aboutsummaryrefslogtreecommitdiffstats
path: root/Source/BGHUDButtonCell.h
diff options
context:
space:
mode:
authorMax <post@wickenrode.com>2017-06-06 21:20:30 +0200
committerMax <post@wickenrode.com>2017-06-06 21:20:30 +0200
commite52546bccb26e1afc8207053cf260e37f2596c37 (patch)
treee3324032263bd94bc0c78bfbb09c3728dfb27b48 /Source/BGHUDButtonCell.h
parent50e309bc30a10158762b5cf528d67123462dc3a0 (diff)
downloadsequelpro-e52546bccb26e1afc8207053cf260e37f2596c37.tar.gz
sequelpro-e52546bccb26e1afc8207053cf260e37f2596c37.tar.bz2
sequelpro-e52546bccb26e1afc8207053cf260e37f2596c37.zip
Change the way index deletion works (part of #2811)
* Instead of trying to figure out if an index is used by an FK ahead of time Sequel Pro will now simply run the query and check for the error code (and only for error 1553 will it attempt to also remove the FK) * This means that a user will receive two remove dialogs in this rare case, but I think that is actually preferable. Common wisdom shows that users never read the first warning dialog, so in the past they may have agreed to something that they didn’t intend to do. The second dialog should actually make them pause and read it. Also there is a different confirmation button now. * This also fixes the code to detect which FK in particular MySQL is referring to. SP should now correctly handle compound indexes and multi-column FKs as well as ambiguous results.
Diffstat (limited to 'Source/BGHUDButtonCell.h')
0 files changed, 0 insertions, 0 deletions