Language inconsitency to be fixed for errata

["Robert D Anderson" <robander@us.ibm.com>]

Jarno was updating DITA-OT processing for filtering and noticed an inconsistency in our description of conditional processing groups. This is the DITA 1.3 feature that allows you to create groups within filter/flag attributes, such as
product="database(dbA dbB)"

When describing rules for groups, we say:
> Groups consist of a name immediately followed by a parenthetical group of one or more space-delimited string values.

But immediately following the rules we say:
> An empty group within an attribute is equivalent to omitting that group from the attribute. For example, <ph product="database() A"> is equivalent to <ph product="A">.

The inconsistency is that we first say "A group can't be empty because it has one or more values", but immediately after we basically say "Empty groups are legal, if meaningless, and should be handled this way."

Our second statement already indicates that empty groups are legal. The explicit intent is clear: if an empty group is encountered, it should be allowed but ignored (rather than treated as invalid syntax and throwing an error).

Given that intent, the easiest way to resolve the inconsistency here is simply to change our first statement from "one or more" to "zero or more":
> Groups consist of a name immediately followed by a parenthetical group of zero or more space-delimited string values.

For reference, the latest copy of the spec topic:
http://docs.oasis-open.org/dita/dita/v1.3/errata01/os/complete/part3-all-inclusive/archSpec/base/usage-of-conditional-processing-attributes.html

Regards, Robert D. Anderson DITA-OT lead and Co-editor DITA 1.3 specification, Digital Services Group
E-mail: robander@us.ibm.com Digital Services Group
11501 BURNET RD,, TX, 78758-3400, AUSTIN, USA