On Monday, September 3, 2001, at 06:38 PM, Craig R. McClanahan wrote:
> On Mon, 3 Sep 2001, robert burrell donkin wrote:
<snip>
>> that sounds right but i'm still a little confused...
>>
>> the concrete example in the test case has rules with identical patterns
>> '*/bar', say.
>> one of which has associated namespace foo the other has no (null)
>> associated namespace.
>>
>> there are 3 logical matching cases for 'root/bar' with varying
>> namespaces.
>> 1.[null namespace] match(null,'root/bar')
>> 2.[matching namespace] match('foo','root/bar')
>> 3.[not matching namespace] match('notfoo','root/bar')
>>
>> what result should i expect for each?
>>
>
> IMHO, 1 and 2 should match, but 3 should not.
just to confirm both should match for 1 + 2 but only the one with the null
namespace should match 3
that's the way that i think it should work too :)
> That being said, I wasn't
> thinking much about suffix matches when adding this, so it may or may not
> actually work this way yet.
i don't think that's the way it works at the moment
i'll confirm and then prepare a patch (if needed)
- robert
