You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently validation in spikewrap is done in a few (suboptimal) ways, for example on the class and in quick and dirty argument validator method. Some checks are basic on the type / expected value of public argument inputs, others are on things like if folders passed really exist.
Speaking with @lauraporta who also spent a lot of time thinking about best ways to check that user-pased inputs valid, and @niksirbi who I recall is using a nice package, the name of which I forgot.
Anyway, putting this issue here as at some point, depending on how specific validation needs are for each project, we could try and align validation code.
The text was updated successfully, but these errors were encountered:
There are basically two popular Python libraries for validation, pydantic & attrs.
For movement I started using pydantic, but later switched to attrs for various reasons.
For references, see the 3 blog posts mentioned under this issue.
JoeZiminski
changed the title
[Feature] Align validation methods across NUI projects
[Feature] Align validation methods across NIU projects
Oct 10, 2023
Currently validation in
spikewrap
is done in a few (suboptimal) ways, for example on the class and in quick and dirty argument validator method. Some checks are basic on the type / expected value of public argument inputs, others are on things like if folders passed really exist.Speaking with @lauraporta who also spent a lot of time thinking about best ways to check that user-pased inputs valid, and @niksirbi who I recall is using a nice package, the name of which I forgot.
Anyway, putting this issue here as at some point, depending on how specific validation needs are for each project, we could try and align validation code.
The text was updated successfully, but these errors were encountered: