Bug #87712 | Fractional DATETIME setting is not recognized as a change in the API | ||
---|---|---|---|
Submitted: | 8 Sep 2017 15:47 | Modified: | 18 Sep 2017 10:14 |
Reporter: | Andy Schmidt | Email Updates: | |
Status: | Verified | Impact on me: | |
Category: | MySQL Workbench | Severity: | S3 (Non-critical) |
Version: | 6.3.9 | OS: | Any |
Assigned to: | CPU Architecture: | Any | |
Tags: | apply, column definition, datetime, fractional |
[8 Sep 2017 15:47]
Andy Schmidt
[11 Sep 2017 5:40]
Chiranjeevi Battula
Hello Andy, Thank you for the bug report. This is most likely duplicate of Bug #80204, please see Bug #80204 Thanks, Chiranjeevi.
[11 Sep 2017 12:13]
Andy Schmidt
Dear Chiranjeevi, I believe this is a completely DISTINCT problem from the one you have been referencing. (Although both deal with the DATETIME type and it certainly would make sense to fix them at the same time.) Importantly, the other problem occurs earlier in the editing process, likely completely elsewhere in the code, and involves entirely different logic. Also, the two behaviors occur independent from one another and there are no identical work-arounds. This all suggests that these are two different bugs (that just happen to occur with the same data type). Specifically: The related problem you are citing deals with INPUT validation, where a missing number of fractional seconds is not accepted when moving the editing focus to the next column, e.g. DATETIME() is not treated as an equivalent to DATETIME or DATETIME(0). The user is never able to press "APPLY" because the input is rejected as invalid syntax. The easily available workaround is use DATETIME or DATETIME(0) - either will be accepted. The newly reported bug is NOT about INPUT validation. There is NEVER a problem entering the intended datatype of DATETIME(6), and the input is fully accepted! Thus none of the issues reported with the other bug play a part. The problem only occurs after APPLY is pressed - when any ALTER command should be constructed that accommodates all column changes. At that time, a DIFFERENT part of program code does not detected that DATETIME(6) is a change from the previous value of DATETIME. Instead of creating a proper ALTER statement, a notice appears claiming that nothing had changes thus no ALTER statement is offered. Finally, the newly reported bug has NO workaround. The user has to manually construct and submit the necessary ALTER statements.
[18 Sep 2017 10:14]
Chiranjeevi Battula
Hello Andy, Thank you for the feedback. Verified this behavior on MySQL Workbench in 6.3.9 version Thanks, Chiranjeevi.
[18 Sep 2017 10:15]
Chiranjeevi Battula
Screenshot
Attachment: Bug_87712.JPG (image/jpeg, text), 169.81 KiB.