bamaer commented on PR #8927: URL: https://github.com/apache/hop/pull/8927#issuecomment-6221024471
@mattcasters, thanks, that was quick. I built 588ac42 and ran the tests, plus some extra cases. `[d]+1`, `[d]-1`, `[d]+0.5` and date differences across both DST changes in Brussels, New York, Sydney and Lord Howe now all give whole days, and `DAYS` matches PDI again. Timestamp fields compare and add correctly. One thing left: **Result type with more than one copy** (`OpenFormula.java:165-168`): `discoverReturnType` now runs on every row and sets `needDataConversion` on the `OpenFormulaMetaFunction`, which all copies of the transform share. I ran two copies of `IF([a]>0;[a];0)` into an Integer field, one seeing positive values and one negative, and 58 of 400,000 rows came out as BigDecimal instead of Long. Keeping the flag per copy in `OpenFormulaData`, next to `returnType`, would fix 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]
