Made it possible for the TO_* functions to take more varied arguments #48
Conversation
Pull Request Test Coverage Report for Build 234
💛 - Coveralls |
|
Would you please add new tests for this? |
|
Sure, I have now added test cases for some of the other parser types that NewValue allows. |
| 'SELECT to_tsvector(LOWER(c0_.text)) AS sclr_0 FROM ContainsText c0_', | ||
| 'SELECT to_tsvector(1 + 1) AS sclr_0 FROM ContainsText c0_', | ||
| 'SELECT to_tsvector(1) AS sclr_0 FROM ContainsText c0_', | ||
| 'SELECT to_tsvector(LENGTH(c0_.text)) AS sclr_0 FROM ContainsText c0_', |
There was a problem hiding this comment.
The checks from lines 24, 25 and 26 won't produce meaningful SQL, as to_tsvector(integer) is not a valid function. You can safely drop those scenarios.
| \sprintf('SELECT TO_TSQUERY(UPPER(e.text)) FROM %s e', ContainsText::class), | ||
| \sprintf('SELECT TO_TSQUERY(1+1) FROM %s e', ContainsText::class), | ||
| \sprintf('SELECT TO_TSQUERY(true) FROM %s e', ContainsText::class), | ||
| \sprintf('SELECT TO_TSQUERY(LENGTH(e.text)) FROM %s e', ContainsText::class), |
There was a problem hiding this comment.
The checks from lines 35, 36 and 37 won't produce meaningful SQL, as to_tsquery(integer) is not a valid function. You can safely drop those scenarios.
|
I suppose that considering that to_tsquery and to_tsvector only takes strings, this doesn't make much sense for them, so i've undone those changes. I added the same tests for to_json and to_jsonb though, since the accept anyelement. |
|
Thank you @magnusnordlander for this contribution! It's lovely to see new people enhancing the project. |
We want to be able to pass any kind of StringPrimary into the
TO_JSONBfunction, however, I think it makes sense to be able to pass e.g. a SimpleArithmeticExpression as well, and it seems NewValue is the best corresponding type for that.Also, even though we're only using it in
TO_JSONB, obviously this makes sense for the otherTO_*functions too.