In commit 905ac48bb we activated exceptions when assets were not found,
in order to detect places where we were trying to load non-existent
images.
We got an exception for that reason: we were loading images based on the
current locale, but for some locales there was no images.
We're now using fallbacks and loading another image when the original
one isn't available.
Note we're copying the English images to images with a generic name for
the case where there's no fallback with an image. We're copying the
files instead of using symbolic links to make sure they can be
overwritten independently in other CONSUL installations.
Also note we're updating the HTML so the section gets the ID instead of
the header. That way the system test is simple.
When configuring phases in a process, we were validating the start date
or the end date is present, the other date is present too.
However, in other parts of the application we were checking whether a
phase is enabled and then assumed its dates were present if the phase
was enabled. However, we weren't validating this behavior, so it was
possible to enable a phase and leaving its dates blank, causing the
application to crash.
So, as suggested by Alberto, we're changing the validation rule so
phase dates are mandatory when a phase is enabled.
With this rule, the old validation rules are not necessary. I've
considered leaving them in order to avoid database inconsistencies.
However, I realized records having a disabled phase with its start and
end dates have always been valid. This means applications checking for
the presence of these dates instead of checking whether the phase is
enabled have never worked properly.
We don't have to change the logic anywhere else because as mentioned we
were already checking phases are enabled before using their dates.
Our `namespace` helper returns a string. However, Rails version 5.2.4.6
doesn't allow strings as arguments to polymorphic_path [1]
Since returning a symbol in our `namespace` helper would break other
places in the application, we're converting it to a symbol in the
methods calling `polymorphic_path`.
[1] https://github.com/advisories/GHSA-hjg4-8q5f-x6fm
On small screens, sometimes the bottom of the footer didn't have the
footer's background color.
I'm not sure why the `min-height` rule affects this outcome. However,
since this rule usually results in footer with quite a bit of empty
space at the bottom, we can simpliy remove the rule and use padding to
guarantee there's a bit of space between the text in the footer and the
bottom of the screen.
We were getting an error since we started using the postgres 9.6 image:
```
Attaching to app_1, database_1
database_1 | Error: Database is uninitialized and superuser password is not specified.
database_1 | You must specify POSTGRES_PASSWORD to a non-empty value for the
database_1 | superuser. For example, "-e POSTGRES_PASSWORD=password" on "docker run".
database_1 |
database_1 | You may also use "POSTGRES_HOST_AUTH_METHOD=trust" to allow all
database_1 | connections without a password. This is *not* recommended.
database_1 |
database_1 | See PostgreSQL documentation about "trust":
database_1 | https://www.postgresql.org/docs/current/auth-trust.html
```
Version 9.4 hasn't been maintained since February 2020, so we aren't
supporting it either. And we might start using `UPSERT` instead of
`find_or_create`, which was introduced in PostgreSQL 9.5.
We're still supporting PostgreSQL 9.5 even if it's also unmaintained
because it has only been officially unmaintained for a couple of months.
In order to ensure compatibility with existing CONSUL installations, we
disabled all settings related to SDG. However, we also made it much
harder to enable SDG globally on the site, since administrators first
had to enable the SDG feature and then enable it for each process.
Most people will expect SDG is enabled for all processes once they
enable the SDG feature, so that's what we're doing. They can of course
disable specific processes should they wish to do so.
Before commit 28caabecd, it was clear which budgets were in draft mode
because their phase was "drafting".
Now the phase isn't "drafting" anymore, so we have to make it clear
somehow that the budget is a draft.
I'm using styles similar to the ones we added in commit 2f636eaf7 for
completed budgets but at the same time making them slightly different so
it's easy to differenciate completed and drafting budgets.
Particularly the line with `within "tr", text: "Finished budget" do` is
now easier to read.
This way we avoid a potential pitfall. Imagine that the factory which
creates a finished budget generated a budget with the name "COMPLETED
Budget 1". Then the test:
```
within "#budget_#{finished_budget.id}" do
expect(page).to have_content("COMPLETED")
end
```
Would pass even if we didn't add the text "COMPLETED" anywhere else,
because it would be included in the name of the budget.
Since the target branch was in a different repository, the action failed
since it couldn't find the reference.
The code here is based on a recent change in Pronto [1] and with a comparison
between the repo.url property of pull_request.head and pull_request.base
to determine if the pull request was created from a forked repository
[1] https://github.com/prontolabs/pronto/commit/4fe28418b6
