gabotorresruiz commented on code in PR #44999:
URL: https://github.com/apache/superset/pull/44999#discussion_r4286574900
##########
superset-frontend/src/utils/semanticViewStructure.ts:
##########
@@ -53,7 +53,7 @@ export const mapSemanticTypeToGenericDataType = (
}
if (
- /(\b(u?int\d*)\b|\bfloat\d*\b|\bdouble\b|\bdecimal\d*\b|\bnumber\b)/.test(
+
/(\b(u?int\d*|integer|bigint|smallint|tinyint|numeric|real)\b|\bfloat\d*\b|\bdouble\b|\bdecimal\d*\b|\bnumber\b)/.test(
Review Comment:
Confirmed closed. I ran the old and the new pattern over 47 type spellings,
the full pyarrow set plus adversarial ones, and the diff changes exactly one
classification: `halffloat` goes from `undefined` to `Numeric`. Nothing else
moves, and the word boundaries still reject `myhalffloat`, `halffloat_t` and
`floatish`. Thanks for turning it around so fast.
##########
superset-frontend/src/filters/components/Range/RangeFilterPlugin.tsx:
##########
@@ -176,6 +177,23 @@ const validateRange = (
max,
});
+ if (values.some(value => value !== null && !Number.isFinite(value))) {
+ return { isValid: false, errorMessage: requiredError };
+ }
+
+ // Allow empty ranges if enableEmptyFilter is false
+ if (!enableEmptyFilter && inputMin === null && inputMax === null) {
Review Comment:
Closed too, thanks. The new `test.each` over the three single-value modes
with `required` both ways pins the SQL half, which is the case that had nothing
holding it.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
