Support singular output_validator/ for 2023-07-draft spec #424
Comments
mpsijm
changed the title
mpsijm
added a commit
that referenced
this issue
Feb 15, 2025
* Implement 'type' field in problem.yaml from 2023-07-draft spec * [problem] Move determining fields interactive, multipass, and custom_output to constructor of ProblemSettings * Rename variables `multipass` to `multi_pass` * [skel] Read type instead of validation * [problem] Mark TODOs for #424 (support singular output_validator/) * [skel] Remove support for `bt new_problem --type default` in favour of `pass-fail` * [problem] If we don't find output validator paths, abort when problem type is interactive and/or multi-pass
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Follow-up of #421 (comment).
Currently (in the
legacyandlegacy-icpcspecs), there can be multipleoutput_validators/. But according to the new spec (https://icpc.io/problem-package-format/spec/2023-07-draft.html#problem-package-structure-overview), there can be only one output validator, that lives inoutput_validator/. This makes sense, because if there are multiple output validators, this forces the problem packager to decide in which order these should be executed (usingbuildandrunscripts). We'd like to support this in BAPCtools as well, however:bt export.{input,output}_visualizer/andstatic_validator/at some point, so it makes sense to set up our architecture to support these as well. But as long as we don't, it may not be worth it.See also the comments in
problem.pyandvalidate.pymarked with# TODO #424, and DOMjudge/domjudge#2887.The text was updated successfully, but these errors were encountered: