#657: Clarified that members of list must be literals/IRIs and the same order is kept #710
Conversation
HolgerKnublauch
requested review from
afs,
recalcitrantsupplant,
robert-david and
simonstey
Co-authored-by: Ted Thibodeau Jr <[email protected]>
Co-authored-by: Ted Thibodeau Jr <[email protected]>
Co-authored-by: Ted Thibodeau Jr <[email protected]>
Co-authored-by: Ted Thibodeau Jr <[email protected]>
Co-authored-by: Ted Thibodeau Jr <[email protected]>
Co-authored-by: Ted Thibodeau Jr <[email protected]>
| Note that some <a>named parameter functions</a> — such as <code>shnex:IntersectionExpression</code> — | ||
| also use a <a>SHACL list</a> as an object of the <a>key parameter</a>, similar to <a>list parameter functions</a> which always have a <a>SHACL list</a> as the object of their <a>list parameter property</a>. | ||
| However, these may produce more than one <a>output nodes</a>, and also accept lists as input nodes. | ||
| However, these may produce more than one <a>output node</a>, and also accept lists as input nodes. |
There was a problem hiding this comment.
| However, these may produce more than one <a>output node</a>, and also accept lists as input nodes. | |
| However, these may produce more than one <a>output node</a>, and also accept a list of input nodes. |
There was a problem hiding this comment.
Does it accept a list of lists? I guess not.
There was a problem hiding this comment.
Hi @robert-david I think you are looking at a different part of the spec than this PR is about.
The PR is about ListExpressions that are written as plain RDF lists in TTL, such as ( 1 2 3 )
The line you have highlighted is about the general architecture of Node Expressions, which can either be NamedParameterFunctions or ListParameterFunctions.
The example mentioned in this paragraph is something like
sh:values [
shnex:intersection (
[ shnex:pathValues ex:children ]
[ shnex:pathValues ex:friends ]
)
]
which is one example where a named parameter function can take other lists as its input - here the list containing the two pathValues expressions.
So unless I am missing something, this is neither incorrect nor related to the PR.
Closes #657