assembly-objectfile 1.8.2 → 1.10.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.github/ISSUE_TEMPLATE/bug_report.md +30 -0
- data/.github/pull_request_template.md +12 -0
- data/.rubocop.yml +121 -6
- data/.rubocop_todo.yml +85 -55
- data/.travis.yml +3 -3
- data/Gemfile +2 -0
- data/README.md +2 -1
- data/Rakefile +2 -3
- data/assembly-objectfile.gemspec +9 -3
- data/config/boot.rb +4 -2
- data/lib/assembly-objectfile.rb +4 -21
- data/lib/assembly-objectfile/content_metadata.rb +53 -176
- data/lib/assembly-objectfile/content_metadata/config.rb +24 -0
- data/lib/assembly-objectfile/content_metadata/file.rb +63 -0
- data/lib/assembly-objectfile/content_metadata/file_set.rb +73 -0
- data/lib/assembly-objectfile/content_metadata/file_set_builder.rb +65 -0
- data/lib/assembly-objectfile/content_metadata/nokogiri_builder.rb +55 -0
- data/lib/assembly-objectfile/object_file.rb +5 -3
- data/lib/assembly-objectfile/object_fileable.rb +43 -19
- data/lib/assembly-objectfile/version.rb +3 -1
- data/spec/content_metadata_spec.rb +734 -568
- data/spec/object_file_spec.rb +14 -0
- data/spec/spec_helper.rb +6 -4
- metadata +75 -13
data/assembly-objectfile.gemspec
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
$LOAD_PATH.push File.expand_path('lib', __dir__)
|
2
4
|
require 'assembly-objectfile/version'
|
3
5
|
|
@@ -12,14 +14,18 @@ Gem::Specification.new do |s|
|
|
12
14
|
s.description = 'Get exif data, file sizes and more.'
|
13
15
|
s.license = 'ALv2'
|
14
16
|
|
15
|
-
s.rubyforge_project = 'assembly-objectfile'
|
16
|
-
|
17
17
|
s.files = `git ls-files`.split("\n")
|
18
18
|
s.test_files = `git ls-files -- spec/*`.split("\n")
|
19
19
|
s.bindir = 'exe'
|
20
20
|
s.executables = s.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
21
21
|
s.require_paths = ['lib']
|
22
22
|
|
23
|
+
s.required_ruby_version = '~> 2.5'
|
24
|
+
|
25
|
+
s.add_dependency 'activesupport', '>= 5.2.0'
|
26
|
+
s.add_dependency 'deprecation'
|
27
|
+
s.add_dependency 'dry-struct', '~> 1.0'
|
28
|
+
s.add_dependency 'dry-types', '~> 1.1'
|
23
29
|
s.add_dependency 'mime-types', '> 3'
|
24
30
|
s.add_dependency 'mini_exiftool'
|
25
31
|
s.add_dependency 'nokogiri'
|
@@ -29,5 +35,5 @@ Gem::Specification.new do |s|
|
|
29
35
|
s.add_development_dependency 'rspec', '~> 3.0'
|
30
36
|
s.add_development_dependency 'rubocop'
|
31
37
|
s.add_development_dependency 'rubocop-rspec'
|
32
|
-
s.add_development_dependency 'simplecov'
|
38
|
+
s.add_development_dependency 'simplecov', '~> 0.17.0' # CodeClimate cannot use SimpleCov >= 0.18.0 for generating test coverage
|
33
39
|
end
|
data/config/boot.rb
CHANGED
@@ -1,9 +1,11 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require 'rubygems'
|
2
4
|
|
3
5
|
environment = ENV['ENVIRONMENT'] ||= 'development'
|
4
|
-
project_root = File.expand_path(File.dirname(__FILE__)
|
6
|
+
project_root = File.expand_path("#{File.dirname(__FILE__)}/..")
|
5
7
|
|
6
8
|
# Load config for current environment.
|
7
|
-
$LOAD_PATH.unshift(project_root
|
9
|
+
$LOAD_PATH.unshift("#{project_root}/lib")
|
8
10
|
|
9
11
|
require 'assembly-objectfile'
|
data/lib/assembly-objectfile.rb
CHANGED
@@ -1,36 +1,19 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
module Assembly
|
2
4
|
# the path to the gem, used to access profiles stored with the gem
|
3
|
-
PATH_TO_GEM = File.expand_path(File.dirname(__FILE__)
|
5
|
+
PATH_TO_GEM = File.expand_path("#{File.dirname(__FILE__)}/..")
|
4
6
|
|
5
7
|
# if input image is not one of these mime types, it will not be regarded as a valid image for the purpose of generating a JP2 derivative
|
6
8
|
VALID_IMAGE_MIMETYPES = ['image/jpeg', 'image/tiff', 'image/tif', 'image/png'].freeze
|
7
9
|
|
8
10
|
# if input file has one of these extensions in a 3D object, it will get the 3d resource type
|
9
|
-
VALID_THREE_DIMENSION_EXTENTIONS = ['.obj'
|
11
|
+
VALID_THREE_DIMENSION_EXTENTIONS = ['.obj'].freeze
|
10
12
|
|
11
13
|
# the list of mimetypes that will be "trusted" by the unix file command; if a mimetype other than one of these is returned
|
12
14
|
# by the file command, then a check will be made to see if exif data exists...if so, the mimetype returned by the exif data will be used
|
13
15
|
# if no exif data exists, then the mimetype returned by the unix file command will be used
|
14
16
|
TRUSTED_MIMETYPES = ['text/plain', 'plain/text', 'application/pdf', 'text/html', 'application/xml'].freeze
|
15
|
-
|
16
|
-
# default publish/preserve/shelve attributes used in content metadata
|
17
|
-
FILE_ATTRIBUTES = {}
|
18
|
-
# if no mimetype specific attributes are specified for a given file, define some defaults, and override for specific mimetypes below
|
19
|
-
FILE_ATTRIBUTES['default'] = { preserve: 'yes', shelve: 'no', publish: 'no' }
|
20
|
-
FILE_ATTRIBUTES['image/tif'] = { preserve: 'yes', shelve: 'no', publish: 'no' }
|
21
|
-
FILE_ATTRIBUTES['image/tiff'] = { preserve: 'yes', shelve: 'no', publish: 'no' }
|
22
|
-
FILE_ATTRIBUTES['image/jp2'] = { preserve: 'no', shelve: 'yes', publish: 'yes' }
|
23
|
-
FILE_ATTRIBUTES['image/jpeg'] = { preserve: 'yes', shelve: 'no', publish: 'no' }
|
24
|
-
FILE_ATTRIBUTES['audio/wav'] = { preserve: 'yes', shelve: 'no', publish: 'no' }
|
25
|
-
FILE_ATTRIBUTES['audio/x-wav'] = { preserve: 'yes', shelve: 'no', publish: 'no' }
|
26
|
-
FILE_ATTRIBUTES['audio/mp3'] = { preserve: 'no', shelve: 'yes', publish: 'yes' }
|
27
|
-
FILE_ATTRIBUTES['audio/mpeg'] = { preserve: 'no', shelve: 'yes', publish: 'yes' }
|
28
|
-
FILE_ATTRIBUTES['application/pdf'] = { preserve: 'yes', shelve: 'yes', publish: 'yes' }
|
29
|
-
FILE_ATTRIBUTES['plain/text'] = { preserve: 'yes', shelve: 'yes', publish: 'yes' }
|
30
|
-
FILE_ATTRIBUTES['text/plain'] = { preserve: 'yes', shelve: 'yes', publish: 'yes' }
|
31
|
-
FILE_ATTRIBUTES['image/png'] = { preserve: 'yes', shelve: 'yes', publish: 'no'}
|
32
|
-
FILE_ATTRIBUTES['application/zip'] = { preserve: 'yes', shelve: 'no', publish: 'no' }
|
33
|
-
FILE_ATTRIBUTES['application/json'] = { preserve: 'yes', shelve: 'yes', publish: 'yes' }
|
34
17
|
end
|
35
18
|
|
36
19
|
require 'assembly-objectfile/content_metadata'
|
@@ -1,10 +1,20 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require 'nokogiri'
|
4
|
+
require 'deprecation'
|
5
|
+
require 'active_support'
|
6
|
+
require 'assembly-objectfile/content_metadata/file'
|
7
|
+
require 'assembly-objectfile/content_metadata/file_set'
|
8
|
+
require 'assembly-objectfile/content_metadata/file_set_builder'
|
9
|
+
require 'assembly-objectfile/content_metadata/config'
|
10
|
+
require 'assembly-objectfile/content_metadata/nokogiri_builder'
|
2
11
|
|
3
12
|
module Assembly
|
4
13
|
SPECIAL_DPG_FOLDERS = %w[31 44 50].freeze # these special dpg folders will force any files contained in them into their own resources, regardless of filenaming convention
|
5
14
|
# these are used when :bundle=>:dpg only
|
6
15
|
|
7
16
|
DEPRECATED_STYLES = %i[book_with_pdf book_as_image].freeze
|
17
|
+
VALID_STYLES = %i[simple_image simple_book file map document 3d webarchive-seed].freeze
|
8
18
|
|
9
19
|
# This class generates content metadata for image files
|
10
20
|
class ContentMetadata
|
@@ -24,6 +34,7 @@ module Assembly
|
|
24
34
|
# :book_as_image, as simple_book, but with contentMetadata type="book", resource type="image" (same rule applies for resources with non images) - NOTE: THIS IS DEPRECATED
|
25
35
|
# :map, like simple_image, but with contentMetadata type="map", resource type="image"
|
26
36
|
# :3d, contentMetadata type="3d", ".obj" and other configured 3d extension files go into resource_type="3d", everything else into resource_type="file"
|
37
|
+
# :webarchive-seed, contentMetadata type="webarchive-seed", resource type="image"
|
27
38
|
# :bundle = optional - a symbol containing the method of bundling files into resources, allowed values are
|
28
39
|
# :default = all files get their own resources (default)
|
29
40
|
# :filename = files with the same filename but different extensions get bundled together in a single resource
|
@@ -43,195 +54,61 @@ module Assembly
|
|
43
54
|
# :auto_labels = optional - Will add automated resource labels (e.g. "File 1") when labels are not provided by the user. The default is true.
|
44
55
|
# Example:
|
45
56
|
# Assembly::ContentMetadata.create_content_metadata(:druid=>'druid:nx288wh8889',:style=>:simple_image,:objects=>object_files,:add_file_attributes=>false)
|
46
|
-
def self.create_content_metadata(
|
47
|
-
|
48
|
-
|
49
|
-
|
50
|
-
|
51
|
-
|
52
|
-
|
57
|
+
def self.create_content_metadata(druid:, objects:, auto_labels: true,
|
58
|
+
add_exif: false, bundle: :default, style: :simple_image,
|
59
|
+
add_file_attributes: false, file_attributes: {},
|
60
|
+
preserve_common_paths: false, flatten_folder_structure: false,
|
61
|
+
include_root_xml: nil)
|
62
|
+
|
63
|
+
common_path = find_common_path(objects) unless preserve_common_paths # find common paths to all files provided if needed
|
64
|
+
|
65
|
+
filesets = FileSetBuilder.build(bundle: bundle, objects: objects, style: style)
|
66
|
+
config = Config.new(auto_labels: auto_labels,
|
67
|
+
flatten_folder_structure: flatten_folder_structure,
|
68
|
+
add_file_attributes: add_file_attributes,
|
69
|
+
file_attributes: file_attributes,
|
70
|
+
add_exif: add_exif,
|
71
|
+
type: object_level_type(style))
|
72
|
+
|
73
|
+
builder = NokogiriBuilder.build(druid: druid,
|
74
|
+
filesets: filesets,
|
75
|
+
common_path: common_path,
|
76
|
+
config: config)
|
77
|
+
|
78
|
+
if include_root_xml == false
|
79
|
+
builder.doc.root.to_xml
|
80
|
+
else
|
81
|
+
builder.to_xml
|
82
|
+
end
|
83
|
+
end
|
53
84
|
|
54
|
-
|
55
|
-
|
56
|
-
|
57
|
-
auto_labels = (params[:auto_labels].nil? ? true : params[:auto_labels])
|
58
|
-
add_file_attributes = params[:add_file_attributes] || false
|
59
|
-
file_attributes = params[:file_attributes] || {}
|
60
|
-
preserve_common_paths = params[:preserve_common_paths] || false
|
61
|
-
flatten_folder_structure = params[:flatten_folder_structure] || false
|
62
|
-
include_root_xml = params[:include_root_xml]
|
85
|
+
def self.special_dpg_folder?(folder)
|
86
|
+
SPECIAL_DPG_FOLDERS.include?(folder)
|
87
|
+
end
|
63
88
|
|
64
|
-
|
65
|
-
objects.flatten.
|
89
|
+
def self.find_common_path(objects)
|
90
|
+
all_paths = objects.flatten.map do |obj|
|
66
91
|
raise "File '#{obj.path}' not found" unless obj.file_exists?
|
67
92
|
|
68
|
-
|
93
|
+
obj.path # collect all of the filenames into an array
|
69
94
|
end
|
70
95
|
|
71
|
-
|
72
|
-
|
73
|
-
|
74
|
-
content_type_descriptions = { file: 'file', image: 'image', book: 'book', map: 'map', '3d': '3d' }
|
75
|
-
resource_type_descriptions = { object: 'object', file: 'file', image: 'image', book: 'page', map: 'image', '3d': '3d' }
|
76
|
-
|
77
|
-
# global sequence for resource IDs
|
78
|
-
sequence = 0
|
96
|
+
Assembly::ObjectFile.common_path(all_paths) # find common paths to all files provided if needed
|
97
|
+
end
|
98
|
+
private_class_method :find_common_path
|
79
99
|
|
80
|
-
|
81
|
-
|
100
|
+
def self.object_level_type(style)
|
101
|
+
Deprecation.warn(self, "the style #{style} is now deprecated and should not be used. This will be removed in assembly-objectfile 2.0") if DEPRECATED_STYLES.include? style
|
102
|
+
raise "Supplied style (#{style}) not valid" unless (VALID_STYLES + DEPRECATED_STYLES).include? style
|
82
103
|
|
83
|
-
# set the object level content type id
|
84
104
|
case style
|
85
105
|
when :simple_image
|
86
|
-
|
87
|
-
when :file
|
88
|
-
content_type_description = content_type_descriptions[:file]
|
106
|
+
'image'
|
89
107
|
when :simple_book, :book_with_pdf, :book_as_image
|
90
|
-
|
91
|
-
when :map
|
92
|
-
content_type_description = content_type_descriptions[:map]
|
93
|
-
when :'3d'
|
94
|
-
content_type_description = content_type_descriptions[:'3d']
|
95
|
-
else
|
96
|
-
raise "Supplied style (#{style}) not valid"
|
97
|
-
end
|
98
|
-
|
99
|
-
puts "WARNING - the style #{style} is now deprecated and should not be used." if DEPRECATED_STYLES.include? style
|
100
|
-
|
101
|
-
# determine how many resources to create
|
102
|
-
# setup an array of arrays, where the first array is the number of resources, and the second array is the object files containined in that resource
|
103
|
-
case bundle
|
104
|
-
when :default # one resource per object
|
105
|
-
resources = objects.collect { |obj| [obj] }
|
106
|
-
when :filename # one resource per distinct filename (excluding extension)
|
107
|
-
# loop over distinct filenames, this determines how many resources we will have and
|
108
|
-
# create one resource node per distinct filename, collecting the relevant objects with the distinct filename into that resource
|
109
|
-
resources = []
|
110
|
-
distinct_filenames = objects.collect(&:filename_without_ext).uniq # find all the unique filenames in the set of objects, leaving off extensions and base paths
|
111
|
-
distinct_filenames.each { |distinct_filename| resources << objects.collect { |obj| obj if obj.filename_without_ext == distinct_filename }.compact }
|
112
|
-
when :dpg # group by DPG filename
|
113
|
-
# loop over distinct dpg base names, this determines how many resources we will have and
|
114
|
-
# create one resource node per distinct dpg base name, collecting the relevant objects with the distinct names into that resource
|
115
|
-
resources = []
|
116
|
-
distinct_filenames = objects.collect(&:dpg_basename).uniq # find all the unique DPG filenames in the set of objects
|
117
|
-
distinct_filenames.each do |distinct_filename|
|
118
|
-
resources << objects.collect { |obj| obj if obj.dpg_basename == distinct_filename && !is_special_dpg_folder?(obj.dpg_folder) }.compact
|
119
|
-
end
|
120
|
-
objects.each { |obj| resources << [obj] if is_special_dpg_folder?(obj.dpg_folder) } # certain subfolders require individual resources for files within them regardless of file-naming convention
|
121
|
-
when :prebundled
|
122
|
-
