Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Adds a feature of dimension anotations #202

Merged
merged 9 commits into from
Oct 1, 2024
Merged
Show file tree
Hide file tree
Changes from 5 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions .changeset/thick-papayas-turn.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
---
"cube-link": patch
---

Adding dimension annotations (`meta:annotation`) re https://gitlab.ldbar.ch/bafu/visualize/-/issues/542
73 changes: 73 additions & 0 deletions meta/core.md
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,34 @@ To be able to distinguish Dimensions that are defined inside a Cube from Dimensi

</aside>

### meta:Limit {#Limit}

A type of [annotation](#annotation) which can be used to express a limit or target value for a dimension.

<aside class='example' title='Dimension with a target'>

```turtle
<cube/shape> sh:property [
a cube:KeyDimension ;
sh:path ex:year ;
] , [
a cube:MeasureDimension ;
meta:annotation [
a meta:Limit ;
schema:value 95 ;

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Here, this part, schema:value is too generic. Do we mean a lower or an upper limit?

We can avoid this confusion in at least three ways I could think of:

  1. Introduce two new classes meta:UpperLimit and meta:LowerLimit
  2. Add some other specification, like an additional meta:limitType "upper" or something like that
  3. Be clear, on what we mean, by replacing schema:value by something more precise, like a schema:maxValue for an upper limit or a schema:minValue for a lower limit.

I would advocate together with Thomas for option three, as it reuses schema:maxValue/schema:minValue instead of introducing new things.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, I think I understood now.

Let me clarify one more point. Would it ever make sense to have both schema:minValue and schema:maxValue on a limit. From what you write, I guess that is pointless?

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Depends on how you want to look at it.

It could be, that something is supposed to follow an upper and a lower limit. One could then add a schema:minValue as well as a schema:maxValue to the same meta:Limit. Or be more precise and add two meta:Limit.

From a semantic's point it might be clearer to add two meta:Limit in such a case. So yes, there's not really a point to have both predicates for one meta:Limit. Or would you like to disagree @Rdataflow ?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I just imagined the case of a target range, being specified by means of lower and upper limit. Yes, for such cases it could make sense to have both at the some node ;-)

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Two thoughts:

  1. It could make sense to have min and max, together, so to allow multiple "limits" with corresponding upper and lower limit. (e.g. Target Band 2025, Target Band 2026) so I propose to keep connected min / max together.
  2. It should be possible, to give a limit annotation a title, and probably also a sub-title/description. There might be multiple limit annotations, and earlier or later even a possibility to choose one, or multiple to show in an interface (e.g. visualize).

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Alright, you got me conviced. I was steering more towards thinking of target ranges as "two limits in a trenchcoat" and thus wanting to split them up. But that would just be more "wordy". So let's keep the possibility.

As for your second thought @l00mi , I agree. a dcterms:title would make sense.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In principle, any property is allowed. The shapes aren't closed. Do you think it's worth mentioning that explicitly in the spec?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

schema:name "Target 2020" ;
meta:annotationContext [
sh:path ex:year ;
sh:hasValue <https://schema.ld.admin.ch/year/2020> ;
] ;
] ;
] .
```

</aside>

`schema:minValue` and `schema:maxValue` can be used to express a range of limit values.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@tpluscode it would be helpful for the reader to explicitly state lower limit values and upper limit values 😉 as there are also situations where only one sort occurs and this information is needed to check and display whether the values are within the acceptable range.

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good catch. We seem to have dropped that along the way - or at least I think we had that covered once.
In any case, we should include that somehow.
Either with different classes? I.e. meta:UpperLimit and meta:LowerLimit? Or maybe more elegant with schema:minValue/schema:maxValue instead of the generic schema:value?

Or would you see a case where a LIMIT could no specified to be an upper or lowr Limit @Rdataflow ? It would not really be a limit without that specification?

Copy link

@Kronmar-Bafu Kronmar-Bafu Oct 24, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

After a small chat with @Rdataflow, we both agree that the generic schema:value is not well suited for meta:Limit. At least for what we try to express here.

One thing we should distinguish is between the mathematical concept of a limit (or limes) and limit as we envisioned here. Here, we think about something like a speed limit. Or some value, which should not be exceeded (or exceedence leads to some consequence), or below which a measurement can't be quantified like a detection limit.

We would argue, that these kinds of limits do generally have an upper or lower meaning associated with them. So the use of schema:value can be and should be replaced with schema:minValue and/or schema:maxValue, to make the meaning of this limit clearer.

Edit: wrote schema:minInclusive which of course was rubbish, my bad. Replaced it with schema:minValue.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@Kronmar-Bafu I do not follow, which part is in your opinion missing from the spec? The snippet shows schema:limit but it also says that minValue/maxValue can be used.


## Properties

### meta:dataKind (temporal / spatial) {#dataKind}
Expand Down Expand Up @@ -84,3 +112,48 @@ With `meta:nextInHierarchy` the next lower level is attached to the higher level

This property is used on a Dimension Constraint to express a relation with other properties through a [meta:Relation](#Relation) instance, the nature of this relationship is determined by the properties used on the instance.
See [this example](../#relexample).

### meta:annotation {#annotation}

This property is used to add additional information to a dimension.

### meta:annotationContext {#annotationContext}

Links to dimension values to which the annotation applies.
The objects of `meta:annotationContext` MUST be well-formed [Property Shapes](https://www.w3.org/TR/shacl/#property-shapes).
The value of their `sh:path` MUST be an IRI of a cube's key dimension.
Only a subset of SHACL Constraints are supported which defined which observations that the annotation
applies to, namely:

1. `sh:hasValue` - to select specific observation value
2. `sh:in` - to select multiple observation values
2. `sh:minInclusive` - to select values greater or equal to a specific value
3. `sh:maxInclusive` - to select values smaller or equal to a specific value
4. `sh:minExclusive` - to select values greater than a specific value
5. `sh:maxExclusive` - to select values smaller than a specific value
tpluscode marked this conversation as resolved.
Show resolved Hide resolved

In case of temporal dimensions, the constraint values are expected to be literals with datatypes `xsd:date`,
`xsd:dateTime` or `xsd:gYear`, or the IRIs of [temporal entities](https://lindas.admin.ch/governance/core-entities/).
Comment on lines +135 to +136
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I could not find a simple enough way to include this in the validation shape


<aside class='example' title='Dimension with a continuous limit on a temporal dimension'>

```turtle
<cube/shape> sh:property [
a cube:KeyDimension ;
sh:path ex:year ;
] , [
a cube:MeasureDimension ;
meta:annotation [
a meta:Limit ;
schema:value 95 ;
schema:name "Target 2020" ;
meta:annotationContext [
sh:path ex:year ;
sh:minInclusive "2020-01-01"^^xsd:date ;
sh:maxInclusive "2020-12-31"^^xsd:date ;
] ;
] ;
] .
```

</aside>
Original file line number Diff line number Diff line change
@@ -0,0 +1,81 @@
PREFIX ex: <http://example.org/>
PREFIX dcterms: <http://purl.org/dc/terms/>
@prefix relation: <https://cube.link/relation/> .
@prefix meta: <https://cube.link/meta/> .
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix cube: <https://cube.link/> .
@prefix observation: <https://environment.ld.admin.ch/foen/nfi/observation/max_min_undefined> .
@prefix sh: <http://www.w3.org/ns/shacl#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
@prefix schema: <http://schema.org/> .
@base <https://example.org/> .

<cube> a cube:Cube ;
cube:observationConstraint <shape> ;
cube:observationSet <observationSet> .

<observationSet> cube:observation <observationA> .

<observationA> a cube:Observation ;
cube:observedBy <observer> ;
<measure> 4.9 ;
<year> <https://ld.admin.ch/time/year/2020> ;
.

<shape> a cube:Constraint ;
sh:targetClass cube:Observation ;
sh:closed true ;
sh:property
[
sh:path rdf:type ;
sh:nodeKind sh:IRI ;
sh:minCount 1 ;
sh:maxCount 1
] ;
sh:property
[
sh:path cube:observedBy ; ;
sh:nodeKind sh:IRI ;
sh:minCount 1 ;
sh:maxCount 1
] ;
sh:property
[
a cube:MeasureDimension ;
sh:datatype xsd:decimal ;
sh:path <measure> ;
schema:name "measure" ;
sh:minCount 1 ;
sh:maxCount 1 ;
meta:annotation
[
a meta:Limit ;
schema:value 97 ;
schema:name "Target 2020+" ;
meta:annotationContext
[
sh:path <year> ;
sh:minInclusive <https://ld.admin.ch/time/year/2020> ;
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why two lower bounds?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That's a test cases which should fail

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ah, I see, to check they are not accepted :-)

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I tried to make this even more accurate but I seem to have hit a limitation of shat SHACL can do when there are multiple properties with same value. I failed to make a validation to catch for example

 sh:minInclusive <https://ld.admin.ch/time/year/2020>
 sh:minExclusive <https://ld.admin.ch/time/year/2020>

sh:minExclusive <https://ld.admin.ch/time/year/2022> ;
] ;
] ;
] ;
sh:property
[
a meta:KeyDimension ;
tpluscode marked this conversation as resolved.
Show resolved Hide resolved
sh:path <year> ;
schema:name "year" ;
sh:minCount 1 ;
sh:maxCount 1 ;
sh:nodeKind sh:IRI ;
] ;
sh:property
[
a meta:KeyDimension ;
tpluscode marked this conversation as resolved.
Show resolved Hide resolved
sh:path <station> ;
schema:name "station" ;
sh:minCount 1 ;
sh:maxCount 1 ;
sh:nodeKind sh:IRI ;
] ;
.
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
@prefix sh: <http://www.w3.org/ns/shacl#> .
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
@prefix schema: <http://schema.org/> .
@prefix cube: <https://cube.link/> .

