Port remaining tests in functions.rs to sqllogictest #6608
Conversation
github-actions
bot
added
core
jiangzhx
changed the title
| @@ -87,5 +87,5 @@ pub fn decimal_to_str(value: Decimal) -> String { | |||
| } | |||
|
|
|||
| pub fn big_decimal_to_str(value: BigDecimal) -> String { | |||
| value.round(12).normalized().to_string() | |||
| value.round(16).normalized().to_string() | |||
There was a problem hiding this comment.
Sqrt(2) returns a double value of 1.4142135623730951.
the precision of value.round(12) is not sufficient, 1.414213562373 is returned.
To improve accuracy, we could increase the rounding precision to value.round(16)
There was a problem hiding this comment.
will cause other testcase failed:
- pg_compat_simple.slt:76
- window.slt:441
- decimal.slt:391
- predicates.slt:47
- pg_compat_window.slt:692
so i reverted change.
| SELECT sqrt(column1) FROM t | ||
| ---- | ||
| 1.4142135623730951 |
There was a problem hiding this comment.
The original test case had a value of 1, but when I called sqrt(1) in SQLLogicTest, it returned a value of 1, which should have been displayed as 1.0 for clarity. To make the response more reasonable, I changed the value in the test case from 1 to 2, it's will retrun 1.4142135623730951
| # case_sensitive_identifiers_functions | ||
| statement ok | ||
| create table t as values ( |
There was a problem hiding this comment.
This data could replace the original test case. Can you guys provide some feedback?
There was a problem hiding this comment.
I think using 2 is a better value than 1 👍
| # case_sensitive_identifiers_functions | ||
| statement ok | ||
| create table t as values ( |
There was a problem hiding this comment.
I think using 2 is a better value than 1 👍
|
I clicked by mistake, sorry ... |
Which issue does this PR close?
Closes #6300.
Rationale for this change
What changes are included in this PR?
Are these changes tested?
Are there any user-facing changes?