Attribute selectors should be case-sensitive by default and only be case-insensitive for certain attributes values

• Component: CSS
• Status: RESOLVED DUPLICATE

[Niels Leenheer (HTML5test)]

Current behavoir:
The values of all attributes are compared by using a case-insensitive method.

Correct behavoir:
The CSS specification says that the case-sensitivity is determined by the document language. The HTML specification specifically tells us which attributes are case-sensitive, case-neutral or case-insensitive. Because attributes values are case-sensitive by default, we should treat unknown attributes in a case-sensitive way. 

The list of case-insensitive attributes is as follows:
lang, dir, http-equiv, text, link, vlink, alink, compact, align, frame, rules,
valign, scope, axis, nowrap, hreflang, rel, rev, charset, codetype, declare,
valuetype, shape, nohref, media, bgcolor, clear, color, face, noshade,
noresize, scrolling, target, method, enctype, accept-charset, accept, checked,
multiple, selected, disabled, readonly, language, defer, type

See also the following bugs:
http://bugzilla.mozilla.org/show_bug.cgi?id=357614
http://bugzilla.mozilla.org/show_bug.cgi?id=356936
http://bugs.kde.org/show_bug.cgi?id=135505

and

http://rakaz.nl/item/css_selector_bugs_case_sensitivity

[Eric Seidel (no email)]

The code which defines this behaivor is in:

CSSParser::parseContent

            if (fname == "attr(") {
                if (args->size() != 1)
                    return false;
                Value* a = args->current();
                String attrName = domString(a->string);
                if (document()->isHTMLDocument())
                    attrName = attrName.lower();
                parsedValue = new CSSPrimitiveValue(attrName, CSSPrimitiveValue::CSS_ATTR);


Note that that code will be changed slightly by the patch attached to bug 15302, since the mathml.css file (which contains attr selectors) is parsed before there is any document() set.

[Robert Blaut]

Created attachment 17316 [details]
minimal test case

[Robert Blaut]

Is this bg already fixed? I don't see error behavior.

[Alexey Proskuryakov]

*** This bug has been marked as a duplicate of 15470 ***