'concore run' is misleading

[GREENRAT-K405]

The concore run command does not run a workflow. It compiles a .graphml definition into platform-specific scripts via mkconcore.py and drops them in out/. After it "completes", users are shown a panel telling them to manually cd out && ./build && ./run. This is the exact opposite of what a user expects from a command called "run"

What I suggest is replace "concore run" => "concore build" everywhere

[GREENRAT-K405]

@pradeeban @Rahuljagwani @mayureshkothare please have a look

[pradeeban]

Interesting. I believe this is a new addition of a feature. Thanks for catching.

[GREENRAT-K405]

@pradeeban I think our aim with concore CLI commands should be complete replacement (not immediately but smoothly after we cover the basic scripts) and even more improvements to the already existing OS dependent rigid scripts

[GREENRAT-K405]

for this, the CLIs should not reference or be too dependent on the traditional scripts too much and they should be capable to drive the executions on their own.

[pradeeban]

Fair point. But we do not want to completely make breaking changes either for historic reasons, for a research project as mature as CONTROL-CORE.

[GREENRAT-K405]

Fair point. But we do not want to completely make breaking changes either for historic reasons, for a research project as mature as CONTROL-CORE.

Absolutely!
@pradeeban Also I wanted to ask (I know we shouldn't touch this but,) should we move the studies like nintan, ratc, ratc2 into a single folder under the name "Legacy_Studies" or something like this? So developers don't confuse it with what needs to be worked on.

[GREENRAT-K405]

@pradeeban Or shouldn't we move them to concore-studies repo entirely? The legacy studies I believe contain extremely important practical records.

But if we think from the perspective of end-user who would just import concore as a package they wouldn't find a need to explore the existing studies and also from the perspective of developers/ contributors they most likely won't and shouldn't touch these studies.

Just wanted to have a thought regarding this, no need for immediate changes.

[pradeeban]

@GREENRAT-K405 No, we better not touch those. A more conservative approach would be to simply create a new repository such as "CONTROL-CORE3" or something like that if our changes are drastic (I don't think that is needed).

[GREENRAT-K405]

@pradeeban understood!