Skip to content

Conversation

alpha0010
Copy link
Contributor

Fixes #51

@alpha0010
Copy link
Contributor Author

Hmm... it looks like we found a bug in Sphinx. I will investigate.

@alpha0010
Copy link
Contributor Author

Reported: http://sphinxsearch.com/bugs/view.php?id=2519

@oohnoitz
Copy link
Contributor

once its fixed in sphinx, i'll get this merged in.

@alpha0010
Copy link
Contributor Author

It appears the problem is our HAVING test (which is unrelated to this pull). Technically that specific query can be rewritten as:

$result = SphinxQL::create(self::$conn)->select(SphinxQL::expr('count(*) as cnt'))
    ->from('rt')
    ->where('gid', 304)
    ->groupBy('gid')
    ->execute();

Maybe the Sphinx team decided HAVING should only be used in cases that WHERE is not applicable (for example, COUNT(*) > 123). I suppose we will have to wait for their developers to say if this is intended behaviour.

@oohnoitz
Copy link
Contributor

it does seem like it's a regression since the test hadn't failed on the beta release on previous build versions.

although rewriting the query to the one you've provided would cause the test to pass, it would defeat the purpose of having the test in the first place, since we are checking to make sure that the HAVING query built is working and valid. but as you've stated, we'll just need to wait hear from the team themselves.

@oohnoitz oohnoitz merged commit 13a46d3 into FoolCode:master Aug 23, 2016
@oohnoitz
Copy link
Contributor

HAVING tests will be updated in the #107 PR once tests pass.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants