There is now the `create_etablissement` method which
create etablissement with EtablissementAdapter
and enqueue api_entreprise jobs to retrieve
all informations we can get based on SIRET
The mailers expect serializable arguments, but were given
ActiveRecord::Relation objects instead. This made the mailers throw an
exception.
But how was that possible ? This code is tested, and the tests were
green.
Well, the specs spy on the mailer implementation, in order to check that
the mailers methods were properly called. Fair enough.
But if the specs mock the mailer code (instead of calling the original
implementation), we may not notice that the original implementation
rejects our method parameters.
Here this is the case: once we actually call the original implementation
the tests start to fail, because some arguments are not converted from
an ActiveRecord::Relation to a serializable array.
This is fixed by ensuring that the mailer code is executed (and doesn't
throw an exception).
When Turbolinks is enabled, a `redirect_to` in a `format: :js` request
will emit code that instructs Turbolinks to navigate to the page.
Turbolinks will then load the redirection target as HTML, and display
it.
But with Turbolinks disabled, the same `redirect_to` will instead
respond with a standard `302: redirect`. `Rails.ajax` will interpret
this redirect as "Please perform the same call again to the redirected
URL, in the same format" – and request our redirection target, but
using `format: :js`.
This breaks the "Publish procedure" button. In that case, we really want
the page to be navigated to. Add an explicit JS redirect, so that the
redirection occurs in HTML.
Dossier has a `default_scope { kept }`.
Because of that, when the parent dossier is discarded, `champ.dossier`
will return nil.
We should kill the default scope. But meanwhile, ensure that
`champ.dossier` returns even a discarded dossier.
When cliking on the "Delete attachment" link, and opening the URL
in a new tab, the `DELETE /attachements/:id` will become
`GET /attachments/:id` – which will cause the `show` action to be
routed with an html format (instead of JS).
In that case, we don't want to throw an error at the user face.
Instead simply re-render the dossier page (if any).
Fix a long-standing error in Sentry.
This helper is:
- no longer used;
- buggy (not all requests increment it);
- discouraged (we should instead match an UI change that signals the end
of an ajax request).
Good riddance.
rails_helper already has `ActiveRecord::Migration.maintain_test_schema!`
which automatically updates the test database if needed.
However, if we raise **before** `maintain_test_schema` had the chance
to do its job, the test database is never-automigrated.
Thus by removing the check, we ensure the test database will be migrated
as needed (and still an error will be raised if the schema cannot be
applied).
Test helpers are separated between two files: spec_helper and
rails_helper. This separation is meant to allow tests that do not
require Rails (like testing standalone libs) to boot faster.
The spec_helper file is always loaded, through `--require spec_helper`
in the `.rspec` config file. When needed, the rails_helper file is
expected to be required manually.
This is fine, but:
- Many test files have a redundant `require 'spec_helper'` line;
- Many test files should require `rails_helper`, but don't.
Not requiring `rails_helper` will cause the Rails-concerned section of
the test environment not to be configured–which may cause subtle bugs
(like the test database not being properly initialized).
Moreover, Spring loads all the Rails files on preloading anyway. So the
gains from using only `spec_helper` are thin.
To streamline this process, this commit:
- Configures `.rspec` to require `rails_helper` by default;
- Remove all manual requires to spec_helper or rails_helper.
Reference: https://stackoverflow.com/questions/24145329/how-is-spec-rails-helper-rb-different-from-spec-spec-helper-rb-do-i-need-it
Turns out that assigning the params to the procedure calls `read` on the
images files. Calling `read` moves the seek offset to the end of the
file; which means that subsequent calls to `read` return an empty
string.
Fix the issue by calling `rewind` before attempting to `read` the file,
to ensure the seek offset is at the beginning of the file.
Fix#4958
