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
When analyzing the top project only (using the --no-project-recurse flag)
of an alire project, we should also filter the alire directory that contains all dependencies (needed projects).
The text was updated successfully, but these errors were encountered:
First of all, what is meant with recursive analysis?
When a project is referenced, what should be analyzed from this project?
A. For projects whose source code is available, real recursive analysis is possible - but is it wanted?
B. For projects where only the interface (*.ads file) are available, analysis can't continue any further - but is it needed?
The user can specify a non-empty list of projects to analyse. However, nothing is specified about the presence/absence of relations between these projects. I.e. is a project and a dependent project thereof allowed together? And if so, is there a difference with only that project [when the --no-project-recurse-flag is absent]?
This causes that there are (at least) two options to filter relations between projects.
A. Using the arguments of dependency_graph_extractor to select a project with the relevant dependent projects and the --no-project-recurse flag.
B. While analyzing the graphml file.
Do we want so much variation that we as a small team should support?
@ajmooij What is your opinion? How is this realized in the C++ variant?
When analyzing the top project only (using the --no-project-recurse flag)
of an alire project, we should also filter the alire directory that contains all dependencies (needed projects).
The text was updated successfully, but these errors were encountered: