digital_scriptorium 0.2.0 → 0.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 90d97605a47a87aec5fb8dc41385f9ce55e415d8d96bd51bc91f0436b8d5ec07
-  data.tar.gz: 873ab9fbb3d1fb275419ec753c36929f1a806fe40a9023aaa3279b2684871218
+  metadata.gz: 62b823f25e2940c6a68ee4ea8db949cebbdc95f9d31682d72a60dbbb62c0a0ff
+  data.tar.gz: 53c69f7e20af8b7efc327214c4bfbb8378f6d80133106458231277961f4ed613
 SHA512:
-  metadata.gz: 0142d571dd96cd21270a782c0327bf0798af63e00d904eb71263d7a43fb8eb69e4fbe2854ad75dcf4b4bed7b56148e81a4a0be1862bff0e76168572126b94532
-  data.tar.gz: e51efd7f188fcc8bc29e846d50b6fceeb6a61ec00ccb1217807ed124303060290a6f6890980b25ab54adae11283896319af3ba0af63b6ca1f4d890a7bf2243e4
+  metadata.gz: 20ae390598e32c3276426c98dd437cd9d097e485d50bcb86757058defbbce42408a3a00b8c20fec9f86f2010078483bec4e1cd3053fc39ad42a462cf283e30e4
+  data.tar.gz: e200d3a053fba432f9819c68ceb3c48db5ad7d6daafc648f39ef5039b0fba345275c04f00215aff8c369fae2cf897e3eef31152bf93d375fdfa560b55b86564c

data/lib/digital_scriptorium/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module DigitalScriptorium
-  VERSION = '0.2.0'
+  VERSION = '0.2.1'
 end

metadata

@@ -1,56 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: digital_scriptorium
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.2.1
 platform: ruby
 authors:
 - Michael Holloway
 bindir: exe
 cert_chain: []
-date: 2025-01-17 00:00:00.000000000 Z
+date: 2025-01-18 00:00:00.000000000 Z
 dependencies:
-- !ruby/object:Gem::Dependency
-  name: multi_json
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.15'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.15'
-- !ruby/object:Gem::Dependency
-  name: representable
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.2'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.2'
-- !ruby/object:Gem::Dependency
-  name: tty-spinner
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '0.9'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '0.9'
 - !ruby/object:Gem::Dependency
   name: wikibase_representable
   requirement: !ruby/object:Gem::Requirement
@@ -161,7 +119,6 @@ files:
 - lib/digital_scriptorium/transformers/uniform_title_claim_transformer.rb
 - lib/digital_scriptorium/version.rb
 - sig/digital_scriptorium.rbs
-- wikibase_to_solr_new.rb
 homepage: https://github.com/mdholloway/digital_scriptorium
 licenses:
 - MIT

data/wikibase_to_solr_new.rb

@@ -1,116 +0,0 @@
-# frozen_string_literal: true
-
-require 'digital_scriptorium'
-require 'json'
-require 'logging'
-require 'optparse'
-require 'set'
-require 'time'
-require 'tty-spinner'
-require 'zlib'
-
-dir = File.dirname __FILE__
-
-input_file = File.expand_path 'wikibase_export.json.gz', dir
-output_file = File.expand_path 'solr_import.json', dir
-config_file = File.expand_path 'property_config.yml', dir
-pretty_print = false
-
-logger = Logging.logger($stdout)
-
-OptionParser.new { |opts|
-  opts.banner = 'Usage: wikibase_to_solr.rb [options]'
-
-  opts.on('-i', '--in FILE', 'The file path to the gzipped Wikibase JSON export file.') do |f|
-    input_file = File.expand_path f, dir
-  end
-
-  opts.on('-o', '--out FILE', 'The file path to output the formatted Solr JSON file.') do |f|
-    output_file = File.expand_path f, dir
-  end
-
-  opts.on('-c', '--config FILE', 'The file path to the property configuration file.') do |f|
-    config_file = File.expand_path f, dir
-  end
-
-  opts.on('-p', '--pretty-print', 'Whether to pretty-print the JSON output.') do
-    pretty_print = true
-  end
-}.parse!
-
-def merge(solr_item, new_props)
-  solr_item.merge(new_props) do |_, old_val, new_val|
-    old_val.nil? ? new_val : (old_val + new_val).uniq
-  end
-end
-
-def base_solr_item(meta)
-  ds_id = meta.manuscript.ds_id
-  {
-    'qid_meta' => [meta.holding.id, meta.manuscript.id, meta.record.id],
-    'id' => [ds_id],
-    'id_display' => [JSON.generate(recorded_value: ds_id)],
-    'id_search' => [ds_id]
-  }
-end
-
-def record?(entity)
-  entity.is_a?(DigitalScriptorium::DsItem) &&
-    entity.claims_by_property_id?(DigitalScriptorium::PropertyId::INSTANCE_OF) &&
-    entity.record?
-end
-
-start_time = Time.now.utc
-
-loading_spinner = TTY::Spinner.new('[:spinner] Loading export data', hide_cursor: true)
-loading_spinner.auto_spin
-
-export_json = Zlib::GzipReader.open(input_file).read
-export_hash = DigitalScriptorium::ExportRepresenter.new(DigitalScriptorium::Export.new)
-                                                   .from_json(export_json)
-                                                   .to_hash
-loaded_time = Time.now.utc
-loading_spinner.success("(#{format('%0.02f', loaded_time - start_time)}s)")
-
-item_count = 0
-generating_spinner = TTY::Spinner.new('[:spinner] Generating Solr documents', hide_cursor: true)
-generating_spinner.auto_spin
-
-File.open(output_file, 'w') do |file|
-  file << '['
-  file << "\n" if pretty_print
-
-  export_hash.each_with_index do |(_, entity), idx|
-    next unless record?(entity)
-
-    meta = DigitalScriptorium::DsMeta.new(entity, export_hash)
-    solr_item = base_solr_item(meta)
-
-    [meta.holding, meta.manuscript, meta.record].each do |item|
-      item.claims.each do |property_id, claims|
-        claims.each do |claim|
-          next unless DigitalScriptorium::Transformers.defined? property_id
-
-          begin
-            transformer = DigitalScriptorium::Transformers.create property_id, claim, export_hash
-            solr_item = merge solr_item, transformer.solr_props
-          rescue StandardError => e
-            logger.error "Error processing #{property_id} claim for item #{item.id}: #{e}"
-          end
-        end
-      end
-    end
-
-    file << (pretty_print ? JSON.pretty_generate(solr_item) : JSON.generate(solr_item))
-    file << ',' if idx < export_hash.size - 1
-    file << "\n" if pretty_print
-
-    item_count += 1
-  end
-
-  file << ']'
-end
-
-finish_time = Time.now.utc
-generating_spinner.success("(#{format('%0.02f', finish_time - loaded_time)}s)")
-puts "Generated #{item_count} Solr documents in #{format('%0.02f', finish_time - start_time)} seconds"