Proposal A discussion points

[glyn]

Issues raised

Applying filters to objects

Raised cburgmer/json-path-comparison#41.

See:

• https://cburgmer.github.io/json-path-comparison/results/filter_expression_after_dot_notation_with_wildcard_after_recursive_descent.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_after_recursive_descent.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_on_object.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_bracket_notation_with_number_on_object.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_equals_on_object.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_value_after_recursive_descent.html

Multiplication, division, and scripting

Raised cburgmer/json-path-comparison#42.

See these examples where an operator is not treated as part of key:

• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_division.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_multiplication.html

Proposal A lists multiplication and division as "to dos". Is this the thin end of the wedge for scripting?

Dot notation without root

Raised cburgmer/json-path-comparison#45.

• https://cburgmer.github.io/json-path-comparison/results/dot_notation_without_root.html

Leading zeroes in numeric literals

Raised cburgmer/json-path-comparison#46.

• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_equals_number_with_leading_zeros.html

Duplicates in filter results

Raised cburgmer/json-path-comparison#47.

See:

• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_equals.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_greater_than.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_greater_than_or_equal.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_less_than.html
• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_less_than_or_equal.html
• (counter example?) https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_negation_and_equals.html

### Typo fix

Raised cburgmer/json-path-comparison#48.

Issues not raised

&& usually binds more tightly than || (and seems to be the case in JavaScript), but:

• https://cburgmer.github.io/json-path-comparison/results/filter_expression_with_different_ungrouped_operators.html

If implementors can't agree pretty unanimously that && binds more tightly than ||, then we can't expect users to expect that. So Proposal A's behaviour of failing to compile the path is a good option.

[glyn]

@cburgmer: I'd prefer to discuss the above with you before raising issues, but perhaps you'd prefer issues so that the discussion is "on the record". What do you think?

[glyn]

See https://github.com/cburgmer/json-path-comparison/tree/master/proposals/Proposal_A for some of the rationale of Proposal A.

[cburgmer]

Yes, let's discuss.
I'd prefer to not branch out into actual implementations, but rather centralise this - quite probably over at json-path-comparison. The goal is to allow other authors to join the discussions (even if we will not capture many of the major implementations it seems).

I don't know how to structure this though. We could start with Github issues. If we find this is hard to structure, we could try moving into the wiki later.

[glyn]

All the issues have been raised.