attr("viewBox") broken by case folding

[[email protected]…]

Hi !

May I let you take a look at this fiddle ?

​http://jsfiddle.net/HTZ9n/6/

I think there is an issue while retrieving the viewBox attribute with inline svg. Do I make any sense ?

Thanks !

[anonymous]

A few more things:

Tested on Firefox 9.0.1, Chrome 17.0.963.33 beta, Safari 5.1.1

worked with jQuery 1.6.4

[timmywil]

Confirmed. ​http://jsfiddle.net/timmywil/HTZ9n/7/show/

The viewBox attribute should stay camelCase. SVG may be the only exception where attribute names are not lowercase. ​http://www.w3.org/TR/SVG/coords.html#ViewBoxAttribute

[anonymous]

Is this perhaps down to the issue described here? ​http://api.jquery.com/prop/ The element seems retrievable with .prop, but not with .attr.

[dmethvin]

It seems like the DOM attribute methods are case insensitive on sane browsers, so perhaps we can just skip the .toLowerCase() part? In a quick check to .attr() I just moved the lowercase to getting the attrHook only and got a full pass including IE7 and IE8. I wasn't able to get a full test in IE6 due to some BrowserStack issues.

In environments where there is case sensitivity (including IE7 HTML) I don't think there is any way to hide that, so documenting that the case must match seems fine to me.

--- a/src/attributes.js
+++ b/src/attributes.js
@@ -305,8 +305,7 @@ jQuery.extend({
                // All attributes are lowercase
                // Grab necessary hook if one is defined
                if ( notxml ) {
-                       name = name.toLowerCase();
-                       hooks = jQuery.attrHooks[ name ] || ( rboolean.test( name ) ? boolHook : nodeHook );
+                       hooks = jQuery.attrHooks[ name.toLowerCase() ] || ( rboolean.test( name ) ? boolHook : nodeHook );
                }

                if ( value !== undefined ) {

[timmywil]

Attributes are not case sensitive in IE when using getAttributeNode, which we use for all html attributes in IE. viewBox is technically an svg attribute that should be rendered with xhtml or xml, which is why it is case-sensitive. I think the best solution would be to have a test for svg, if possible.

[mikesherov]

Unfortunately, we don't support SVG, and even though this worked in the past, it's not something we're supporting so we're not going to fix it for now. If we decide to support SVG in the future, we'll reopen and address this issue.

Thanks!

​http://docs.jquery.com/Won't_Fix#SVG.2FXML.2FVML_Bugs

[anonymous]

Requesting that this be re-opened. jQuery 2.x is advertised as supporting modern browsers, all of which support SVG, so fixing it in at least 2.x would be valuable. Also, there is currently no official documentation stating it won't be fixed, as the page for that on docs.jquery.com is gone.

[markelog]

Replying to anonymous:

Requesting that this be re-opened. jQuery 2.x is advertised as supporting modern browsers, all of which support SVG, so fixing it in at least 2.x would be valuable. Also, there is currently no official documentation stating it won't be fixed, as the page for that on docs.jquery.com is gone.

Sorry – ​http://contribute.jquery.org/wont-fix/

[anonymous]

Ah, so that's where the documentation went. OK, so could this be set as "patchwelcome" and made a blocker of #7584?

[markelog]

Replying to anonymous:

Ah, so that's where the documentation went. OK, so could this be set as "patchwelcome" and made a blocker of #7584?

At that point of time, it's the other way around - doc will be changed, all svg/vml "patchwelcome" tickets will become "wontfix", but some svg related issues could be addressed, if they would affect wide amount of users.

[gibson042]

#15200 is a duplicate of this ticket.

[Genosse]

Well, What is the reason to convert characters in attributes to lowercase? But for normal basic support SVG need only CamelCase notation. And you do not want to do this thing for many years? Why?