ro-crate 0.4.9 → 0.4.13
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/Gemfile.lock +13 -13
- data/README.md +39 -0
- data/lib/ro_crate/model/crate.rb +48 -6
- data/lib/ro_crate/model/data_entity.rb +5 -4
- data/lib/ro_crate/model/directory.rb +2 -4
- data/lib/ro_crate/model/entity.rb +41 -4
- data/lib/ro_crate/model/entry.rb +2 -2
- data/lib/ro_crate/model/file.rb +2 -4
- data/lib/ro_crate/model/metadata.rb +9 -1
- data/lib/ro_crate/model/organization.rb +1 -1
- data/lib/ro_crate/model/preview.rb +3 -15
- data/lib/ro_crate/model/preview_generator.rb +40 -0
- data/lib/ro_crate/model/remote_entry.rb +1 -12
- data/lib/ro_crate/reader.rb +76 -19
- data/lib/ro_crate/writer.rb +4 -4
- data/lib/ro_crate.rb +2 -1
- data/ro_crate.gemspec +3 -3
- data/test/crate_test.rb +58 -3
- data/test/directory_test.rb +21 -21
- data/test/entity_test.rb +114 -0
- data/test/fixtures/biobb_hpc_workflows-condapack.zip +0 -0
- data/test/fixtures/conflicting_data_directory/info.txt +1 -0
- data/test/fixtures/conflicting_data_directory/nested.txt +1 -0
- data/test/fixtures/nested_directory.zip +0 -0
- data/test/fixtures/ro-crate-galaxy-sortchangecase/LICENSE +176 -0
- data/test/fixtures/ro-crate-galaxy-sortchangecase/README.md +6 -0
- data/test/fixtures/ro-crate-galaxy-sortchangecase/ro-crate-metadata.json +133 -0
- data/test/fixtures/ro-crate-galaxy-sortchangecase/sort-and-change-case.ga +118 -0
- data/test/fixtures/ro-crate-galaxy-sortchangecase/test/test1/input.bed +3 -0
- data/test/fixtures/ro-crate-galaxy-sortchangecase/test/test1/output_exp.bed +3 -0
- data/test/fixtures/ro-crate-galaxy-sortchangecase/test/test1/sort-and-change-case-test.yml +8 -0
- data/test/fixtures/sparse_directory_crate/ro-crate-preview.html +60 -59
- data/test/reader_test.rb +83 -40
- data/test/test_helper.rb +5 -1
- data/test/writer_test.rb +59 -2
- metadata +26 -8
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 364757350da7d275b02def348bfd97782ad0ae31728c44f0691bc084e4d94ac9
|
4
|
+
data.tar.gz: 6b56ae58b682a4618dcb92ca78cbae720426316bc8ec50b9377fb9fda8ec9137
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: a2b14f9c64528476a18f84d0292311484e975a3dc6085c00c94f5dac8d84fa144e91bb6138c876db4a4079aeec4222eb6a1f91458e2ef0260defa9d3a0a42d57
|
7
|
+
data.tar.gz: b56d257868b7cb94f341e844223d14988c280fff34bb54dd1a676d47681d96afc8193ebb60a22eecfef7e96dc0d2ce03ed1571190a1717fbe56d1b80ed579053
|
data/Gemfile.lock
CHANGED
@@ -1,31 +1,31 @@
|
|
1
1
|
PATH
|
2
2
|
remote: .
|
3
3
|
specs:
|
4
|
-
ro-crate (0.4.
|
5
|
-
addressable (
|
4
|
+
ro-crate (0.4.13)
|
5
|
+
addressable (>= 2.7, < 2.9)
|
6
6
|
rubyzip (~> 2.0.0)
|
7
7
|
|
8
8
|
GEM
|
9
9
|
remote: https://rubygems.org/
|
10
10
|
specs:
|
11
|
-
addressable (2.
|
11
|
+
addressable (2.8.0)
|
12
12
|
public_suffix (>= 2.0.2, < 5.0)
|
13
13
|
crack (0.4.3)
|
14
14
|
safe_yaml (~> 1.0.0)
|
15
|
-
docile (1.3.
|
15
|
+
docile (1.3.5)
|
16
16
|
hashdiff (1.0.1)
|
17
|
-
|
18
|
-
|
19
|
-
public_suffix (4.0.3)
|
17
|
+
power_assert (1.1.3)
|
18
|
+
public_suffix (4.0.6)
|
20
19
|
rake (13.0.0)
|
21
20
|
rubyzip (2.0.0)
|
22
21
|
safe_yaml (1.0.5)
|
23
|
-
simplecov (0.
|
22
|
+
simplecov (0.21.2)
|
24
23
|
docile (~> 1.1)
|
25
|
-
|
26
|
-
|
27
|
-
simplecov-html (0.
|
28
|
-
|
24
|
+
simplecov-html (~> 0.11)
|
25
|
+
simplecov_json_formatter (~> 0.1)
|
26
|
+
simplecov-html (0.12.3)
|
27
|
+
simplecov_json_formatter (0.1.2)
|
28
|
+
test-unit (3.2.9)
|
29
29
|
power_assert
|
30
30
|
webmock (3.8.3)
|
31
31
|
addressable (>= 2.3.6)
|
@@ -39,7 +39,7 @@ PLATFORMS
|
|
39
39
|
DEPENDENCIES
|
40
40
|
rake (~> 13.0.0)
|
41
41
|
ro-crate!
|
42
|
-
simplecov (~> 0.
|
42
|
+
simplecov (~> 0.21.2)
|
43
43
|
test-unit (~> 3.2.3)
|
44
44
|
webmock (~> 3.8.3)
|
45
45
|
yard (~> 0.9.25)
|
data/README.md
CHANGED
@@ -1,5 +1,7 @@
|
|
1
1
|
# ro-crate-ruby
|
2
2
|
|
3
|
+
![Tests](https://github.com/ResearchObject/ro-crate-ruby/actions/workflows/tests.yml/badge.svg)
|
4
|
+
|
3
5
|
This is a WIP gem for creating, manipulating and reading RO-Crates (conforming to version 1.1 of the specification).
|
4
6
|
|
5
7
|
* RO-Crate - https://researchobject.github.io/ro-crate/
|
@@ -17,6 +19,43 @@ and run `bundle install`.
|
|
17
19
|
|
18
20
|
## Usage
|
19
21
|
|
22
|
+
This gem consists a hierarchy of classes to model RO-Crate "entities": the crate itself, data entities
|
23
|
+
(files and directory) and contextual entities (with a limited set of specializations, such as `ROCrate::Person`).
|
24
|
+
They are all descendents of the `ROCrate::Entity` class, with the `ROCrate::Crate` class representing the crate itself.
|
25
|
+
|
26
|
+
The `ROCrate::Reader` class handles reading of RO-Crates into the above model, from a Zip file or directory.
|
27
|
+
|
28
|
+
The `ROCrate::Writer` class can write out an `ROCrate::Crate` instance into a Zip file or directory.
|
29
|
+
|
30
|
+
**Note:** for performance reasons, the gem is currently not linked-data aware and will allow you to set properties that
|
31
|
+
are not semantically valid.
|
32
|
+
|
33
|
+
### Entities
|
34
|
+
Entities correspond to entries in the `@graph` of the RO-Crate's metadata JSON-LD file. Each entity class is
|
35
|
+
basically a wrapper around a set of JSON properties, with some convenience methods for getting/setting some
|
36
|
+
commonly used properties (`crate.name = "My first crate"`).
|
37
|
+
|
38
|
+
These convenience getter/setter methods will automatically handle turning objects into references and adding them to the
|
39
|
+
`@graph` if necessary.
|
40
|
+
|
41
|
+
##### Getting/Setting Arbitrary Properties of Entities
|
42
|
+
As well as using the pre-defined getter/setter methods, you can get/set arbitrary properties like so.
|
43
|
+
|
44
|
+
To set the "creativeWorkStatus" property of the RO-Crate itself to a string literal:
|
45
|
+
```ruby
|
46
|
+
crate['creativeWorkStatus'] = 'work-in-progress'
|
47
|
+
```
|
48
|
+
|
49
|
+
If you want to reference other entities in the crate, you can get a JSON-LD reference from an entity object by using the `reference` method:
|
50
|
+
```ruby
|
51
|
+
joe = crate.add_person('joe', { name: 'Joe Bloggs' }) # Add the entity to the @graph
|
52
|
+
crate['copyrightHolder'] = joe.reference # Reference the entity from the "copyrightHolder" property
|
53
|
+
```
|
54
|
+
and to resolve those references back to the object, use the `dereference` method:
|
55
|
+
```ruby
|
56
|
+
joe = crate['copyrightHolder'].dereference
|
57
|
+
```
|
58
|
+
|
20
59
|
### Documentation
|
21
60
|
|
22
61
|
[Click here for API documentation](https://www.researchobject.org/ro-crate-ruby/).
|
data/lib/ro_crate/model/crate.rb
CHANGED
@@ -25,6 +25,15 @@ module ROCrate
|
|
25
25
|
super(self, nil, id, properties)
|
26
26
|
end
|
27
27
|
|
28
|
+
##
|
29
|
+
# Lookup an Entity using the given ID (in this Entity's crate).
|
30
|
+
#
|
31
|
+
# @param id [String] The ID to query.
|
32
|
+
# @return [Entity, nil]
|
33
|
+
def dereference(id)
|
34
|
+
entities.detect { |e| e.canonical_id == crate.resolve_id(id) } if id
|
35
|
+
end
|
36
|
+
|
28
37
|
##
|
29
38
|
# Create a new file and add it to the crate.
|
30
39
|
#
|
@@ -168,6 +177,15 @@ module ROCrate
|
|
168
177
|
@preview ||= ROCrate::Preview.new(self)
|
169
178
|
end
|
170
179
|
|
180
|
+
##
|
181
|
+
# Set the RO-Crate preview file
|
182
|
+
# @param preview [Preview] the preview to set.
|
183
|
+
#
|
184
|
+
# @return [Preview]
|
185
|
+
def preview=(preview)
|
186
|
+
@preview = claim(preview)
|
187
|
+
end
|
188
|
+
|
171
189
|
##
|
172
190
|
# All the entities within the crate. Includes contextual entities, data entities, the crate itself and its metadata file.
|
173
191
|
#
|
@@ -220,32 +238,56 @@ module ROCrate
|
|
220
238
|
entity.class.new(crate, entity.id, entity.raw_properties)
|
221
239
|
end
|
222
240
|
|
223
|
-
alias_method :
|
241
|
+
alias_method :own_payload, :payload
|
224
242
|
##
|
225
|
-
#
|
226
|
-
# key is the
|
243
|
+
# The file payload of the RO-Crate - a map of all the files/directories contained in the RO-Crate, where the
|
244
|
+
# key is the path relative to the crate's root, and the value is an Entry where the source data can be read.
|
227
245
|
#
|
228
246
|
# @return [Hash{String => Entry}>]
|
229
|
-
def
|
247
|
+
def payload
|
230
248
|
# Gather a map of entries, starting from the crate itself, then any directory data entities, then finally any
|
231
249
|
# file data entities. This ensures in the case of a conflict, the more "specific" data entities take priority.
|
232
|
-
entries =
|
250
|
+
entries = own_payload
|
233
251
|
non_self_entities = default_entities.reject { |e| e == self }
|
234
252
|
sorted_entities = (non_self_entities | data_entities).sort_by { |e| e.is_a?(ROCrate::Directory) ? 0 : 1 }
|
235
253
|
|
236
254
|
sorted_entities.each do |entity|
|
237
|
-
entity.
|
255
|
+
entity.payload.each do |path, entry|
|
238
256
|
entries[path] = entry
|
239
257
|
end
|
240
258
|
end
|
241
259
|
|
242
260
|
entries
|
243
261
|
end
|
262
|
+
alias_method :entries, :payload
|
244
263
|
|
245
264
|
def get_binding
|
246
265
|
binding
|
247
266
|
end
|
248
267
|
|
268
|
+
##
|
269
|
+
# Remove the entity from the RO-Crate.
|
270
|
+
#
|
271
|
+
# @param entity [Entity, String] The entity or ID of an entity to remove from the crate.
|
272
|
+
# @param remove_orphaned [Boolean] Should linked contextual entities also be removed from the crate they are left
|
273
|
+
# dangling (nothing else is linked to them)?
|
274
|
+
#
|
275
|
+
# @return [Entity, nil] The entity that was deleted, or nil if nothing was deleted.
|
276
|
+
def delete(entity, remove_orphaned: true)
|
277
|
+
entity = dereference(entity) if entity.is_a?(String)
|
278
|
+
return unless entity
|
279
|
+
|
280
|
+
deleted = data_entities.delete(entity) || contextual_entities.delete(entity)
|
281
|
+
|
282
|
+
if deleted && remove_orphaned
|
283
|
+
crate_entities = crate.linked_entities(deep: true)
|
284
|
+
to_remove = (entity.linked_entities(deep: true) - crate_entities)
|
285
|
+
to_remove.each(&:delete)
|
286
|
+
end
|
287
|
+
|
288
|
+
deleted
|
289
|
+
end
|
290
|
+
|
249
291
|
private
|
250
292
|
|
251
293
|
def full_entry_path(relative_path)
|
@@ -3,6 +3,8 @@ module ROCrate
|
|
3
3
|
# A class to represent a "Data Entity" within an RO-Crate.
|
4
4
|
# Data Entities are the actual physical files and directories within the Crate.
|
5
5
|
class DataEntity < Entity
|
6
|
+
properties(%w[name contentSize dateModified encodingFormat identifier sameAs author])
|
7
|
+
|
6
8
|
def self.format_local_id(id)
|
7
9
|
super.chomp('/')
|
8
10
|
end
|
@@ -13,8 +15,6 @@ module ROCrate
|
|
13
15
|
# @return [Class]
|
14
16
|
def self.specialize(props)
|
15
17
|
type = props['@type']
|
16
|
-
id = props['@id']
|
17
|
-
abs = URI(id)&.absolute? rescue false
|
18
18
|
type = [type] unless type.is_a?(Array)
|
19
19
|
if type.include?('Dataset')
|
20
20
|
ROCrate::Directory
|
@@ -24,12 +24,13 @@ module ROCrate
|
|
24
24
|
end
|
25
25
|
|
26
26
|
##
|
27
|
-
#
|
27
|
+
# The payload of all the files/directories associated with this DataEntity, mapped by their relative file path.
|
28
28
|
#
|
29
29
|
# @return [Hash{String => Entry}>] The key is the location within the crate, and the value is an Entry.
|
30
|
-
def
|
30
|
+
def payload
|
31
31
|
{}
|
32
32
|
end
|
33
|
+
alias_method :entries, :payload
|
33
34
|
|
34
35
|
##
|
35
36
|
# A disk-safe filepath based on the ID of this DataEntity.
|
@@ -2,8 +2,6 @@ module ROCrate
|
|
2
2
|
##
|
3
3
|
# A data entity that represents a directory of potentially many files and subdirectories (or none).
|
4
4
|
class Directory < DataEntity
|
5
|
-
properties(%w[name contentSize dateModified encodingFormat identifier sameAs])
|
6
|
-
|
7
5
|
def self.format_local_id(id)
|
8
6
|
super + '/'
|
9
7
|
end
|
@@ -30,11 +28,11 @@ module ROCrate
|
|
30
28
|
end
|
31
29
|
|
32
30
|
##
|
33
|
-
# The
|
31
|
+
# The payload of this directory - a map of all the files/directories, where the key is the destination path
|
34
32
|
# within the crate and the value is an Entry where the source data can be read.
|
35
33
|
#
|
36
34
|
# @return [Hash{String => Entry}>]
|
37
|
-
def
|
35
|
+
def payload
|
38
36
|
entries = {}
|
39
37
|
entries[filepath.chomp('/')] = @entry if @entry
|
40
38
|
|
@@ -129,16 +129,26 @@ module ROCrate
|
|
129
129
|
# @param id [String] The ID to query.
|
130
130
|
# @return [Entity, nil]
|
131
131
|
def dereference(id)
|
132
|
-
crate.
|
132
|
+
crate.dereference(id)
|
133
133
|
end
|
134
|
-
|
135
134
|
alias_method :get, :dereference
|
136
135
|
|
136
|
+
##
|
137
|
+
# Remove this entity from the RO-Crate.
|
138
|
+
#
|
139
|
+
# @param remove_orphaned [Boolean] Should linked contextual entities also be removed from the crate (if nothing else is linked to them)?
|
140
|
+
#
|
141
|
+
# @return [Entity, nil] This entity, or nil if nothing was deleted.
|
142
|
+
def delete(remove_orphaned: true)
|
143
|
+
crate.delete(self, remove_orphaned: remove_orphaned)
|
144
|
+
end
|
145
|
+
|
137
146
|
def id
|
138
147
|
@properties['@id']
|
139
148
|
end
|
140
149
|
|
141
150
|
def id=(id)
|
151
|
+
@canonical_id = nil
|
142
152
|
@properties['@id'] = self.class.format_id(id)
|
143
153
|
end
|
144
154
|
|
@@ -190,13 +200,13 @@ module ROCrate
|
|
190
200
|
#
|
191
201
|
# @return [Addressable::URI]
|
192
202
|
def canonical_id
|
193
|
-
crate.resolve_id(id)
|
203
|
+
@canonical_id ||= crate.resolve_id(id)
|
194
204
|
end
|
195
205
|
|
196
206
|
##
|
197
207
|
# Is this entity local to the crate or an external reference?
|
198
208
|
#
|
199
|
-
# @return [
|
209
|
+
# @return [Boolean]
|
200
210
|
def external?
|
201
211
|
crate.canonical_id.host != canonical_id.host
|
202
212
|
end
|
@@ -226,6 +236,33 @@ module ROCrate
|
|
226
236
|
@properties.has_type?(type)
|
227
237
|
end
|
228
238
|
|
239
|
+
##
|
240
|
+
# Gather a list of entities linked to this one through its properties.
|
241
|
+
# @param deep [Boolean] If false, only consider direct links, otherwise consider transitive links.
|
242
|
+
# @param linked [Hash{String => Entity}] Discovered entities, mapped by their ID, to avoid loops when recursing.
|
243
|
+
# @return [Array<Entity>]
|
244
|
+
def linked_entities(deep: false, linked: {})
|
245
|
+
properties.each do |key, value|
|
246
|
+
value = [value] if value.is_a?(JSONLDHash)
|
247
|
+
|
248
|
+
if value.is_a?(Array)
|
249
|
+
value.each do |v|
|
250
|
+
if v.is_a?(JSONLDHash) && !linked.key?(v['@id'])
|
251
|
+
entity = v.dereference
|
252
|
+
linked[entity.id] = entity if entity
|
253
|
+
if deep
|
254
|
+
entity.linked_entities(deep: true, linked: linked).each do |e|
|
255
|
+
linked[e.id] = e
|
256
|
+
end
|
257
|
+
end
|
258
|
+
end
|
259
|
+
end
|
260
|
+
end
|
261
|
+
end
|
262
|
+
|
263
|
+
linked.values.compact
|
264
|
+
end
|
265
|
+
|
229
266
|
private
|
230
267
|
|
231
268
|
def default_properties
|
data/lib/ro_crate/model/entry.rb
CHANGED
@@ -14,10 +14,10 @@ module ROCrate
|
|
14
14
|
end
|
15
15
|
|
16
16
|
##
|
17
|
-
# Write the source to the destination via a buffer.
|
17
|
+
# Write the entry's source to the destination via a buffer.
|
18
18
|
#
|
19
19
|
# @param dest [#write] An IO-like destination to write to.
|
20
|
-
def
|
20
|
+
def write_to(dest)
|
21
21
|
input = source
|
22
22
|
input = input.open('rb') if input.is_a?(Pathname)
|
23
23
|
while (buff = input.read(4096))
|
data/lib/ro_crate/model/file.rb
CHANGED
@@ -2,14 +2,12 @@ module ROCrate
|
|
2
2
|
##
|
3
3
|
# A data entity that represents a single file.
|
4
4
|
class File < DataEntity
|
5
|
-
properties(%w[name contentSize dateModified encodingFormat identifier sameAs])
|
6
|
-
|
7
5
|
##
|
8
6
|
# Create a new ROCrate::File. PLEASE NOTE, the new file will not be added to the crate. To do this, call
|
9
7
|
# Crate#add_data_entity, or just use Crate#add_file.
|
10
8
|
#
|
11
9
|
# @param crate [Crate] The RO-Crate that owns this file.
|
12
|
-
# @param source [String, Pathname, ::File,
|
10
|
+
# @param source [String, Pathname, ::File, URI, nil, #read] The source on the disk (or on the internet if a URI) where this file will be read.
|
13
11
|
# @param crate_path [String] The relative path within the RO-Crate where this file will be written.
|
14
12
|
# @param properties [Hash{String => Object}] A hash of JSON-LD properties to associate with this file.
|
15
13
|
def initialize(crate, source, crate_path = nil, properties = {})
|
@@ -58,7 +56,7 @@ module ROCrate
|
|
58
56
|
# (for compatibility with Directory#entries)
|
59
57
|
#
|
60
58
|
# @return [Hash{String => Entry}>] The key is the location within the crate, and the value is an Entry.
|
61
|
-
def
|
59
|
+
def payload
|
62
60
|
remote? ? {} : { filepath => source }
|
63
61
|
end
|
64
62
|
|
@@ -17,7 +17,15 @@ module ROCrate
|
|
17
17
|
# @return [String] The rendered JSON-LD as a "prettified" string.
|
18
18
|
def generate
|
19
19
|
graph = crate.entities.map(&:properties).reject(&:empty?)
|
20
|
-
JSON.pretty_generate('@context' =>
|
20
|
+
JSON.pretty_generate('@context' => context, '@graph' => graph)
|
21
|
+
end
|
22
|
+
|
23
|
+
def context
|
24
|
+
@context || CONTEXT
|
25
|
+
end
|
26
|
+
|
27
|
+
def context= c
|
28
|
+
@context = c
|
21
29
|
end
|
22
30
|
|
23
31
|
private
|
@@ -12,26 +12,14 @@ module ROCrate
|
|
12
12
|
# @return [String]
|
13
13
|
attr_accessor :template
|
14
14
|
|
15
|
-
def initialize(crate, properties = {})
|
15
|
+
def initialize(crate, source = nil, properties = {})
|
16
|
+
source ||= PreviewGenerator.new(self)
|
16
17
|
@template = nil
|
17
|
-
super(crate,
|
18
|
-
end
|
19
|
-
|
20
|
-
##
|
21
|
-
# Generate the crate's `ro-crate-preview.html`.
|
22
|
-
# @return [String] The rendered HTML as a string.
|
23
|
-
def generate
|
24
|
-
b = crate.get_binding
|
25
|
-
renderer = ERB.new(template || ::File.read(DEFAULT_TEMPLATE))
|
26
|
-
renderer.result(b)
|
18
|
+
super(crate, source, IDENTIFIER, properties)
|
27
19
|
end
|
28
20
|
|
29
21
|
private
|
30
22
|
|
31
|
-
def source
|
32
|
-
Entry.new(StringIO.new(generate))
|
33
|
-
end
|
34
|
-
|
35
23
|
def default_properties
|
36
24
|
{
|
37
25
|
'@id' => IDENTIFIER,
|
@@ -0,0 +1,40 @@
|
|
1
|
+
require 'erb'
|
2
|
+
|
3
|
+
module ROCrate
|
4
|
+
##
|
5
|
+
# A class to handle generation of an RO-Crate's preview HTML in an IO-like way (to fit into an Entry).
|
6
|
+
class PreviewGenerator
|
7
|
+
##
|
8
|
+
# @param preview [Preview] The RO-Crate preview object.
|
9
|
+
def initialize(preview)
|
10
|
+
@preview = preview
|
11
|
+
end
|
12
|
+
|
13
|
+
def read(*args)
|
14
|
+
io.read(*args)
|
15
|
+
end
|
16
|
+
|
17
|
+
##
|
18
|
+
# Generate the crate's `ro-crate-preview.html`.
|
19
|
+
# @return [String] The rendered HTML as a string.
|
20
|
+
def generate
|
21
|
+
b = crate.get_binding
|
22
|
+
renderer = ERB.new(template)
|
23
|
+
renderer.result(b)
|
24
|
+
end
|
25
|
+
|
26
|
+
def template
|
27
|
+
@preview.template || ::File.read(Preview::DEFAULT_TEMPLATE)
|
28
|
+
end
|
29
|
+
|
30
|
+
def crate
|
31
|
+
@preview.crate
|
32
|
+
end
|
33
|
+
|
34
|
+
private
|
35
|
+
|
36
|
+
def io
|
37
|
+
@io ||= StringIO.new(generate)
|
38
|
+
end
|
39
|
+
end
|
40
|
+
end
|
@@ -2,7 +2,7 @@ module ROCrate
|
|
2
2
|
##
|
3
3
|
# A class to represent a reference within an RO-Crate, to a remote file held on the internet somewhere.
|
4
4
|
# It handles the actual reading/writing of bytes.
|
5
|
-
class RemoteEntry
|
5
|
+
class RemoteEntry < Entry
|
6
6
|
attr_reader :uri
|
7
7
|
|
8
8
|
##
|
@@ -13,17 +13,6 @@ module ROCrate
|
|
13
13
|
@uri = uri
|
14
14
|
end
|
15
15
|
|
16
|
-
def write(dest)
|
17
|
-
raise 'Cannot write to a remote entry!'
