Describe the problem/challenge you have
Something blows up in some Starlark code in my template; but it's hard to get to the root cause because I can't see the values of expressions at runtime.
Describe the solution you'd like
The ability to include trace output during a ytt run. Even better if I could leave such tracing and be able to toggle it on and off from the command-line (e.g. include in --debug or introduce a flag --trace)
Anything else you would like to add:
The current workaround is to use assert.fail() in spots close to where the error occurs and include the desired context that way.
[ ] enumerate use-cases; describe a number of scenarios where runtime information/data/state would be useful.
[ ] explore whether the generalized solution for addressing those use-cases would be best as an enhancement to --debug or a separate mechanism (e.g. another flag)
[ ] explore how complex/custom Starlark values (structs, yamlfragments, ...) would be presented.
[ ] explore what security concerns might be present.
Vote on this request
This is an invitation to the community to vote on issues, to help us prioritize our backlog. Use the "smiley face" up to the right of this comment to vote.
š "I would like to see this addressed as soon as possible"
š "There are other more important things to focus on right now"
We are also happy to receive and review Pull Requests if you want to help working on this issue.
jayunit100
commented
2 years ago
Describe the problem/challenge you have Something blows up in some Starlark code in my template; but it's hard to get to the root cause because I can't see the values of expressions at runtime.
Describe the solution you'd like The ability to include trace output during a
yttrun. Even better if I could leave such tracing and be able to toggle it on and off from the command-line (e.g. include in--debugor introduce a flag--trace)Anything else you would like to add: The current workaround is to use
assert.fail()in spots close to where the error occurs and include the desired context that way.This idea has come up multiple times (in fact, hard to believe there isn't already an issue about this). Most recently in this Slack conversation: https://kubernetes.slack.com/archives/CH8KCCKA5/p1632938666479600
To do:
--debugor a separate mechanism (e.g. another flag)Vote on this request
This is an invitation to the community to vote on issues, to help us prioritize our backlog. Use the "smiley face" up to the right of this comment to vote.
š "I would like to see this addressed as soon as possible" š "There are other more important things to focus on right now"
We are also happy to receive and review Pull Requests if you want to help working on this issue.