# if the user specifies this method, they will pass in an array of arrays, indicating resources, so we don't need to bundle in the gem
|
123
|
-
resources = objects
|
108
|
+
'book'
|
124
109
|
else
|
125
|
-
|
110
|
+
style.to_s
|
126
111
|
end
|
127
|
-
|
128
|
-
resources.delete([]) # delete any empty elements
|
129
|
-
|
130
|
-
builder = Nokogiri::XML::Builder.new do |xml|
|
131
|
-
xml.contentMetadata(objectId: druid.to_s, type: content_type_description) do
|
132
|
-
resources.each do |resource_files| # iterate over all the resources
|
133
|
-
# start a new resource element
|
134
|
-
sequence += 1
|
135
|
-
resource_id = "#{pid}_#{sequence}"
|
136
|
-
|
137
|
-
# grab all of the file types within a resource into an array so we can decide what the resource type should be
|
138
|
-
resource_file_types = resource_files.collect(&:object_type)
|
139
|
-
resource_has_non_images = !(resource_file_types - [:image]).empty?
|
140
|
-
resource_from_special_dpg_folder = resource_files.collect { |obj| is_special_dpg_folder?(obj.dpg_folder) }.uniq
|
141
|
-
|
142
|
-
if bundle == :dpg && resource_from_special_dpg_folder.include?(true) # objects in the special DPG folders are always type=object when we using :bundle=>:dpg
|
143
|
-
resource_type_description = resource_type_descriptions[:object]
|
144
|
-
else # otherwise look at the style to determine the resource_type_description
|
145
|
-
case style
|
146
|
-
when :simple_image
|
147
|
-
resource_type_description = resource_type_descriptions[:image]
|
148
|
-
when :file
|
149
|
-
resource_type_description = resource_type_descriptions[:file]
|
150
|
-
when :simple_book # in a simple book project, all resources are pages unless they are *all* non-images -- if so, switch the type to object
|
151
|
-
resource_type_description = resource_has_non_images && resource_file_types.include?(:image) == false ? resource_type_descriptions[:object] : resource_type_descriptions[:book]
|
152
|
-
when :book_as_image # same as simple book, but all resources are images instead of pages, unless we need to switch them to object type
|
153
|
-
resource_type_description = resource_has_non_images && resource_file_types.include?(:image) == false ? resource_type_descriptions[:object] : resource_type_descriptions[:image]
|
154
|
-
when :book_with_pdf # in book with PDF type, if we find a resource with *any* non images, switch it's type from book to object
|
155
|
-
resource_type_description = resource_has_non_images ? resource_type_descriptions[:object] : resource_type_descriptions[:book]
|
156
|
-
when :map
|
157
|
-
resource_type_description = resource_type_descriptions[:map]
|
158
|
-
when :'3d'
|
159
|
-
resource_extensions = resource_files.collect {|obj| obj.ext}
|
160
|
-
if (resource_extensions & VALID_THREE_DIMENSION_EXTENTIONS).empty? # if this resource contains no known 3D file extensions, the resource type is file
|
161
|
-
resource_type_description = resource_type_descriptions[:file]
|
162
|
-
else # otherwise the resource type is 3d
|
163
|
-
resource_type_description = resource_type_descriptions[:'3d']
|
164
|
-
end
|
165
|
-
end
|
166
|
-
end
|
167
|
-
|
168
|
-
resource_type_counters[resource_type_description.to_sym] += 1 # each resource type description gets its own incrementing counter
|
169
|
-
|
170
|
-
xml.resource(id: resource_id, sequence: sequence, type: resource_type_description) do
|
171
|
-
# create a generic resource label if needed
|
172
|
-
resource_label = (auto_labels == true ? "#{resource_type_description.capitalize} #{resource_type_counters[resource_type_description.to_sym]}" : '')
|
173
|
-
|
174
|
-
# but if one of the files has a label, use it instead
|
175
|
-
resource_files.each { |obj| resource_label = obj.label unless obj.label.nil? || obj.label.empty? }
|
176
|
-
|
177
|
-
xml.label(resource_label) unless resource_label.empty?
|
178
|
-
|
179
|
-
resource_files.each do |obj| # iterate over all the files in a resource
|
180
|
-
mimetype = obj.mimetype if add_file_attributes || add_exif # we only need to compute the mimetype if we are adding file attributes or exif info, otherwise skip it for performance reasons
|
181
|
-
|
182
|
-
# set file id attribute, first check the relative_path parameter on the object, and if it is set, just use that
|
183
|
-
if obj.relative_path
|
184
|
-
file_id = obj.relative_path
|
185
|
-
else
|
186
|
-
# if the relative_path attribute is not set, then use the path attribute and check to see if we need to remove the common part of the path
|
187
|
-
file_id = preserve_common_paths ? obj.path : obj.path.gsub(common_path, '')
|
188
|
-
file_id = File.basename(file_id) if flatten_folder_structure
|
189
|
-
end
|
190
|
-
|
191
|
-
xml_file_params = { id: file_id }
|
192
|
-
|
193
|
-
if add_file_attributes
|
194
|
-
file_attributes_hash = obj.file_attributes || file_attributes[mimetype] || file_attributes['default'] || Assembly::FILE_ATTRIBUTES[mimetype] || Assembly::FILE_ATTRIBUTES['default']
|
195
|
-
xml_file_params.merge!(
|
196
|
-
preserve: file_attributes_hash[:preserve],
|
197
|
-
publish: file_attributes_hash[:publish],
|
198
|
-
shelve: file_attributes_hash[:shelve],
|
199
|
-
role: file_attributes_hash[:role]
|
200
|
-
)
|
201
|
-
xml_file_params.reject! { |_k, v| v.nil? || v.empty? }
|
202
|
-
end
|
203
|
-
|
204
|
-
if add_exif
|
205
|
-
xml_file_params[:mimetype] = mimetype
|
206
|
-
xml_file_params[:size] = obj.filesize
|
207
|
-
end
|
208
|
-
xml.file(xml_file_params) do
|
209
|
-
if add_exif # add exif info if the user requested it
|
210
|
-
xml.checksum(obj.sha1, type: 'sha1')
|
211
|
-
xml.checksum(obj.md5, type: 'md5')
|
212
|
-
xml.imageData(height: obj.exif.imageheight, width: obj.exif.imagewidth) if obj.image? # add image data for an image
|
213
|
-
elsif obj.provider_md5 || obj.provider_sha1 # if we did not add exif info, see if there are user supplied checksums to add
|
214
|
-
xml.checksum(obj.provider_sha1, type: 'sha1') if obj.provider_sha1
|
215
|
-
xml.checksum(obj.provider_md5, type: 'md5') if obj.provider_md5
|
216
|
-
end # add_exif
|
217
|
-
end
|
218
|
-
end # end resource_files.each
|
219
|
-
end
|
220
|
-
end # resources.each
|
221
|
-
end
|
222
|
-
end
|
223
|
-
|
224
|
-
result = if include_root_xml == false
|
225
|
-
builder.doc.root.to_xml
|
226
|
-
else
|
227
|
-
builder.to_xml
|
228
|
-
end
|
229
|
-
|
230
|
-
result
|
231
|
-
end # create_content_metadata
|
232
|
-
|
233
|
-
def self.is_special_dpg_folder?(folder)
|
234
|
-
SPECIAL_DPG_FOLDERS.include?(folder)
|
235
112
|
end
|
236
113
|
end # class
|
237
114
|
end # module
|
@@ -0,0 +1,24 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require 'dry-struct'
|
4
|
+
require 'dry-types'
|
5
|
+
|
6
|
+
module Assembly
|
7
|
+
class ContentMetadata
|
8
|
+
# Types for the configuration
|
9
|
+
module Types
|
10
|
+
include Dry.Types()
|
11
|
+
end
|
12
|
+
|
13
|
+
# Represents a configuration for generating the content metadata
|
14
|
+
class Config < Dry::Struct
|
15
|
+
STYLES = %w[image file book map 3d document webarchive-seed].freeze
|
16
|
+
attribute :auto_labels, Types::Strict::Bool.default(true)
|
17
|
+
attribute :flatten_folder_structure, Types::Strict::Bool.default(false)
|
18
|
+
attribute :add_file_attributes, Types::Strict::Bool.default(false)
|
19
|
+
attribute :add_exif, Types::Strict::Bool.default(false)
|
20
|
+
attribute :file_attributes, Types::Strict::Hash.default({}.freeze)
|
21
|
+
attribute :type, Types::Strict::String.enum(*STYLES)
|
22
|
+
end
|
23
|
+
end
|
24
|
+
end
|
@@ -0,0 +1,63 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require 'active_support/core_ext/module/delegation'
|
4
|
+
|
5
|
+
module Assembly
|
6
|
+
class ContentMetadata
|
7
|
+
# Represents a single File
|
8
|
+
class File
|
9
|
+
# default publish/preserve/shelve attributes used in content metadata
|
10
|
+
# if no mimetype specific attributes are specified for a given file, define some defaults, and override for specific mimetypes below
|
11
|
+
ATTRIBUTES_FOR_TYPE = {
|
12
|
+
'default' => { preserve: 'yes', shelve: 'no', publish: 'no' },
|
13
|
+
'image/tif' => { preserve: 'yes', shelve: 'no', publish: 'no' },
|
14
|
+
'image/tiff' => { preserve: 'yes', shelve: 'no', publish: 'no' },
|
15
|
+
'image/jp2' => { preserve: 'no', shelve: 'yes', publish: 'yes' },
|
16
|
+
'image/jpeg' => { preserve: 'yes', shelve: 'no', publish: 'no' },
|
17
|
+
'audio/wav' => { preserve: 'yes', shelve: 'no', publish: 'no' },
|
18
|
+
'audio/x-wav' => { preserve: 'yes', shelve: 'no', publish: 'no' },
|
19
|
+
'audio/mp3' => { preserve: 'no', shelve: 'yes', publish: 'yes' },
|
20
|
+
'audio/mpeg' => { preserve: 'no', shelve: 'yes', publish: 'yes' },
|
21
|
+
'application/pdf' => { preserve: 'yes', shelve: 'yes', publish: 'yes' },
|
22
|
+
'plain/text' => { preserve: 'yes', shelve: 'yes', publish: 'yes' },
|
23
|
+
'text/plain' => { preserve: 'yes', shelve: 'yes', publish: 'yes' },
|
24
|
+
'image/png' => { preserve: 'yes', shelve: 'yes', publish: 'no' },
|
25
|
+
'application/zip' => { preserve: 'yes', shelve: 'no', publish: 'no' },
|
26
|
+
'application/json' => { preserve: 'yes', shelve: 'yes', publish: 'yes' }
|
27
|
+
}.freeze
|
28
|
+
|
29
|
+
# @param [Symbol] bundle
|
30
|
+
# @param [Assembly::ObjectFile] file
|
31
|
+
# @param style
|
32
|
+
def initialize(file:, bundle: nil, style: nil)
|
33
|
+
@bundle = bundle
|
34
|
+
@file = file
|
35
|
+
@style = style
|
36
|
+
end
|
37
|
+
|
38
|
+
delegate :sha1, :md5, :provider_md5, :provider_sha1, :mimetype, :filesize, :image?, to: :file
|
39
|
+
|
40
|
+
def file_id(common_path:, flatten_folder_structure:)
|
41
|
+
# set file id attribute, first check the relative_path parameter on the object, and if it is set, just use that
|
42
|
+
return file.relative_path if file.relative_path
|
43
|
+
|
44
|
+
# if the relative_path attribute is not set, then use the path attribute and check to see if we need to remove the common part of the path
|
45
|
+
file_id = common_path ? file.path.gsub(common_path, '') : file.path
|
46
|
+
file_id = ::File.basename(file_id) if flatten_folder_structure
|
47
|
+
file_id
|
48
|
+
end
|
49
|
+
|
50
|
+
def file_attributes(provided_file_attributes)
|
51
|
+
file.file_attributes || provided_file_attributes[mimetype] || provided_file_attributes['default'] || ATTRIBUTES_FOR_TYPE[mimetype] || ATTRIBUTES_FOR_TYPE['default']
|
52
|
+
end
|
53
|
+
|
54
|
+
def image_data
|
55
|
+
{ height: file.exif.imageheight, width: file.exif.imagewidth }
|
56
|
+
end
|
57
|
+
|
58
|
+
private
|
59
|
+
|
60
|
+
attr_reader :file
|
61
|
+
end
|
62
|
+
end
|
63
|
+
end
|
@@ -0,0 +1,73 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require 'active_support/core_ext/object/blank'
