cocina-models 0.113.0 → 0.115.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '09669a342a43a6be309144d1d71748372251fdb37ad96c480e7ce88127749ec5'
-  data.tar.gz: 99e35e0634f65f03e545dfabc40d95cb6a638c1f0f0e05ffbf14315738ac895c
+  metadata.gz: 4d9e710c098333b0b95f15f1f624a365b7067e3ef638d48e7f04bcbc72f2b038
+  data.tar.gz: 2bcb210b1f563ddc1916fe8a718f326184c75eb077f3c931372083ad1bf8c179
 SHA512:
-  metadata.gz: 29c840e670181b57df7d29573da1cc019cd6a34300f43548a762f0d394736af09822bb61e85acd7eefa6426e37996fbe5869c4b58c8aa7e9176a201af14f91ec
-  data.tar.gz: 00f39759126179cd79f5cda1a1cabf93b3da2ba19ef0ecd90fa1f1c10735b31ff966e9496e93cac69fb35255b0dc1c1e45176d70f743630005eb883384e38132
+  metadata.gz: ec2b6a2e5a92c4cc712e09e2edc23a6975c37eb71f13133e1400d21d999880f1a79a8962624bdd339cffaa08e88a0fffd4395c716b1ed49360c5045a43edf666
+  data.tar.gz: 69f0f8869d2536477765b033c39387ef4693f648e50747fa49e15d1cb1cd27be7bc830dec49bf5beb06a1a04472ddf78750201c60c31d002f47006ed72c99751

data/.circleci/config.yml

@@ -11,5 +11,5 @@ workflows:
           context: dlss
           before-test:
             - run:
-                name: validate openapi
+                name: validate schema
                 command: bin/validate-schema schema.json

data/.rubocop.yml

@@ -540,7 +540,6 @@ Style/RedundantArrayFlatten: # new in 1.76
   Enabled: true
 Style/ArrayIntersectWithSingleElement: # new in 1.81
   Enabled: true
-
 Style/EmptyClassDefinition: # new in 1.83
   Enabled: true
 Style/ModuleMemberExistenceCheck: # new in 1.82
@@ -552,4 +551,30 @@ Style/ReverseFind: # new in 1.83
 RSpec/LeakyLocalVariable: # new in 3.8
   Enabled: true
 RSpec/Output: # new in 3.9
-  Enabled: true
+  Enabled: true
+Lint/DataDefineOverride: # new in 1.85
+  Enabled: true
+Lint/UnreachablePatternBranch: # new in 1.85
+  Enabled: true
+Style/FileOpen: # new in 1.85
+  Enabled: true
+Style/MapJoin: # new in 1.85
+  Enabled: true
+Style/OneClassPerFile: # new in 1.85
+  Enabled: true
+Style/PartitionInsteadOfDoubleSelect: # new in 1.85
+  Enabled: true
+Style/PredicateWithKind: # new in 1.85
+  Enabled: true
+Style/ReduceToHash: # new in 1.85
+  Enabled: true
+Style/RedundantMinMaxBy: # new in 1.85
+  Enabled: true
+Style/RedundantStructKeywordInit: # new in 1.85
+  Enabled: true
+Style/SelectByKind: # new in 1.85
+  Enabled: true
+Style/SelectByRange: # new in 1.85
+  Enabled: true
+Style/TallyMethod: # new in 1.85
+  Enabled: true

data/Gemfile

@@ -7,3 +7,4 @@ gemspec
 
 gem 'debug'
 gem 'rspec_junit_formatter' # For CircleCI
+gem 'ruby-progressbar'

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    cocina-models (0.113.0)
+    cocina-models (0.115.0)
       activesupport
       deprecation
       dry-struct (~> 1.0)
@@ -59,7 +59,7 @@ GEM
       concurrent-ruby (~> 1.0)
       dry-core (~> 1.1)
       zeitwerk (~> 2.6)
-    dry-struct (1.8.0)
+    dry-struct (1.8.1)
       dry-core (~> 1.1)
       dry-types (~> 1.8, >= 1.8.2)
       ice_nine (~> 0.11)
@@ -86,8 +86,8 @@ GEM
       prism (>= 1.3.0)
       rdoc (>= 4.0.0)
       reline (>= 0.4.2)
-    json (2.18.1)
-    json-schema (6.1.0)
+    json (2.19.1)
+    json-schema (6.2.0)
       addressable (~> 2.8)
       bigdecimal (>= 3.1, < 5)
     json_schemer (2.5.0)
@@ -100,7 +100,7 @@ GEM
     language_server-protocol (3.17.0.5)
     lint_roller (1.1.0)
     logger (1.7.0)
-    mcp (0.7.1)
+    mcp (0.8.0)
       json-schema (>= 4.1)
     minitest (6.0.2)
       drb (~> 2.0)
@@ -124,7 +124,7 @@ GEM
     psych (5.3.1)
       date
       stringio
-    public_suffix (7.0.2)
+    public_suffix (7.0.5)
     racc (1.8.1)
     rainbow (3.1.1)
     rake (13.3.1)
@@ -150,7 +150,7 @@ GEM
     rspec-support (3.13.7)
     rspec_junit_formatter (0.6.0)
       rspec-core (>= 2, < 4, != 2.12.0)
-    rubocop (1.85.0)
+    rubocop (1.85.1)
       json (~> 2.3)
       language_server-protocol (~> 3.17.0.2)
       lint_roller (~> 1.1.0)
@@ -162,7 +162,7 @@ GEM
       rubocop-ast (>= 1.49.0, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 4.0)
-    rubocop-ast (1.49.0)
+    rubocop-ast (1.49.1)
       parser (>= 3.3.7.2)
       prism (~> 1.7)
     rubocop-rake (0.7.1)
@@ -209,6 +209,7 @@ DEPENDENCIES
   rubocop (~> 1.24)
   rubocop-rake
   rubocop-rspec
+  ruby-progressbar
   simplecov
 
 CHECKSUMS
@@ -218,7 +219,7 @@ CHECKSUMS
   attr_extras (7.1.0) sha256=d96fc9a9dd5d85ba2d37762440a816f840093959ae26bb90da994c2d9f1fc827
   base64 (0.3.0) sha256=27337aeabad6ffae05c265c450490628ef3ebd4b67be58257393227588f5a97b
   bigdecimal (4.0.1) sha256=8b07d3d065a9f921c80ceaea7c9d4ae596697295b584c296fe599dd0ad01c4a7
-  cocina-models (0.113.0)
+  cocina-models (0.115.0)
   concurrent-ruby (1.3.6) sha256=6b56837e1e7e5292f9864f34b69c5a2cbc75c0cf5338f1ce9903d10fa762d5ab
   connection_pool (3.0.2) sha256=33fff5ba71a12d2aa26cb72b1db8bba2a1a01823559fb01d29eb74c286e62e0a
   date (3.5.1) sha256=750d06384d7b9c15d562c76291407d89e368dda4d4fff957eb94962d325a0dc0
@@ -230,7 +231,7 @@ CHECKSUMS
   dry-core (1.2.0) sha256=0cc5a7da88df397f153947eeeae42e876e999c1e30900f3c536fb173854e96a1
   dry-inflector (1.3.1) sha256=7fb0c2bb04f67638f25c52e7ba39ab435d922a3a5c3cd196120f63accb682dcc
   dry-logic (1.6.0) sha256=da6fedbc0f90fc41f9b0cc7e6f05f5d529d1efaef6c8dcc8e0733f685745cea2
-  dry-struct (1.8.0) sha256=74c38b559924fb6462ac43ec780c4533a082d7b1d238a8d7857b773b3b8e2966
+  dry-struct (1.8.1) sha256=033868594c45241540172bf1ebbc8bb76b72b4f0717072325deba38ac13e80f1
   dry-types (1.9.1) sha256=baebeecdb9f8395d6c9d227b62011279440943e3ef2468fe8ccc1ba11467f178
   edtf (3.2.0) sha256=a15a0ee274e49c8047a3ebb5d61d793ba44f7f8ffbf0595392c467e3ea8d2447
   equivalent-xml (0.6.0) sha256=8919761efa848ad0846369ff8be1f646b17e5061698c4867b09829000cc3f487
@@ -240,14 +241,14 @@ CHECKSUMS
   ice_nine (0.11.2) sha256=5d506a7d2723d5592dc121b9928e4931742730131f22a1a37649df1c1e2e63db
   io-console (0.8.2) sha256=d6e3ae7a7cc7574f4b8893b4fca2162e57a825b223a177b7afa236c5ef9814cc
   irb (1.17.0) sha256=168c4ddb93d8a361a045c41d92b2952c7a118fa73f23fe14e55609eb7a863aae
-  json (2.18.1) sha256=fe112755501b8d0466b5ada6cf50c8c3f41e897fa128ac5d263ec09eedc9f986
-  json-schema (6.1.0) sha256=6bf70a2cfb6dfd5a06da28093fa8190f324c88eabd36a7f47097f227321dc702
+  json (2.19.1) sha256=dd94fdc59e48bff85913829a32350b3148156bc4fd2a95a2568a78b11344082d
+  json-schema (6.2.0) sha256=e8bff46ed845a22c1ab2bd0d7eccf831c01fe23bb3920caa4c74db4306813666
   json_schemer (2.5.0) sha256=2f01fb4cce721a4e08dd068fc2030cffd0702a7f333f1ea2be6e8991f00ae396
   jsonpath (1.1.5) sha256=29f70467193a2dc93ab864ec3d3326d54267961acc623f487340eb9c34931dbe
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
   lint_roller (1.1.0) sha256=2c0c845b632a7d172cb849cc90c1bce937a28c5c8ccccb50dfd46a485003cc87
   logger (1.7.0) sha256=196edec7cc44b66cfb40f9755ce11b392f21f7967696af15d274dde7edff0203
-  mcp (0.7.1) sha256=fa967895d6952bad0d981ea907731d8528d2c246d2079d56a9c8bae83d14f1c7
+  mcp (0.8.0) sha256=ae8bd146bb8e168852866fd26f805f52744f6326afb3211e073f78a95e0c34fb
   minitest (6.0.2) sha256=db6e57956f6ecc6134683b4c87467d6dd792323c7f0eea7b93f66bd284adbc3d
   multi_json (1.19.1) sha256=7aefeff8f2c854bf739931a238e4aea64592845e0c0395c8a7d2eea7fdd631b7
   nokogiri (1.19.1-arm64-darwin) sha256=dfe2d337e6700eac47290407c289d56bcf85805d128c1b5a6434ddb79731cb9e
@@ -260,7 +261,7 @@ CHECKSUMS
   prettyprint (0.2.0) sha256=2bc9e15581a94742064a3cc8b0fb9d45aae3d03a1baa6ef80922627a0766f193
   prism (1.9.0) sha256=7b530c6a9f92c24300014919c9dcbc055bf4cdf51ec30aed099b06cd6674ef85
   psych (5.3.1) sha256=eb7a57cef10c9d70173ff74e739d843ac3b2c019a003de48447b2963d81b1974
-  public_suffix (7.0.2) sha256=9114090c8e4e7135c1fd0e7acfea33afaab38101884320c65aaa0ffb8e26a857
+  public_suffix (7.0.5) sha256=1a8bb08f1bbea19228d3bed6e5ed908d1cb4f7c2726d18bd9cadf60bc676f623
   racc (1.8.1) sha256=4a7f6929691dbec8b5209a0b373bc2614882b55fc5d2e447a21aaa691303d62f
   rainbow (3.1.1) sha256=039491aa3a89f42efa1d6dec2fc4e62ede96eb6acd95e52f1ad581182b79bc6a
   rake (13.3.1) sha256=8c9e89d09f66a26a01264e7e3480ec0607f0c497a861ef16063604b1b08eb19c
@@ -273,8 +274,8 @@ CHECKSUMS
   rspec-mocks (3.13.8) sha256=086ad3d3d17533f4237643de0b5c42f04b66348c28bf6b9c2d3f4a3b01af1d47
   rspec-support (3.13.7) sha256=0640e5570872aafefd79867901deeeeb40b0c9875a36b983d85f54fb7381c47c
   rspec_junit_formatter (0.6.0) sha256=40dde674e6ae4e6cc0ff560da25497677e34fefd2338cc467a8972f602b62b15
-  rubocop (1.85.0) sha256=317407feb681a07d54f64d2f9e1d6b6af1ce7678e51cd658e3ad8bd66da48c01
-  rubocop-ast (1.49.0) sha256=49c3676d3123a0923d333e20c6c2dbaaae2d2287b475273fddee0c61da9f71fd
+  rubocop (1.85.1) sha256=3dbcf9e961baa4c376eeeb2a03913dca5e3987033b04d38fa538aa1e7406cc77
+  rubocop-ast (1.49.1) sha256=4412f3ee70f6fe4546cc489548e0f6fcf76cafcfa80fa03af67098ffed755035
   rubocop-rake (0.7.1) sha256=3797f2b6810c3e9df7376c26d5f44f3475eda59eb1adc38e6f62ecf027cbae4d
   rubocop-rspec (3.9.0) sha256=8fa70a3619408237d789aeecfb9beef40576acc855173e60939d63332fdb55e2
   ruby-progressbar (1.13.0) sha256=80fc9c47a9b640d6834e0dc7b3c94c9df37f08cb072b7761e4a71e22cff29b33
@@ -294,4 +295,4 @@ CHECKSUMS
   zeitwerk (2.7.5) sha256=d8da92128c09ea6ec62c949011b00ed4a20242b255293dd66bf41545398f73dd
 
 BUNDLED WITH
-  4.0.7
+  4.0.8

data/README.md

@@ -6,7 +6,7 @@
 
 The cocina-models gem is a Ruby implementation of the Stanford Digital Repository (SDR) data model, which we named "Cocina." The data being modeled is oriented around digital repository objects.
 
-The data model is expressed in an OpenAPI specification that lives in this codebase. Expressing the model in such a spec allows for rich validation (using gems such as `json_schemer`). The gem provides a set of generators (see below) to generate Ruby classes from the specification, with modeling provided by dry-struct / dry-types. Together, these provide a way for consumers to validate objects against models and to manipulate those objects.
+The data model is expressed in an JSON Schema specification that lives in this codebase. Expressing the model in such a spec allows for rich validation (using gems such as `json_schemer`). The gem provides a set of generators (see below) to generate Ruby classes from the specification, with modeling provided by dry-struct / dry-types. Together, these provide a way for consumers to validate objects against models and to manipulate those objects.
 
 Note that the data model encodes properties as camelCase, which the team believes to be consistent with other HTTP APIs and the original design of the Cocina data model. While using camelCase in Ruby code may look and feel wrong, we did explore automagic conversion between camelCase in the model and snake_case in the Ruby context. We ultimately concluded that we have enough representations of the data model in enough codebases to reasonably worry about data inconsistency problems, none of which we need in our work on SDR.
 
@@ -55,7 +55,10 @@ Beyond what is necessary to test the generator, the Cocina model classes are not
 
 ## Testing validation changes
 
-If there is a possibility that a model, mapping, or validation change will conflict with some existing objects then [validate-cocina](https://github.com/sul-dlss/dor-services-app/blob/main/bin/validate-cocina) should be used for testing. This must be run on the `sdr-infra` VM since it requires deploying a branch of cocina-models.
+If there is a possibility that a model, mapping, or validation change will conflict with some existing objects then `bin/validate-data` should be used for testing.  This operates on a sample of objects from the repository and reports any validation errors. You may get the sample by running the script [bin/export-cocina-head-versions](https://github.com/sul-dlss/dor-services-app?tab=readme-ov-file#export-data) and downloading the data file to your computer. Running a full validation takes about 2 hours.
+
+
+Alternatively, you can use [validate-cocina](https://github.com/sul-dlss/dor-services-app/blob/main/bin/validate-cocina) for testing. This must be run on the `sdr-infra` VM since it requires deploying a branch of cocina-models.  It is slower than using `bin/validate-data`, but all of the data is completely up to date.
 
 For background on object validation, as it relates to migrating versions, see: https://github.com/sul-dlss/dor-services-app/wiki/Migrating-Cocina
 
@@ -153,21 +156,7 @@ This list of services is known to include:
 * [sul-dlss/sdr-api](https://github.com/sul-dlss/sdr-api)
 * [sul-dlss/dor-services-app](https://github.com/sul-dlss/dor-services-app/)
 
-
-#### Step 3A: Update API specifications
-
-**NOTE**: You can skip step 3A if there have not been any changes to the `cocina-models` OpenAPI spec since the prior release.
-
-The cocina-models gem is used in applications that have an API specification that accepts Cocina models.
-
-#### Step 3B: Bump gems and create the PRs
-
-If you updated the `schema.json` in step 3A, use the same PR for step 3B. Why? When [dor-services-app](https://github.com/sul-dlss/dor-services-app), for example, is updated to use the new models (via the auto-update script), these clients should be updated at the same time or there is risk of models produced by dor-services-app not being acceptable to the clients.
-
-1. Perform `bundle update --conservative cocina-models dor-services-client` in the  services above and make PRs for those repos if they don't already exist. You may first need to update how these gems are pinned in the `Gemfile` in order to bump them.
-2. Note that sdr-client is not currently used in these applications, but if it were, would also need to be bumped to the latest release.
-
-#### Step 3C: Merge 'em
+Perform `bundle update --conservative cocina-models dor-services-client` in the services above and make PRs for those repos. You may first need to update how these gems are pinned in the `Gemfile` in order to bump them.
 
 Get the directly coupled services PRs merged before the deploy in step 5.
 

data/bin/validate-data

@@ -0,0 +1,287 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+# Validate JSONL (XZ compressed) files against the schema
+
+# Trap Ctrl+C to exit gracefully
+Signal.trap('INT') do
+  puts "\nTerminated"
+  exit(1)
+end
+
+Signal.trap('TERM') do
+  puts "\nTerminated"
+  exit(1)
+end
+
+require 'bundler/setup'
+require 'cocina/models'
+require 'json'
+require 'ruby-progressbar'
+require 'optparse'
+
+# Parse command line options
+def parse_options # rubocop:disable Metrics/MethodLength
+  options = {
+    processes: 12,
+    count: nil,
+    batch_size: 100
+  }
+
+  OptionParser.new do |opts|
+    opts.banner = 'Usage: validate-data FILENAME [options]'
+
+    opts.on('-p', '--processes NUM', Integer, 'Number of processes to use (default: 12)') do |p|
+      options[:processes] = p
+    end
+
+    opts.on('-c', '--count NUM', Integer, 'Total line count (skips counting pass if provided)') do |c|
+      options[:count] = c
+    end
+
+    opts.on('-b', '--batch-size NUM', Integer, 'Batch size for worker processing (default: 100)') do |b|
+      options[:batch_size] = b
+    end
+
+    opts.on('-h', '--help', 'Display this help message') do
+      puts opts
+      exit
+    end
+  end.parse!
+
+  # Filename is required as a positional argument
+  if ARGV.empty?
+    puts 'Error: FILENAME is required'
+    puts 'Usage: validate-data FILENAME [options]'
+    puts 'Run with --help for more information'
+    exit 1
+  end
+
+  options[:filename] = ARGV[0]
+  options
+end
+
+# Count lines in the file
+def count_lines(filename)
+  count = 0
+  IO.popen(['xzcat', filename]) do |io|
+    io.each_line { count += 1 }
+  end
+  count
+end
+
+# Get total line count (either from option or by counting)
+def get_total_lines(filename, provided_count)
+  if provided_count
+    puts "Using provided line count: #{provided_count}"
+    provided_count
+  else
+    puts 'Counting lines...'
+    total = count_lines(filename)
+    puts "Total lines to validate: #{total}"
+    total
+  end
+end
+
+# Worker process that reads batches from a pipe and validates
+def worker_process(reader) # rubocop:disable Metrics/MethodLength
+  errors = []
+
+  loop do
+    # Read length prefix (4 bytes)
+    length_data = reader.read(4)
+    break if length_data.nil? || length_data.empty?
+
+    length = length_data.unpack1('N')
+    data = reader.read(length)
+    batch = Marshal.load(data) # rubocop:disable Security/MarshalLoad
+
+    # Process each line in the batch
+    batch.each do |line_num, line_content|
+      json = JSON.parse(line_content)
+      Cocina::Models.build(json)
+    rescue JSON::ParserError => e
+      errors << { line: line_num, error: "JSON Parse Error: #{e.message}" }
+    rescue Cocina::Models::ValidationError => e
+      errors << { line: line_num, error: "Validation Error: #{json['externalIdentifier']} - #{e.message}" }
+    rescue Cocina::Models::UnknownTypeError => e
+      errors << { line: line_num, error: "Unknown Type Error: #{e.message}" }
+    rescue StandardError => e
+      errors << { line: line_num, error: "Error: #{e.class} - #{e.message}" }
+    end
+  end
+
+  errors
+end
+
+# Spawn worker processes
+def spawn_workers(num_processes) # rubocop:disable Metrics/MethodLength
+  workers = []
+  result_readers = []
+
+  num_processes.times do
+    work_reader, work_writer = IO.pipe
+    result_reader, result_writer = IO.pipe
+
+    pid = fork do
+      # Child process
+
+      # Close all writer pipes inherited from the parent for OTHER workers.
+      workers.each { |w| w[:writer].close }
+
+      # Now, close the handles for this specific worker that the child doesn't need.
+      work_writer.close
+      result_reader.close
+
+      errors = worker_process(work_reader)
+
+      # Send results back
+      result_writer.write(Marshal.dump(errors))
+      result_writer.close
