One part of the matching was to check whether domain part of fully qualified name is a name of a domain or any of its subdomains. The problem is that at the time of first request we don't yet have lists of subdomains.
One solution would be to issue a request to data provider, asking for a list of subdomains but mixing subdomain logic with re_expression matching doesn't seem right. Therefore this patch deletes the check for specific domain name, leaving only check for a specific format.
Thanks Jan
On 06/13/2012 05:14 PM, Jan Zelený wrote:
One part of the matching was to check whether domain part of fully qualified name is a name of a domain or any of its subdomains. The problem is that at the time of first request we don't yet have lists of subdomains.
Yeah, that's an interesting problem.
One solution would be to issue a request to data provider, asking for a list of subdomains but mixing subdomain logic with re_expression matching doesn't seem right. Therefore this patch deletes the check for specific domain name, leaving only check for a specific format.
Sadly this completely breaks the multiple re_expression concept. The only reason we can have multiple re_expression's work and not conflict, is because we enforce the following during matching:
* The domain's re_expression must match * The domain's re_expression match must produce the domain's name (or one of its subdomains).
Without the later condition, it's trivial to have re_expressions overlap between domains.
To me it (unfortunately) seems that given both subdomains and multiple re_expressions, resolving a qualified name requires looking up those subdomains.
Cheers,
Stef
On 06/13/2012 05:14 PM, Jan Zelený wrote:
One part of the matching was to check whether domain part of fully qualified name is a name of a domain or any of its subdomains. The problem is that at the time of first request we don't yet have lists of subdomains.
Yeah, that's an interesting problem.
One solution would be to issue a request to data provider, asking for a list of subdomains but mixing subdomain logic with re_expression matching doesn't seem right. Therefore this patch deletes the check for specific domain name, leaving only check for a specific format.
Sadly this completely breaks the multiple re_expression concept. The only reason we can have multiple re_expression's work and not conflict, is because we enforce the following during matching:
- The domain's re_expression must match
- The domain's re_expression match must produce the domain's name (or one of its subdomains).
Without the later condition, it's trivial to have re_expressions overlap between domains.
To me it (unfortunately) seems that given both subdomains and multiple re_expressions, resolving a qualified name requires looking up those subdomains.
Before proposing the patch, I was going through the original review thread looking exactly for this kind of information but it wasn't clear to me if the domain matching is completely necessary. I think I understand it now, thanks for the clarification.
Do you want to implement fix for the issue yourself or should I give it a shot?
Thanks Jan
On 06/14/2012 12:50 PM, Jan Zelený wrote:
Before proposing the patch, I was going through the original review thread looking exactly for this kind of information but it wasn't clear to me if the domain matching is completely necessary. I think I understand it now, thanks for the clarification.
Yeah, it was a discussion on #sssd.
Do you want to implement fix for the issue yourself or should I give it a shot?
TBH, I don't really know how the nuts and bolts of how to implement this fix.
Cheers,
Stef
On 06/14/2012 12:50 PM, Jan Zelený wrote:
Before proposing the patch, I was going through the original review thread looking exactly for this kind of information but it wasn't clear to me if the domain matching is completely necessary. I think I understand it now, thanks for the clarification.
Yeah, it was a discussion on #sssd.
Do you want to implement fix for the issue yourself or should I give it a shot?
TBH, I don't really know how the nuts and bolts of how to implement this fix.
Ok, no problem. I have a rough idea how to do this, I'll give it a shot and let you know once I have something usable prepared.
Jan
sssd-devel@lists.fedorahosted.org