todo.txt

Salut Dario,

Pardon encore pour le lag.
Je ne me rappelais plus, mais il semble que cela soit calculé dans le backend (dans networkCreator.js:77)

Voila.
Ogier



/**

* Apply transformation to node values (normalize, max, split scholar names, ...)

*/

const 
postProcessNodes = (affAcronyms,
nodes) 
=> {

    const 
nullAffinitiesObj = affAcronyms.reduce((o,
v) 
=> ({ ...o, 
[v]: 0 }), {})



    const 
averageLab = nodes.reduce((lab,
node) 
=> {

        affAcronyms.forEach(affinity
=>

            lab[affinity] +=
node.metrics.values[affinity] /
nodes.length

        )

        return 
lab

    }, affAcronyms.reduce((o,
v) 
=> ({ ...o, 
[v]: 0 }), {}))



    // standardize metrics

    nodes.forEach(node
=> {

        node.metrics.std = {}

        affAcronyms.forEach(affinity
=> 
node.metrics.std[affinity] =
node.metrics.values[affinity] ?
node.metrics.values[affinity] /
averageLab[affinity] :
0)

    })



    // Find scholars picks

    nodes.forEach(lab
=> {

        lab.metrics.max =
lab.network.nodes.reduce((o,
node) 
=> {

            affAcronyms.forEach(k
=>

                o[k] =
o[k] >
node.metrics.values[k] ?
o[k] :
node.metrics.values[k]

            )

            return 
o

        }, { ...nullAffinitiesObj })

    })

}