And also print a small hint that --verbose should be used to try to see all errors.
Tests are just tweaked a little to catch this in the output, but a sample of the new rendering...
Hopefully, this can guide people towards --verbose more often when they need it, and the "best deep match" might surface better things for certain schemas. The whole space of guessing which error to show is kind of dicey though -- I tried various priorities for different validators and found that it always depends on how the schema was written.
And also print a small hint that
--verboseshould be used to try to see all errors.Tests are just tweaked a little to catch this in the output, but a sample of the new rendering...
Hopefully, this can guide people towards
--verbosemore often when they need it, and the "best deep match" might surface better things for certain schemas. The whole space of guessing which error to show is kind of dicey though -- I tried various priorities for different validators and found that it always depends on how the schema was written.