Removes REST and ElasticSearch

Why are these changes being introduced:

* We have moved to OpenSearch and GraphQL

Relevant ticket(s):

* https://mitlibraries.atlassian.net/browse/TIMX-220
* https://mitlibraries.atlassian.net/browse/TIMX-213

How does this address that need:

* Removes v2 feature flag and all code in the "else" blocks of v2 checks
* Removes ElasticSearch code
* Removes no longer used dependencies
* Removes most of the REST API. The auth portion will remain until the
  user model is removed in https://mitlibraries.atlassian.net/browse/TIMX-221
* All unused cassettes were removed (this was tedious)
* Updates Actions workflow to no longer run the tests twice
* Updates documentation (public and dev facing)

Document any side effects to this change:

* The auth portion of the REST API remains as adding removal of the user
  model and associated code was too much scope creep. An additional
  ticket has been added to remove the User model and auth portion of
  the REST API.
* The dockerfile is still very elasticsearch focused. There is a
  separate ticket to address this so I chose to leave it unchanged in
  this commit.
* There is a separte ticket to move to our shared Actions CI workflow

Author: JPrevost

.env.test

@@ -1,6 +1,4 @@
 EMAIL_FROM=[email protected]
 EMAIL_URL_HOST=localhost:3000
 JWT_SECRET_KEY=3862fc949629030de4259b88f6e8f7c3702b2fabfc68d00d46fb7f9f70110690b526997ef4d77765ffa010d8aba440286af39947d0c85287174d99be2db14987
-ELASTICSEARCH_INDEX=timdex-prod
 OPENSEARCH_INDEX=timdex-prod
-V2=true

.github/workflows/ci.yml

@@ -22,12 +22,9 @@ jobs:
         run: |
           bundle config path vendor/bundle
           bundle install --jobs 4 --retry 3
-      - name: run tests graphql elasticsearch
+      - name: run tests
         run: |
-          bundle exec rails test --exclude /graphqlv2/
-      - name: run tests graphql opensearch
-        run: |
-          bundle exec rails test --exclude /graphqlv1/
+          bundle exec rails test
       - name: Coveralls
         uses: coverallsapp/[email protected]
         with:

Gemfile

@@ -3,14 +3,11 @@ git_source(:github) { |repo| "https://github.com/#{repo}.git" }
 
 ruby '3.1.2'
 
-gem 'aws-sdk-elasticsearchservice'
 gem 'bootsnap', require: false
 gem 'devise'
-gem 'elasticsearch', '~>6.8'
 gem 'faraday_middleware-aws-sigv4'
 gem 'flipflop'
 gem 'graphql'
-gem 'jbuilder'
 gem 'jwt'
 gem 'lograge'
 gem 'mitlibraries-theme', git: 'https://github.com/mitlibraries/mitlibraries-theme', tag: 'v1.0.2'
@@ -40,7 +37,6 @@ end
 
 group :development do
   gem 'annotate'
-  gem 'graphiql-rails'
   gem 'jekyll'
   gem 'jekyll-remote-theme'
   gem 'jekyll-seo-tag'

Gemfile.lock

@@ -82,15 +82,6 @@ GEM
       rake (>= 10.4, < 14.0)
     ast (2.4.2)
     aws-eventstream (1.2.0)
-    aws-partitions (1.730.0)
-    aws-sdk-core (3.170.1)
-      aws-eventstream (~> 1, >= 1.0.2)
-      aws-partitions (~> 1, >= 1.651.0)
-      aws-sigv4 (~> 1.5)
-      jmespath (~> 1, >= 1.6.1)
-    aws-sdk-elasticsearchservice (1.69.0)
-      aws-sdk-core (~> 3, >= 3.165.0)
-      aws-sigv4 (~> 1.1)
     aws-sigv4 (1.5.2)
       aws-eventstream (~> 1, >= 1.0.2)
     bcrypt (3.1.18)
@@ -127,14 +118,6 @@ GEM
     dotenv-rails (2.8.1)
       dotenv (= 2.8.1)
       railties (>= 3.2)
-    elasticsearch (6.8.0)
-      elasticsearch-api (= 6.8.0)
-      elasticsearch-transport (= 6.8.0)
-    elasticsearch-api (6.8.0)
-      multi_json
-    elasticsearch-transport (6.8.0)
-      faraday
-      multi_json
     em-websocket (0.5.3)
       eventmachine (>= 0.12.9)
       http_parser.rb (~> 0)
@@ -156,17 +139,11 @@ GEM
     globalid (1.1.0)
       activesupport (>= 5.0)
     google-protobuf (3.22.2)
-    graphiql-rails (1.9.0)
-      railties
-      sprockets-rails
     graphql (2.0.19)
     hashdiff (1.0.1)
     http_parser.rb (0.8.0)
     i18n (1.12.0)
       concurrent-ruby (~> 1.0)
-    jbuilder (2.11.5)
-      actionview (>= 5.0.0)
-      activesupport (>= 5.0.0)
     jekyll (4.3.2)
       addressable (~> 2.4)
       colorator (~> 1.0)
@@ -194,7 +171,6 @@ GEM
       jekyll (>= 3.8, < 5.0)
     jekyll-watch (2.2.1)
       listen (~> 3.0)
-    jmespath (1.6.2)
     json (2.6.3)
     jwt (2.7.0)
     kramdown (2.4.0)
@@ -404,29 +380,22 @@ PLATFORMS
 
 DEPENDENCIES
   annotate
-  aws-sdk-elasticsearchservice
   bootsnap
   byebug
   capybara
   climate_control
   devise
   dotenv-rails
-  elasticsearch (~> 6.8)
   faraday_middleware-aws-sigv4
   flipflop
-  graphiql-rails
   graphql
-  jbuilder
   jekyll
   jekyll-remote-theme
   jekyll-seo-tag
   jwt
   listen
   lograge
   mitlibraries-theme!
-  net-imap
-  net-pop
-  net-smtp
   opensearch-ruby
   pg
   puma

README.md

@@ -2,10 +2,7 @@
 
 # TIMDEX Is Making Discovery EXcellent @ MIT
 
-This application interfaces with an ElasticSearch backend and exposes a set of
-API Endpoints to allow registered users to query our data.
-
-The backend is populated via [pipelines](https://github.com/MITLibraries/mario).
+This application interfaces with an OpenSearch backend and exposes a GraphQL endpoint to allow anonymous users to query our data.
 
 ## Architecture Decision Records
 
@@ -22,26 +19,6 @@ additional records with a standardized template.
 - don't commit your .env or .env.development, but do commit .env.test after
   confirming your test values are not actual secrets that need protecting
 
-### Updating the data model
-Updating the data model is somewhat complicated because many files need to be
-edited across multiple repositories and deployment steps should happen in a
-particular order so as not to break production services.
-- Start by updating the data model in [Mario](https://github.com/MITLibraries/mario). Instructions for that can be found
-  in the [Mario README](https://github.com/MITLibraries/mario/blob/master/README.md). Then complete the following steps here in TIMDEX.
-- Update `app/models/search.rb` to build/update/remove queries for the added/
-  edited/deleted fields as appropriate. Make sure to update filters and
-  aggregations if relevant to the changed fields.
-- Update `app/views/api/[version]/search/_base_json_jbuilder` to
-  add/update/remove changed fields OR update
-  `views/api/[version]/search/_extended_json_jbuilder` if the changed fields
-  aren’t/shouldn’t be in the brief record result.
-- If changed fields should be aggregated, update
-  `views/api/[version]/search/_aggregations_json_jbuilder` as appropriate.
-- Update tests as necessary. Make sure to test with all current data
-  source samples ingested into a local ES instance.
-- Update `openapi.json` to make sure our spec matches any changes made
-  (including bumping the version number).
-
 ## Publishing User Facing Documentation
 
 ### Running jekyll documentation locally
@@ -79,46 +56,32 @@ locally.
   `yourapp.herokuapp.com`. However, if you use a custom domain in production,
   that should be the value you use in production.
 - `JWT_SECRET_KEY`: generate with `rails secret`
-- `ELASTICSEARCH_INDEX`: Elasticsearch index or alias to query
-- `ELASTICSEARCH_URL`: defaults to `http://localhost:9200`
 
 ## Production required Environment Variables
 
-- `AWS_ACCESS_KEY`
-- `AWS_ELASTICSEARCH`: boolean. Set to true to enable AWSv4 Signing
-- `AWS_SECRET_ACCESS_KEY`
+- `AWS_OPENSEARCH`: boolean. Set to true to enable AWSv4 Signing
+- `AWS_OPENSEARCH_ACCESS_KEY_ID`
+- `AWS_OPENSEARCH_SECRET_ACCESS_KEY`
 - `AWS_REGION`
+- `OPENSEARCH_INDEX`: Opensearch index or alias to query, default will be to search all indexes which is generally not
+                      expected. `timdex` or `all-current` are aliases used consistently in our data pipelines, with
+                      `timdex` being most likely what most use cases will want.
+- `OPENSEARCH_URL`: Opensearch URL, defaults to `http://localhost:9200`
 - `SMTP_ADDRESS`
 - `SMTP_PASSWORD`
 - `SMTP_PORT`
 - `SMTP_USER`
 
-### Additional required Environment Variables when Opensearch is enabled (aka v2=true)
-
-- `v2`: set to `true`
-- `OPENSEARCH_INDEX`: Opensearch index or alias to query
-- `OPENSEARCH_URL`: Opensearch URL
-- `OPENSEARCH_LOG` set to `true`
-
-- `AWS_OPENSEARCH`
-- `AWS_OPENSEARCH_ACCESS_KEY_ID`
-- `AWS_OPENSEARCH_SECRET_ACCESS_KEY`
 ## Optional Environment Variables (all ENVs)
 
-- `ELASTICSEARCH_LOG` if `true`, verbosely logs ElasticSearch queries.
+- `OPENSEARCH_LOG` if `true`, verbosely logs OpenSearch queries.
 
   ```text
   NOTE: do not set this ENV at all if you want ES logging fully disabled.
   Setting it to `false` is still setting it and you will be annoyed and
   confused.
   ```
 
-- `ES_LOG_LEVEL` set elasticsearch transport log level. Defaults to `INFO`.
-
-```text
-NOTE: `ELASTICSEARCH_LOG` must also be set for logging to function.
-```
-
 - `PREFERRED_DOMAIN` - set this to the domain you would like to to use. Any
   other requests that come to the app will redirect to the root of this domain.
   This is useful to prevent access to herokuapp.com domains.
@@ -127,47 +90,3 @@ NOTE: `ELASTICSEARCH_LOG` must also be set for logging to function.
   Default is 1.
 - `SENTRY_DSN`: client key for Sentry exception logging
 - `SENTRY_ENV`: Sentry environment for the application. Defaults to 'unknown' if unset.
-
-## Docker Compose Orchestrated Local Environment
-
-This section will describe how to use the included docker compose files to spin up ElasticSearch
-and optionally use Mario to load sample data for testing.
-
-You may set `ELASTICSEARCH_URL` to `http://0.0.0.0:9200` to use this ES instance in development if you
-choose to not use the included Dockerfile
-
-### Startup ElasticSearch and Timdex
-
-`make up`
-
-### Shutdown ElasticSearch and Timdex when you are done
-
-`make down`
-
-### Optionally, load sample data
-
-After ElasticSearch is running from `make up` command:
-
-`make sampledata`
-
-Note: if you run this and it fails, try again in a few seconds as ES may still be loading
-
-### Run arbitrary Mario commands
-
-You can also run arbitrary Mario commands using a syntax like this after first running `make up`.
-
-`docker run --network timdex_default mitlibraries/mario --url http://elasticsearch:9200 YOUR_MARIO_COMMAND_HERE [e.g. indexes]`
-
-Note: if you have no indexes loaded, many mario commands will fail. Try `make sampledata` or load the data you
-need before proceeding.
-
-### Quick curl examples with sample data in mind
-
-`curl 'http://0.0.0.0:3000/api/v1/ping'`
-
-`curl 'http://0.0.0.0:3000/api/v1/search?q=archives'`
-
-You can also use the playground via your browser, see the index page of the running app for a link
-and example queries.
-
-http://0.0.0.0:3000