chore: Migrate github.com/jinzhu/gorm to gorm.io/gorm#863
chore: Migrate github.com/jinzhu/gorm to gorm.io/gorm#863eternal-flame-AD merged 12 commits intomasterfrom
Conversation
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #863 +/- ##
==========================================
+ Coverage 78.76% 79.14% +0.38%
==========================================
Files 56 56
Lines 2670 2225 -445
==========================================
- Hits 2103 1761 -342
+ Misses 476 360 -116
- Partials 91 104 +13 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
This reverts commit b494a3b. Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
jmattheis
left a comment
There was a problem hiding this comment.
Code looks good. I'll try to test this with all supported databases on sunday, or next week.
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
|
I checked the unique index now generates correctly on postgres. Thanks for the catch. The primary key one doesn't look super actionable right now, the current behavior looks fine to me (migrate the foreign key to a supertype but keep primary key as is on an existing DB) |
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
jmattheis
left a comment
There was a problem hiding this comment.
postgres and sqlite looks good, but mysql/mariadb has some problems.
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
Signed-off-by: eternal-flame-AD <yume@yumechi.jp>
jmattheis
left a comment
There was a problem hiding this comment.
, please squash the commits.
Closes #862 .
I removed the manual column fixup: The default column type is already bloba and longblob (I tested on the docker container for MariaDB and postgres).
I also dropped a test that is basically if the database returns error then the function has to return an error, feels a little mechanical coverage driven test not really useful.