Ergonomic, hermetic, user-friendly software build system for Python, Go, Java, Kotlin, Protobuf, Scala, Shell. Pants lets you fearlessly scale up your codebase!

Pants

yeah I think we’ll get to more configurability eventually

there may be tunables, but i don't know what they would be

i don't see why we would try to solve this very general problem of how to display progress in the right rules for pants output when we don't have any experience looking at what the less important output might be

I think “more time spent in a particular subgraph” is a good way to highlight a `top` like view/aggregate

and like i'm thinking any method of saying "hey, this rule's progress might be more interesting to the user" or vice versa would be useful and then naturally leads to a fully automatically scheduled solution eventually

the bit I’m worried about with `top` style is that after some given threshhold the larger things will begin to dominate teh display of smaller things - at the natural threshhold at which most of the smaller things run at

i guess i wasn't thinking hard enough about the immense and changing variety of output there is without any automated whatever, which is a lot

so if something is iteratively computing a bunch of small things those would all get sort of lost and it might look like pants is hung w/o telling you much of whats happening

is a list of everything fancy? not sure i understand the concern

it also punts on showing the relevant stuff

the point of mentioning `top` is that top is focused on a specific thing

and we may want to focus on just the raw thing

what do we want to display about small tasks vs large long running tasks

(maybe with "amount of i/o" as another parameter?)

with a simpler execution model I think the mapping of worker -&gt; thing its executing makes perfect sense

the new async stuff complicates it for sure

but definitely some kind of graph/stack aware outer-most-edge-first + aggregation

i feel like we can compress info about smaller tasks