You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In GEXF each attributes can have a title.
Some GEXF producer like tab2net uses this field.
But the parser does not import this information as graphology data model does not handle this title/label concept.
Similarly the write does not propose a way to populate those title fields unless to provide a customize node/edge format method.
I discovered this as some table2net users complained not finding those title after opening a table2net gexf in Gephi Lite : gephi/gephi-lite#163
To solve this one approach could be to expose the node and edge model as a graph attribute both at parsing and writing.
At parsing the models would be added as graph attribute and let consumers decide what to do with those extra informations.
At writing provider could decide to provide models as graph attributes if they want. If not the current inferring models would work as usual. Actually we probably want to merge the provided model with the inferred one.
In GEXF each attributes can have a title.
Some GEXF producer like tab2net uses this field.
But the parser does not import this information as graphology data model does not handle this
title
/label
concept.Similarly the write does not propose a way to populate those title fields unless to provide a customize node/edge format method.
I discovered this as some table2net users complained not finding those title after opening a table2net gexf in Gephi Lite : gephi/gephi-lite#163
To solve this one approach could be to expose the node and edge model as a graph attribute both at parsing and writing.
At parsing the models would be added as graph attribute and let consumers decide what to do with those extra informations.
At writing provider could decide to provide models as graph attributes if they want. If not the current inferring models would work as usual. Actually we probably want to merge the provided model with the inferred one.
What do y ou think @Yomguithereal?
The text was updated successfully, but these errors were encountered: