Skip to content

Releases: mickhansen/graphql-sequelize

v9.5.1

16 Nov 09:28
Compare
Choose a tag to compare

Add missing files for previous release.

v9.5.0

16 Nov 09:19
Compare
Choose a tag to compare

sequelize@6, graphql@16, graphql-relay@0.10 support via #715 by @intellix

v9.0.0

17 Jul 11:01
Compare
Choose a tag to compare
  • added DateType [#621]
  • typeMapper / attributeFields now converts Sequelize.DATE into DateType (used to be String) [#621]

v8.0.0

01 Mar 20:55
Compare
Choose a tag to compare

relay.createConnection: the after callback is now always called, regardless of edges being requested or not

v7.0.1

25 Feb 16:25
Compare
Choose a tag to compare

Fixes regressions introduced in v7.0.0

v7.0.0

21 Feb 14:32
Compare
Choose a tag to compare
  • Added support for having target in resolver/relay.sequelizeConnection being a thunk/callback
  • Removed the default generated orderBy enum on connections

v6.0.0

21 Feb 08:22
Compare
Choose a tag to compare
  • dataloader-sequelize is no longer directly used by graphql-sequelize
  • graphql-sequelize will now always generate N+1 queries unless

dataloader-sequelize has been removed to allow users to make their own decisions regarding batching and to avoid automatically polluting the sequelize object.

For improved performance and less database load (in terms of query count) we suggest implementing createContext from https://github.com/mickhansen/dataloader-sequelize

v5.0.0

30 Jan 07:56
Compare
Choose a tag to compare

Internal JSON type renamed from JSON to SequelizeJSON and now exported as a reusable type:

#403

v4.0.0

17 Oct 13:46
Compare
Choose a tag to compare

resolver no longer adds context to findOptions by default as this can lead to memory leaks when Sequelize does deep cloning.

v3.0.0

17 Oct 13:46
Compare
Choose a tag to compare

This major release changes the general philosophy of the project to rely on batching rather than includes.
Almost all use cases will experience great performance improvements.
There is a minor performance degradation in cases where the query is only one level deep with a single belongsTo relation.