Skip to content

Various SRL findings #1191

Description

@simonstey
  1. https://github.com/w3c/data-shapes/blob/gh-pages/sparql12-rl/index.html#L1097-L1098 -> V_all is defined as V_N where N is the length of the rule-element sequence. But V_i is defined at index.html:1092-1094 as the union of V_0 and all vars_j for j less than i. Therefore V_N = V_0 ∪ vars_1 ∪ ... ∪ vars_(N-1)
    => it excludes vars_N, the variables contributed by the last element of the body. A rule whose head variable is bound only by its final body element is therefore not a well-formed rule per the text, or am I missing something?

Suggested fix:

Let Vall be the union of V0 and all varsj for j from 1 to N, where N is the length of the sequence.


  1. index.html:1891 -> inside evalFunction:
return F(evalFunction(expr1, row), ..., evalFunction(exprN, row))

=> shouldnt row be replaced with μ?


  1. sparql-rl-grammar.bnf:16 -> (ForClause ::= 'FOR' Var 'IN' iri), referenced from sparql-rl-grammar.bnf:12 (Rule1) and :13 (Rule2)

-> CHANGES.md:22 states: "FOR ?var IN <shape>" removed.

=>But it was not removed from the grammar. (it's also still used in syntax-ruleset-structure-10.srl)


  1. sparql-rl-grammar.bnf:13** (Rule2 ::= 'IF' iri? ForClause? 'DATA'? BodyPattern 'THEN' HeadTemplate)

=> this is also not part of the spec anymore

Metadata

Metadata

Assignees

Labels

Type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions