Conversation
>= instead of > fix comment
labkey-tchad
left a comment
labkey-tchad
left a comment
There was a problem hiding this comment.
I think we should focus on whitespace and repeated characters for now.
I'm not ready to exclude backslashes from our fuzz testing. It is one of the mostly likely characters to cause problems and should be a prime target for fuzz testing.
If there is a bug in Selenium, we need to open an issue.
| { | ||
| if (++attempts >= MAX_RANDOM_TRIES) | ||
| throw new IllegalStateException("Failed to generate " + size + " unique text choices after " + MAX_RANDOM_TRIES + " attempts"); | ||
| String generated = randomString(randomInt(1, 25), ";").trim(); |
There was a problem hiding this comment.
I missed this in the last PR. Why are we excluding ; from text choices?
If it is a valid character, we should include it.
There was a problem hiding this comment.
https://github.com/LabKey/internal-issues/issues/1096#issuecomment-4331544773
It breaks MVTC conversion tests. I can create a separate method without semicolon for conversion tests. What do you think?
There was a problem hiding this comment.
I'd say we should have a separate method for those tests. Or maybe have the conversion tests adjust their expectations depending on whether the choices have semicolons or not.
Co-authored-by: Trey Chadick <tchad@labkey.com>
2 participants
Rationale
Fuzz generator made our tests to fail intermittent because of several things: double spaces, repeat characters, Selenium special characters.
Related Pull Requests
Changes
Backslash excluded because Selenium can mess if read it from UI