[SYNPY-1686] Suport Pattern column in data model #1285
Conversation
thomasyu888
changed the title
BryanFauble
left a comment
BryanFauble
left a comment
There was a problem hiding this comment.
Thanks for this! The changes look good to me.
| if node.format is not None: | ||
| schema["format"] = node.format.value | ||
|
|
||
| if hasattr(node, "pattern") and node.pattern is not None: |
There was a problem hiding this comment.
Why if hasattr(node, "pattern")?
There was a problem hiding this comment.
in the loop starting on line 5280, assuming the node has the pattern attribute raises an exception when it is not present
@andrewelamb
There was a problem hiding this comment.
Could you elaborate @SageGJ ?
In any case, the node here is an instance of the TraversalNode dataclass, and one of its attributes is pattern, so you shouldn't ever have to check that the attribute exists via hasattr, right?
There was a problem hiding this comment.
if no pattern is specified for the node then the TraversalNode instance of that node will not have a pattern attribute
tests/unit/synapseclient/extensions/schema_files/data_models/example.model.csv
Outdated
Show resolved
Hide resolved
tests/unit/synapseclient/extensions/schema_files/data_models/example.model.csv
Outdated
Show resolved
Hide resolved
Problem:
It is necessary to switch from using validation rules to specify regex validation to specifying the regex pattern in a column of the data model.
Solution:
A new optional column has been added to the data model and the use of the regex validation rule has been deprecated.
Testing:
Tests have been updated as well as testing data.
Testing also uncovered a new bug causing failing tests, tracked here