What’s the difference between cache_key’s and fingerprints? The cache_key contains the fingerprint? +plus something else if I remember correctly. Trying to understand the differences/unions between the 2 values below.

versioned_target.cache_key.hash.split(-1)
versioned_target.target.playload.fingerprint()

if you're discussing the result of a

self.invalidated()

call, the

vt

usually refers to a

VersionedTarget

-- see https://github.com/pantsbuild/pants/blob/b0d59b440a2e616eb7246325dc021bc186f758a5/src/python/pants/invalidation/cache_manager.py#L170

Yeah, you are probably correct. The code uses

vt

which I replaced with

virtual_target

, but

versioned_target

is likely the correct expansion.

stu's response in the channel answers the main question though, which is that

vt.target.payload.fingerprint()

is just a cache key for the current target's payload, which may contain sources, etc. the cache key for the versioned target may be calculated based on dependencies or other things, again depending on the task. the cache key is what is specifically used for invalidation for

self.invalidated()

-- i think it can be considered a superset (meaning strictly more granular) than the target's

.fingerprint()

(which is calculated from the payload as you have above, usually -- this can be modified as well), but it again depends on the task.