Hmmm. Ok. But that use would not be consistent with the definition in the spec:
“This attribute is provided to allow for overriding execution of rules set at higher levels”
There can be no rule set at a higher level than <file>. And <file> cannot be nested. So what rule could @disable at the <file> level ever disable?
From: Yves Savourel [mailto:]
Sent: Sunday, November 02, 2014 4:08 PM
To: Schnabel, Bryan S;
Subject: RE: [xliff] @disabled in Validation Module. Fix example. Add constraint?
Hi Bryan,
I would agree with the first point (fixing the text for isPresent).
But I think we should keep @disabled allowed for rules at the <file> level: it simply allows to disabled the rule where it is set.
-ys
From: [mailto:]
On Behalf Of Schnabel, Bryan S
Sent: Sunday, November 2, 2014 10:56 AM
To:
Subject: [xliff] @disabled in Validation Module. Fix example. Add constraint?
Hello,
In 5.8.5.9 disabled, in the Validation Module section, we say:
“This attribute is provided to allow for overriding execution of rules set at higher levels . . .”
And
“In the following example, the isNotPresent rule is applied in its entirety to the first unit, but not to the second.”
I have 2 observations.
(1)
The example uses isPresent, not isNotPresent. So I think we should change the sentence to reflect isPresent.
(2)
Since there is no higher level that a rule can be set than <file>, I think we should add a constraint that says @disabled MUST NOT be used on a rule, whose parent validation is a direct child of file. There may be a clearer way of stating that. Probably
somebody who thinks less XPath-y than me could come up with a constraint that is worded more clearly.
Thanks,
Bryan