_:report a sh:ValidationReport ;
sh:result [
rdf:type sh:ValidationResult ;
sh:resultSeverity sh:Violation ;
sh:sourceConstraintComponent sh:XoneConstraintComponent ;
sh:sourceShape <https://cube.link/shape/standalone-constraint-constraint#AnnotationContext> ;
sh:focusNode _:b5 ;
sh:value _:b5 ;
sh:resultMessage "annotation context only allows constraints sh:hasValue, sh:in, sh:minInclusive, sh:maxInclusive, sh:minExclusive, sh:maxExclusive and they cannot be mixed" ;
] ;
sh:conforms false .
Original file line number Diff line number Diff line change
@@ -0,0 +1,87 @@
PREFIX ex: <http://example.org/>
PREFIX dcterms: <http://purl.org/dc/terms/>
@prefix relation: <https://cube.link/relation/> .
@prefix meta: <https://cube.link/meta/> .
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix cube: <https://cube.link/> .
@prefix observation: <https://environment.ld.admin.ch/foen/nfi/observation/max_min_undefined> .
@prefix sh: <http://www.w3.org/ns/shacl#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
@prefix schema: <http://schema.org/> .
@base <https://example.org/> .

<cube> a cube:Cube ;
cube:observationConstraint <shape> ;
cube:observationSet <observationSet> .

<observationSet> cube:observation <observationA> .

<observationA> a cube:Observation ;
cube:observedBy <observer> ;
<measure> 4.9 ;
<year> <https://ld.admin.ch/time/year/2020> ;
.

<shape> a cube:Constraint ;
sh:targetClass cube:Observation ;
sh:closed true ;
sh:property
[
sh:path rdf:type ;
sh:nodeKind sh:IRI ;
sh:minCount 1 ;
sh:maxCount 1
] ;
sh:property
[
sh:path cube:observedBy ; ;
sh:nodeKind sh:IRI ;
sh:minCount 1 ;
sh:maxCount 1
] ;
sh:property
[
a cube:MeasureDimension ;
sh:datatype xsd:decimal ;
sh:path <measure> ;
schema:name "measure" ;
sh:minCount 1 ;
sh:maxCount 1 ;
meta:annotation
[
a meta:Limit ;
schema:value 95 ;
meta:annotationContext
[
sh:path <non-dimension> ;
sh:hasValue 10 ;
] ;
] ,
[
a meta:Limit ;
schema:value 95 ;
meta:annotationContext
[
sh:path <station> ;
sh:hasValue ex:Station ;
] ;
] ;
] ;
sh:property
[
a meta:KeyDimension ;
tpluscode marked this conversation as resolved.
Show resolved Hide resolved
sh:path <year> ;
schema:name "year" ;
sh:minCount 1 ;
sh:maxCount 1 ;
sh:nodeKind sh:IRI ;
] ;
sh:property
[
sh:path <station> ;
schema:name "station" ;
sh:minCount 1 ;
sh:maxCount 1 ;
sh:nodeKind sh:IRI ;
] ;
.
Original file line number Diff line number Diff line change
@@ -0,0 +1,48 @@
@prefix sh: <http://www.w3.org/ns/shacl#> .
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
@prefix schema: <http://schema.org/> .
@prefix cube: <https://cube.link/> .

_:report a sh:ValidationReport ;
sh:result [
rdf:type sh:ValidationResult ;
sh:resultSeverity sh:Violation ;
sh:sourceConstraintComponent sh:NotConstraintComponent ;
sh:sourceShape <https://cube.link/shape/standalone-constraint-constraint#AnnotationContextDimensionType> ;
sh:focusNode _:b5 ;
sh:value _:b5 ;
sh:resultMessage "annotation context must have exactly one sh:path which is cube's key dimension" ;
], [
rdf:type sh:ValidationResult ;
sh:resultSeverity sh:Violation ;
sh:sourceConstraintComponent sh:XoneConstraintComponent ;
sh:sourceShape [
sh:path _:b706 ;
sh:xone (
[
sh:class <https://cube.link/meta/KeyDimension> ;
]
[
sh:property [
sh:path [
sh:inversePath <https://cube.link/meta/annotationContext> ;
] ;
sh:minCount 1 ;
] ;
]
) ;
] ;
sh:focusNode _:b7 ;
sh:value _:b9 ;
sh:resultPath _:b706 ;
] ;
sh:conforms false .

_:b706 rdf:first sh:path ;
rdf:rest (
[
sh:inversePath sh:path ;
]
) .
Loading
Loading