When writing tests, for performance reasons we often want to reuse
the same data-source instance for many tests suites. At the same time,
we want to keep such test suites independent and allow them to reuse
the same model name for different model classes.
Juggler does support redefinition of a model with the same name.
This change is adding a new API called that allows tests to remove all
old models before creating new ones. This API would be typically
called from a `before` hook.
Signed-off-by: Miroslav Bajtoš <mbajtoss@gmail.com>
Defer automigrate/autoupdate until we are connected, so that connection
errors can be reported back to callers.
Fix postInit handler to not report connection error to console.log
and via dataSource "error" event in case there is already an operation
queued. When this happens, we want the error to be handled by the
queued operation and reported to its caller.
Signed-off-by: Miroslav Bajtoš <mbajtoss@gmail.com>
add beginTransaction method which calls begin
method from the Transaction class which in turn
calls the connector's beginTransaction method if
it supports transactions.
Co-Authored-By: Miroslav Bajtoš <mbajtoss@gmail.com>
Before this change, when an "access" hook modified the "query" object,
the "near" condition from the original "query" object were still
applied. As a result, the query can end up being more restrictive and
return only a subset of models that should have matched the conditions.
With this change in place, after "access" hook observers are invoked,
we update the "near" condition using the "query" object provided by
hook observers.
Signed-off-by: Miroslav Bajtoš <mbajtoss@gmail.com>
Before this change, two "loaded" events were emitted for every model
instance found. This commit fixes the problem and restores the intended
behavior.
The bug was probably introduced by 5e0c73bec7.
Fix DataSource constructor to create a shallow copy of the settings
object provided by the caller. This prevents surprising behavior
where changes made to `ds.settings` were picked up by the provided
config object, as observed e.g. in tests of our MongoDB connector.
- On include = true or include = 1 the lib crashes with
"TypeError: includes.forEach is not a function".
- checking for boolean and number type and return empty array.
- fix: include symbol and function in normalization
- fix: review changes; adding debug stmt + early exit for tests
Before this change, when a property was configured with a default value
at LoopBack side and the database was returned a record with a missing
value for such property, then we would supply use the configured
default.
This behavior is problematic for reasons explained in #1692.
In this commit, we are fixing DAO operations like `find` and
`findOrCreate` so that they do NOT apply default property values on
data returned by the database (connector).
Please note that most of the other CRUD methods were already not
applying default values on database data as long as the connector
provided native implementation of the operation.
When converting plain-data object values into model instances,
correctly handle the case where the constructor functions is a class
constructor and must be invoked via `new`.
This is a breaking change. Juggler methods are no longer returning
instances of Bluebird Promise, but return a native Promise instead.
As a result, API consumers can no longer use sugar APIs provided by
Bluebird, e.g. `spread`, `delay`, etc.
Migration guide: rewrite your code from `.then()` style into `async`
functions using `await` statements.
The new version of our config enabled function-paren-newline rule,
this commit fixes the codebase to use more consistent handling
of newlines when calling functions.
Fix the regression introduced by 9af79cf51a where updateAttributes
and replaceAttributes was incorrectly handling the response returned
by the database.
This commit restores the old behaviour where `context.data` is updated
only by a connector-provided function and reverts an incorrect change
of a test file made as part of the faulty fix.
Besides the fixes, this patch also renames `data` to `dbResponse` and
add comments explaining the structure of the response object for more
clarity.
Before this change, many Operation Hooks were not correctly propagating
changes made to `ctx.data` via reassigning ctx.data to a new object.
This change modifies existing tests to account for this different
scenario, adds few more tests for scenarios that were not covered by
tests before and finally fixes the problem discovered.
darthmaim
Miroslav Bajtoš
Biniam Admikew
spurreiter
Bill Matson
Agnes Lin
Hage Yaapa
Raymond Feng
leon3s
Matthew Gabeler-Lee
Dimitris
Tom Kirkpatrick
virkt25