+      work_reader.close
+      exit(0)
+    end
+
+    # Parent process
+    work_reader.close
+    result_writer.close
+
+    workers << { pid: pid, writer: work_writer }
+    result_readers << result_reader
+  end
+
+  [workers, result_readers]
+end
+
+# Send a batch to a worker
+def send_batch(worker, batch)
+  return if batch.empty?
+
+  data = Marshal.dump(batch)
+  worker[:writer].write([data.bytesize].pack('N'))
+  worker[:writer].write(data)
+end
+
+# Stream file and distribute work to workers
+def distribute_work(filename, workers, batch_size, total_lines) # rubocop:disable Metrics/MethodLength
+  line_number = 0
+  current_worker = 0
+  batch = []
+
+  # Create progress bar
+  progressbar = ProgressBar.create(
+    title: 'Validating',
+    total: total_lines,
+    format: '%t: |%B| %p%% %c/%C %a %e',
+    throttle_rate: 0.1
+  )
+
+  IO.popen(['xzcat', filename]) do |io|
+    io.each_line do |line|
+      line_number += 1
+      batch << [line_number, line]
+
+      # When batch is full, send to worker
+      if batch.size >= batch_size
+        worker = workers[current_worker]
+        send_batch(worker, batch)
+        batch = []
+        current_worker = (current_worker + 1) % workers.length
+      end
+
+      # Update progress bar
+      progressbar.increment
+    end
+  end
+
+  # Send any remaining lines in the last batch
+  if batch.any?
+    worker = workers[current_worker]
+    send_batch(worker, batch)
+  end
+
+  # Final progress update
+  progressbar.finish
+end
+
+# Collect results from all workers
+def collect_results(workers, result_readers)
+  # Close all worker input pipes to signal completion
+  workers.each { |w| w[:writer].close }
+
+  puts 'Collecting results from workers...'
+  all_errors = []
+
+  result_readers.each do |reader|
+    data = reader.read
+    worker_errors = Marshal.load(data) # rubocop:disable Security/MarshalLoad
+    all_errors.concat(worker_errors)
+    reader.close
+  end
+
+  # Wait for all workers to complete
+  workers.each { |w| Process.wait(w[:pid]) }
+
+  all_errors
+end
+
+# Print validation summary
+def print_summary(total_lines, errors, elapsed_time) # rubocop:disable Metrics/MethodLength
+  puts '=' * 80
+  puts 'VALIDATION SUMMARY'
+  puts '=' * 80
+  puts "Total lines processed: #{total_lines}"
+  puts "Lines with errors: #{errors.length}"
+  puts "Success rate: #{((total_lines - errors.length).to_f / total_lines * 100).round(2)}%"
+  puts "Time elapsed: #{elapsed_time.round(2)} seconds"
+  puts "Throughput: #{(total_lines / elapsed_time).round(0)} lines/second"
+
+  return unless errors.any?
+
+  puts "\n"
+  puts 'Error details:'
+  puts '-' * 80
+  # Sort errors by line number for better readability
+  errors.sort_by { |e| e[:line] }.each do |error|
+    puts "Line #{error[:line]}: #{error[:error]}"
+  end
+  puts "\n"
+  puts "Line numbers with errors: #{errors.sort_by { |e| e[:line] }.map { |e| e[:line] }.join(', ')}"
+end
+
+# Main execution
+def main
+  options = parse_options
+
+  puts "Validating file: #{options[:filename]}"
+  puts "Using #{options[:processes]} processes with batch size #{options[:batch_size]}"
+  puts '=' * 80
+
+  # Get total line count
+  total_lines = get_total_lines(options[:filename], options[:count])
+  puts '=' * 80
+
+  # Spawn worker processes
+  workers, result_readers = spawn_workers(options[:processes])
+
+  # Start timing
+  start_time = Time.now
+
+  # Distribute work to workers
+  distribute_work(options[:filename], workers, options[:batch_size], total_lines)
+
+  # Collect results
+  all_errors = collect_results(workers, result_readers)
+
+  # Calculate elapsed time
+  elapsed_time = Time.now - start_time
+
+  # Print summary
+  print_summary(total_lines, all_errors, elapsed_time)
+
+  # Exit with appropriate code
+  exit(all_errors.empty? ? 0 : 1)
+end
+
+# Run the script
+main

