the Browser, but offered as a file download.
I don't think this is a cool thing, because currently Atom Feeds can be rendered within the browser with XSL stylesheets.
Perhaps there is a way to keep nospam@example.com out of the Atom1.0 feed. I don't know if it is possible with Smarty (would require a string-compare).Person constructs MAY contain an atom:email element, but MUST NOT contain more than one.
Code: Select all
{if $email != "nospam@example.com"}<email>$email</email>{/if}
Do you happen to have a RFC/Spec where it says that? Before I change that, I would like to read about it, just be sure that I don't break anything. I believe you are right, but double-checking is never badAnyway, I think "application/x.atom+xml" shoud be updated in feed_atom1.0.tpl.
Of course, here you are:garvinhicking wrote:Do you happen to have a RFC/Spec where it says that? Before I change that, I would like to read about it, just be sure that I don't break anything.
And as a second source, see the IANA-List of MIME-Types.[url=http://www.ietf.org/rfc/rfc4287.txt]RFC 4287[/url] (Section 2, Atom Documents) wrote: Both kinds of Atom Documents are specified in terms of the XML
Information Set, serialized as XML 1.0 [W3C.REC-xml-20040204] and
identified with the "application/atom+xml" media type.
Vertrauen ist gut, Kontrolle ist bessergarvinhicking wrote:I believe you are right, but double-checking is never bad
No problem, I have to thank yougarvinhicking wrote:Thanks for that pointer. I just committed the change!
It has been over a year since my best friend Lilly... Sorry wrong tape... ;-)garvinhicking wrote:Code: Select all
+ header('Content-Type: application/atom+xml; charset=utf8');
(English: The HTTP-Header-Encoding (UTF8) differs from the XML-declaration (UTF-8))Validome.org wrote:Die Zeichenkodierung im HTTP-Header (UTF8) weicht von der Zeichenkodierung der XML-Deklaration UTF-8 ab.
Code: Select all
+ header('Content-Type: application/atom+xml; charset=utf-8');
Code: Select all
header('Content-Script-Type: text/javascript');Actually s9y should always use <script type="text/javascript"> for each <script> container.While I'm at thise little validator-bugs. I don't know if s9y is generally affected by this, but my installation (custom theme and plugins) was:
http://www.validome.org/forum/viewtopicp-552-1.htmCode: Select all
header('Content-Script-Type: text/javascript');
I know and I think it's better to keep it as an option for a while.garvinhicking wrote:Hm, in the current rss.php that header() call is not enabled.
My Firefox deals with it.Once I enable it, my firefox won't show the Atom feed inside the browser anymore...?! Is application/atom+xml really the content-type that should be enforced if current browsers don'T really deal with it?
I'll enforce it for Atom 1.0, but as said before I think it should not be default.Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.1) Gecko/20060601 Firefox/2.0.0.1 (Ubuntu-edgy)
Great. This was just a proposal in case it isn't done that way.Actually s9y should always use <script type="text/javascript"> for each <script> container.
ah, okay. I'm still using 1.5, so that might be a difference.My Firefox deals with it.