### What
Leverage yarn workspaces and Turborepo for integration-tests in order to accelerate development and reduce DevX complexity.
### Why
The current solution for running integration tests requires using `medusa-dev-cli` in order to publish packages to a local npm repository. The package where the command is executed will have its package.json altered for any known medusa dependency in order to install from the local npm. The process is taxing on the host machine resources and prevents rapid iterations when working with integration tests.
For more information, see documentation: f0cc1b324c/docs/content/usage/local-development.md (run-api-integration-tests)
### How
By declaring `integeration-tests/**/*` as a workspace, Turborepo can now be leveraged to build and run integration test as if there were packages. The build process will take care of interdependency between package in order to ensure local dependency are met.
In addition, within each integration-tests "packages", we can declare local dependencies as "*" which will translate to using the one that is part of the current build, regardless of the dependency's version number. No more fiddling with version numbers.
Github actions pertaining to integration-tests have been streamlined to use the new behavior.
The integration-tests packages have been marked as `private:true` in order to avoid publishing them to the public npm registry.
### Testing
```
cd root-of-medusajs-medusa-repo/
yarn install
yarn build
yarn test:integration:api
yarn test:integration:plugins
```
After a code change, `yarn build` must be run before re-running an integration test, which is the same procedure as for unit tests.
Resolves: CORE-845
**What**
I have created a new method on the cart service which is `addLineItems`, allowing a user to add one or multiple items in an optimized way. Also updated the `generate` method from the line item service which now also accept a object data or a collection of data which. Various places have been optimized and cache support has been added to the price selection strategy.
The overall optimization allows to reach another 9000% improvement in the response time as a median (Creating a cart with 6 items):
| | Min (ms) | Median (ms) | Max (ms) | Median Improvement (%)
|---|:-:|---|---|---|
| Before optimisation | 1200 | 9999 | 12698 | N/A
| After optimisation | 63 | 252 | 500 | 39x
| After re optimisation | 56 | 82 | 399 | 121x
| After including addressed feedback | 65 | 202 | 495 | 49x
FIXES CORE-722
* use correct authentication middleware
* remove guard from get-session since it's guarded by middleware doing the same check
* Add integration tests
* Create lazy-swans-agree.md
Co-authored-by: olivermrbl <oliver@mrbltech.com>
Co-authored-by: Oliver Windall Juhl <59018053+olivermrbl@users.noreply.github.com>
**What**
- Adds the use of price selection strategy to the endpoint `GET /admin/variants`
- Updates medusa-js to reflect this change (expanding the parameters).
**Testing**
- Adds a new integration test validating that returned variants are now of type PricedVariant, with the expected fields: original_price, calculated_price, etc.
**Why**
- Our current RMA flows (in our admin dashboard) relied heavily on simply using `order.tax_rate` to calculate variant prices in the different RMA menus. As taxes in Medusa, have become feature complete this approach had become very naive and has several potential issues. Moving the responsibility for calculating the correct prices guarantees that we always show the correct prices to admins.
### What
Creating a swap on an order with a discount leads to an incorrect difference due.
**Scenario**
- Create a store with minimum 2 products (Prod A, Prod B)
- Create a discount that only works for Prod A
- Create an order for Prod A with the discount applied
- Create a swap between Prod A and Prod B
**Expected outcome**
We would expect the difference_due amount to come out to the sum of:
- -1 * (price of prod a - discount applied to prod a)
- price of prod b
**Actual outcome**
Instead the discount is applied across both products when calculating difference due. This results in a total that is instead the sum of:
- -1 * (price of prod a - discount applied to prod a)
- price of prod b - discount on prod b ignoring the condition
### How
Adds `line_item.adjustments` to relations in cart retrieval prior to setting the difference_due to car total
Fixes CORE-361
What
Support sales channel remove product batch in medusa, medusa-js and medusa-react
How
By implementing a new endpoint and the associated service method as well as the repository methods.
Medusa-js new removeProductd method in the resource
Medusa-react new hook in the mutations
Tests
Endpoint test
Service test
Integration test
Hook tests
Fixes CORE-292
* fix: caching deps + add true parallelization to integration tests api
* fix: github action
* chore: upgrade to yarn berry (#1762)
* init migration
* remove: yarn.lock from all pkgs
* fix: build script in pkgs
* adjust yarn config
* fix: yarn.lock and yarnrc.yml
* fix: github actions
* fix: wrong type
* fix (medusa-react): use dts-cli instead of tsdx
* fix: yarn.lock
* fix: yarn v
* fix: prepare script
* add: comment on why we need to downgrade yarn before medusa-dev
* chore: move to Turborepo (#1763)
* increase number of parallel nodes
* fix (medusa-fulfillment-webshipper): build script
* fix: use new version of medusa-dev
* fix: rename cache-bootstrap to cache-deps
* feat(medusa): Add /admin/products/:id/variants end point
* test(medusa): Fix get-variants test casees
* feat(medusa): Include the config to the ProdutService#retrieveVariants as a method parameter
* feat(medusa): Improve get-variants endpoint
* feat(medusa): Improve get-variants endpoint
* test(medusa): Fix unit tests
* test(medusa): Fix unit tests
* feat(medusa): Improve typings
* feat(medusa): Update according to feedback
* feat(medusa): Update according to feedback
* test(medusa): Fix list-variants tests
* feat(medusa): Getting the variants from the product end point should use the productVariantService
* fix(medusa): list-variants expand possibly undefined
* Fix(medusa): List-variants endpoint
* fix(medusa): Tests suite for list-variant
* test(integration-tests): Fix yarn lock
* test(integration-tests): Fix yarn lock
* adds bulk add/remove products to/from collection. Adds endpoint updateProducts on collections that uses these bulk operations
* fix integration tests and test description
* undo change to swap
* made requested changes
* added removeProducts endpoint
* made requested changes
* fix: set collection_id null
* updated collection_id to type string | undefined
* added default relations to retrieve collection endpoint
* push snapshot
* added created at and updated at to collection list endpoint
* updated integration tests
* added q param to list collections
* updated snapshots
* changed validator so that reqiurement shipping_option_id is now optional + added integration test that confirms that when an update contains a requirement without an ID it is created
* fix: formatting
* fix: un-bump babel-preset-medusa-package
* chore: update yarn.lock
* fix: implemented suggested changes, need to validate behaviour on clean branch so NOT ready for merging just yet
* fix: implemented suggested changes, need to validate behaviour on clean branch so NOT ready for merging just yet
* afix: made it impossible to set a min. subtotal requirement that is greater than max. subtotal
* fix: added explanation to error
* fix: Error when removing requirement on update
Co-authored-by: olivermrbl <oliver@mrbltech.com>
* Modifies schema to allow SQLite as a DB driver. SQLite is preinstalled in most OSes allowing for minimal prerequisites in the installation process.
* Removes Redis dependency and replaces "real" redis instance with ioredis-mock this is not feature complete and errors are expected.
* Updates medusa new command to only ask for Postgres credentials if the starter template has database_type === "postgres" in medusa-config.js
* Small improvements to bin resolution
* Improvements to endpoint stability