data/lib/cocina/generator/datatype.rb

@@ -2,7 +2,7 @@
 
 module Cocina
   module Generator
-    # Class for generating from an openapi schema
+    # Class for generating from a JSON schema
     class Datatype < SchemaBase
       def generate
         <<~RUBY

data/lib/cocina/generator/generator.rb

@@ -4,7 +4,7 @@ require 'fileutils'
 
 module Cocina
   module Generator
-    # Class for generating Cocina models from openapi.
+    # Class for generating Cocina models from JSON Schema.
     class Generator < Thor # rubocop:disable Metrics/ClassLength
       include Thor::Actions
 

data/lib/cocina/generator/schema.rb

@@ -2,7 +2,7 @@
 
 module Cocina
   module Generator
-    # Class for generating from an openapi schema
+    # Class for generating from a JSON schema
     class Schema < SchemaBase
       def schema_properties
         @schema_properties ||= (properties + all_of_properties + one_of_properties).uniq(&:key)
@@ -112,7 +112,7 @@ module Cocina
                                                    key: key,
                                                    # The property does less validation because may vary between
                                                    # different oneOf schemas. Validation is still performed
-                                                   # by openAPI.
+                                                   # by JSON Schema.
                                                    relaxed: true,
                                                    parent: self,
                                                    schemas: schemas)

data/lib/cocina/generator/schema_array.rb

@@ -2,7 +2,7 @@
 
 module Cocina
   module Generator
-    # Class for generating from an openapi array
+    # Class for generating from a JSON Schema array
     class SchemaArray < SchemaBase
       GENERIC_ITEMS_NAME = 'items'
 

data/lib/cocina/generator/schema_base.rb

@@ -2,7 +2,7 @@
 
 module Cocina
   module Generator
-    # Base class for generating from openapi
+    # Base class for generating from a JSON Schema
     class SchemaBase
       attr_reader :schema_doc, :key, :required, :nullable, :parent, :relaxed, :schemas, :lite
 
@@ -74,7 +74,7 @@ module Cocina
       def relaxed_comment
         return '' unless relaxed
 
-        "# Validation of this property is relaxed. See the openapi for full validation.\n"
+        "# Validation of this property is relaxed. See the schema.json for full validation.\n"
       end
 
       # dry-types-based types contain the word `Types` (e.g., `Types::String`), and custom types (e.g., `SourceId`) do not

data/lib/cocina/generator/schema_ref.rb

@@ -2,7 +2,7 @@
 
 module Cocina
   module Generator
-    # Class for generating from an openapi reference
+    # Class for generating from a JSON Schema reference
     class SchemaRef < SchemaBase
       def generate
         if required && !relaxed

data/lib/cocina/generator/schema_value.rb

@@ -2,7 +2,7 @@
 
 module Cocina
   module Generator
-    # Class for generating from an openapi value
+    # Class for generating from a JSON Schema value
     class SchemaValue < SchemaBase
       def generate
         if required && !relaxed

data/lib/cocina/generator.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 module Cocina
-  # Module for generating Cocina models from openapi.
+  # Module for generating Cocina models from a JSON Schema.
   module Generator
   end
 end

data/lib/cocina/json_schema_wrapper.rb

@@ -3,8 +3,7 @@
 module Cocina
   # Wrapper for JSON Schema support using json_schemer
   class JsonSchemaWrapper
-    class OpenApiError < StandardError; end
-    class MissingReferenceError < OpenApiError; end
+    class MissingReferenceError < StandardError; end
 
     def initialize(spec_hash, strict_reference_validation: true)
       @spec = spec_hash

data/lib/cocina/models/access.rb

@@ -4,15 +4,15 @@ module Cocina
   module Models
     class Access < Struct
       # Access level.
-      # Validation of this property is relaxed. See the openapi for full validation.
+      # Validation of this property is relaxed. See the schema.json for full validation.
       attribute? :view, Types::Strict::String.optional.default('dark')
       # Download access level.
-      # Validation of this property is relaxed. See the openapi for full validation.
+      # Validation of this property is relaxed. See the schema.json for full validation.
       attribute? :download, Types::Strict::String.optional.default('none')
       # Not used for this access type, must be null.
-      # Validation of this property is relaxed. See the openapi for full validation.
+      # Validation of this property is relaxed. See the schema.json for full validation.
       attribute? :location, Types::Strict::String.optional
-      # Validation of this property is relaxed. See the openapi for full validation.
+      # Validation of this property is relaxed. See the schema.json for full validation.
       attribute? :controlledDigitalLending, Types::Strict::Bool.optional.default(false)
     end
   end