Bug #42266 | Incorrect SQL produced using syncronization feature for datatype DOUBLE | ||
---|---|---|---|
Submitted: | 22 Jan 2009 11:16 | Modified: | 1 Apr 2009 21:10 |
Reporter: | Stuart Ramsay | Email Updates: | |
Status: | Duplicate | Impact on me: | |
Category: | MySQL Workbench | Severity: | S3 (Non-critical) |
Version: | 5.0.29SE Rev 3359 | OS: | Windows (XP SP3) |
Assigned to: | CPU Architecture: | Any | |
Tags: | CHECKED, double, sql error, syncronization |
[22 Jan 2009 11:16]
Stuart Ramsay
[22 Jan 2009 11:17]
Stuart Ramsay
Test Workbench Model
Attachment: testdouble.mwb (application/x-zip-compressed, text), 3.86 KiB.
[22 Jan 2009 16:04]
MySQL Verification Team
Thank you for the bug report.
[13 Feb 2009 11:45]
Frank Quosdorf
confirmed. This is especially annoying in conjunction with bug #37709. I decided to create a new Workbench project after it was impossible to get a reasonable alter script generated (caused by #37709). After creating the new Workbench project from the latest SQL schema file, all fields with DOUBLE(15,2) in the old project became DOUBLE in the new project. The issue became obvious when I added a field of type DOUBLE using the same (wrong) declaration as seen in other fields in the same table (could not remember that there was this manual (15,2) thing before). Applying the alteration failed with the result described by the bug reporter. With bug #37709 and this one (#42266) around, I have no choice but to manually add (15,2) to all existing fields of type DOUBLE to at least be able to alter the schema, but it will then again produce an alter script even though nothing changed in the model (#37709). This is really a dreadful combination of bugs in a production environment. OS: Windows Vista Workbench version: 5.0.29 OSS
[1 Apr 2009 21:10]
Alfredo Kojima
marking as duplicate of bug #41290