Fix Dockerfile build error.

[dkinzer]

Fixes build error do to && being in wrong spot.

Fixes #2264

[dkinzer]

OK, ... still seeing an error that I didn't used to get before

app_1   |
app_1   | Done generating test app
app_1   | rake aborted!
app_1   | NameError: uninitialized constant SolrWrapper
app_1   | tasks/blacklight.rake:56:in `block (2 levels) in <top (required)>'
app_1   | /usr/local/bundle/gems/rake-13.0.0/exe/rake:27:in `<top (required)>'
app_1   | /usr/local/bundle/bin/bundle:23:in `load'
app_1   | /usr/local/bundle/bin/bundle:23:in `<main>'
app_1   | Tasks: TOP => blacklight:server
app_1   | (See full trace by running task with --trace)
blacklight_app_1 exited with code 1

🤦‍♂

[dkinzer]

OK, was able to get blacklight site up by removing solr_wrapper from Rakefile, and managed to get the blacklight site up on my local:

[cdmo]

I seem to have run into a different error when I tried this. Log:

docker-compose-up.log

<snip>
app_1   |
app_1   | == 20200316185531 AddDeviseGuestsToUsers: migrating ===========================
app_1   | -- change_table(:users)
app_1   |    -> 0.0133s
app_1   | == 20200316185531 AddDeviseGuestsToUsers: migrated (0.0137s) ==================
app_1   |
app_1   | [DEPRECATED] `Bundler.clean_system` has been deprecated in favor of `Bundler.unbundled_system`. If you instead want to run the command in the environment before bundler was originally loaded, use `Bundler.original_system` (called at /usr/local/bundle/gems/engine_cart-2.3.0/lib/engine_cart/tasks/engine_cart.rake:132)
app_1   | Done generating test app
blacklight_app_1 exited with code 0

🤔

[dkinzer]

@cdmo I'm not able to reproduce that issue locally. Maybe it's caching issue.

Have you tried running?:

docker-compose up --build

or better yet

docker-compose up --build --force-recreate

[cdmo]

I ran docker-compose up --build and got the following:

docker-compose-build.log

Looks like webpacker issues...

Same looking problems in the force option of compose up.

docker-up-force.log

I'm not a docker expert, so I'm not able to diagnose. Plus there's the PR #2263 that maybe overlaps with this one? Should this PR get closed and we should just work off that one? Or are they orthogonal somehow and I'm missing the point?

[dkinzer]

Yup, that's fine.