|
4
|
+
|
5
|
+
module Assembly
|
6
|
+
class ContentMetadata
|
7
|
+
# Represents a groups of related Files, such as a single master file and the derivatives
|
8
|
+
class FileSet
|
9
|
+
# @param [Boolean] dpg (false) is it a dpg bundle?
|
10
|
+
# @param [Array<Assembly::ObjectFile>] resource_files
|
11
|
+
# @param style
|
12
|
+
def initialize(resource_files:, style:, dpg: false)
|
13
|
+
@dpg = dpg
|
14
|
+
@resource_files = resource_files
|
15
|
+
@style = style
|
16
|
+
end
|
17
|
+
|
18
|
+
# objects in the special DPG folders are always type=object when we using :bundle=>:dpg
|
19
|
+
# otherwise look at the style to determine the resource_type_description
|
20
|
+
def resource_type_description
|
21
|
+
@resource_type_description ||= special_dpg_resource? ? 'object' : resource_type_descriptions
|
22
|
+
end
|
23
|
+
|
24
|
+
def label_from_file(default:)
|
25
|
+
resource_files.find { |obj| obj.label.present? }&.label || default
|
26
|
+
end
|
27
|
+
|
28
|
+
def files
|
29
|
+
resource_files.map { |file| File.new(file: file) }
|
30
|
+
end
|
31
|
+
|
32
|
+
private
|
33
|
+
|
34
|
+
attr_reader :dpg, :resource_files, :style
|
35
|
+
|
36
|
+
def special_dpg_resource?
|
37
|
+
return false unless dpg
|
38
|
+
|
39
|
+
resource_files.collect { |obj| ContentMetadata.special_dpg_folder?(obj.dpg_folder) }.include?(true)
|
40
|
+
end
|
41
|
+
|
42
|
+
# rubocop:disable Metrics/CyclomaticComplexity
|
43
|
+
def resource_type_descriptions
|
44
|
+
# grab all of the file types within a resource into an array so we can decide what the resource type should be
|
45
|
+
resource_file_types = resource_files.collect(&:object_type)
|
46
|
+
resource_has_non_images = !(resource_file_types - [:image]).empty?
|
47
|
+
|
48
|
+
case style
|
49
|
+
when :simple_image, :map, :'webarchive-seed'
|
50
|
+
'image'
|
51
|
+
when :file
|
52
|
+
'file'
|
53
|
+
when :simple_book # in a simple book project, all resources are pages unless they are *all* non-images -- if so, switch the type to object
|
54
|
+
resource_has_non_images && resource_file_types.include?(:image) == false ? 'object' : 'page'
|
55
|
+
when :book_as_image # same as simple book, but all resources are images instead of pages, unless we need to switch them to object type
|
56
|
+
resource_has_non_images && resource_file_types.include?(:image) == false ? 'object' : 'image'
|
57
|
+
when :book_with_pdf # in book with PDF type, if we find a resource with *any* non images, switch it's type from book to object
|
58
|
+
resource_has_non_images ? 'object' : 'page'
|
59
|
+
when :document
|
60
|
+
'document'
|
61
|
+
when :'3d'
|
62
|
+
resource_extensions = resource_files.collect(&:ext)
|
63
|
+
if (resource_extensions & VALID_THREE_DIMENSION_EXTENTIONS).empty? # if this resource contains no known 3D file extensions, the resource type is file
|
64
|
+
'file'
|
65
|
+
else # otherwise the resource type is 3d
|
66
|
+
'3d'
|
67
|
+
end
|
68
|
+
end
|
69
|
+
end
|
70
|
+
# rubocop:enable Metrics/CyclomaticComplexity
|
71
|
+
end
|
72
|
+
end
|
73
|
+
end
|