The abstract string types should be used only for function arguments.
Concrete types such as nsString and nsAutoString should be used for new
instances of strings.
Cheers,
Josh
On 04/04/2013 01:10 PM, [email protected] wrote:
Hello,
I'm a bit confused when it comes to abstract string classes, i.e. nsAString. The Mozilla
internal string guide speaks of them as "historic", so I wondered that they are
still expected as input value in nsDOMTokenList::Contains
In nsScriptLoader::ProcessScriptElement I want to do the following:
nsCOMPtr<nsIContentSecurityPolicy> csp;
nsAString& str1 = NS_LITERAL_STRING("string1");
nsAString& str2 = csp->getMyStringValue();
str1.Append(str2);
// call Contains(str1) now, to check if the string "string1string2"
// is a value of an element attribut
Explanation:
- getMyStringValue() is supposed to be a JavaScript function written
in contentSecurityPolicy.js and delivers a value from CSPRep, very
smililar to what getAllowsEval does with _allowEval. The returned value
is always a mixed string consisting of upper and lower case letters and
numbers, e.g. "Abc123"
- after the Append operation str1 should contain "string1string2"
Question 1: My above handling of this issue seems to be erroneous. How do I
use the strings correctly inside this context?
Question 2: nsDOMTokenList defines Contains(const nsAString& aToken,
ErrorResult& aError). However, I found this function nowhere used
in this way. Can I simply call it via Contains(str1) as it is
done in lines 111 or 148?
Thanks for your support.
Jeremy
_______________________________________________
dev-security mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-security
