Improved validation of tools during workflow execution.#1302
Improved validation of tools during workflow execution.#1302guerler merged 3 commits intogalaxyproject:devfrom
Conversation
|
6 failures instead of 44. Making progress. |
a5af35c to
b4c4eff
Compare
|
Ugh - none of these fail for me locally. Odd, maybe because I rebased against dev before testing - forced push and we will try again. |
|
Probably a problem with |
|
Progress: Will fix the logging typos and log the actual reason the validation is failing. |
41590e3 to
918ce64
Compare
|
With extra logging tests discovered a fairly substantial and unfortunate bug, fixed with 918ce64. |
There was a problem hiding this comment.
Should we log the exception and raise it afterwards, like in the run.py or the other way around?
Do a more complete validation of everything after workflow parameters have been replaced and inputs connected up. Attempt 2 at the idea in galaxyproject#1284.
bff1785 to
ce787ad
Compare
|
👍 |
Improved validation of tools during workflow execution.
Prior to PR galaxyproject#1302 workflows wouldn't validate call validate at all on inputs. 1302 was perhaps an over correction that causes validate to be called on datasets that aren't yet READY but require metadata. This rolls that back somewhat, other parameters will still validate - but validation asserting the existence of metadata won't get called in workflows. This is still not ideal but after 1302 and this - we are still in a better, more secure place than beforehand. One might argue that since we need to introduce a new validation point (right before a job is scheduled and set to QUEUED) - that we would be better off rolling back galaxyproject#1302 entirely. I've considered that, but I still think it is better to fail fast on the other problems, also I would rather call the validation code too many times than not enough and it would be very late in the release cycle for such a change.
Prior to PR galaxyproject#1302 workflows wouldn't validate call validate at all on inputs. 1302 was perhaps an over correction that causes validate to be called on datasets that aren't yet READY but require metadata. This rolls that back somewhat, other parameters will still validate - but validation asserting the existence of metadata won't get called in workflows. This is still not ideal but after 1302 and this - we are still in a better, more secure place than beforehand. One might argue that since we need to introduce a new validation point (right before a job is scheduled and set to QUEUED) - that we would be better off rolling back galaxyproject#1302 entirely. I've considered that, but I still think it is better to fail fast on the other problems, also I would rather call the validation code too many times than not enough and it would be very late in the release cycle for such a change.
Do a more complete validation of everything after workflow parameters have been replaced and inputs connected up.
Attempt 2 at the idea in #1284.
Lets see if the tests go any better now.