|
18
|
-
end
|
19
|
-
|
20
|
-
##
|
21
|
-
# Read from the source.
|
22
|
-
#
|
23
|
-
def read
|
24
|
-
source.read
|
25
|
-
end
|
26
|
-
|
27
16
|
##
|
28
17
|
# @return [IO] An IO object for the remote resource.
|
29
18
|
#
|
data/lib/ro_crate/reader.rb
CHANGED
@@ -84,7 +84,10 @@ module ROCrate
|
|
84
84
|
def self.read_zip(source, target_dir: Dir.mktmpdir)
|
85
85
|
unzip_to(source, target_dir)
|
86
86
|
|
87
|
-
|
87
|
+
# Traverse the unzipped directory to try and find the crate's root
|
88
|
+
root_dir = detect_root_directory(target_dir)
|
89
|
+
|
90
|
+
read_directory(root_dir)
|
88
91
|
end
|
89
92
|
|
90
93
|
##
|
@@ -100,8 +103,12 @@ module ROCrate
|
|
100
103
|
entry == ROCrate::Metadata::IDENTIFIER_1_0 }
|
101
104
|
|
102
105
|
if metadata_file
|
103
|
-
|
104
|
-
|
106
|
+
metadata_json = ::File.read(::File.join(source, metadata_file))
|
107
|
+
metadata = JSON.parse(metadata_json)
|
108
|
+
entities = entities_from_metadata(metadata)
|
109
|
+
context = metadata['@context']
|
110
|
+
|
111
|
+
build_crate(entities, source, context: context)
|
105
112
|
else
|
106
113
|
raise 'No metadata found!'
|
107
114
|
end
|
@@ -110,10 +117,9 @@ module ROCrate
|
|
110
117
|
##
|
111
118
|
# Extracts all the entities from the @graph of the RO-Crate Metadata.
|
112
119
|
#
|
113
|
-
# @param
|
120
|
+
# @param metadata [Hash] A Hash containing the parsed metadata JSON.
|
114
121
|
# @return [Hash{String => Hash}] A Hash of all the entities, mapped by their @id.
|
115
|
-
def self.entities_from_metadata(
|
116
|
-
metadata = JSON.parse(metadata_json)
|
122
|
+
def self.entities_from_metadata(metadata)
|
117
123
|
graph = metadata['@graph']
|
118
124
|
|
119
125
|
if graph
|
@@ -126,6 +132,7 @@ module ROCrate
|
|
126
132
|
# Do some normalization...
|
127
133
|
entities[ROCrate::Metadata::IDENTIFIER] = extract_metadata_entity(entities)
|
128
134
|
raise "No metadata entity found in @graph!" unless entities[ROCrate::Metadata::IDENTIFIER]
|
135
|
+
entities[ROCrate::Preview::IDENTIFIER] = extract_preview_entity(entities)
|
129
136
|
entities[ROCrate::Crate::IDENTIFIER] = extract_root_entity(entities)
|
130
137
|
raise "No root entity (with @id: #{entities[ROCrate::Metadata::IDENTIFIER].dig('about', '@id')}) found in @graph!" unless entities[ROCrate::Crate::IDENTIFIER]
|
131
138
|
|
@@ -136,25 +143,50 @@ module ROCrate
|
|
136
143
|
end
|
137
144
|
|
138
145
|
##
|
139
|
-
# Create
|
146
|
+
# Create and populate crate from the given set of entities.
|
140
147
|
#
|
141
148
|
# @param entity_hash [Hash{String => Hash}] A Hash containing all the entities in the @graph, mapped by their @id.
|
142
149
|
# @param source [String, ::File, Pathname] The location of the RO-Crate being read.
|
150
|
+
# @param crate_class [Class] The class to use to instantiate the crate,
|
151
|
+
# useful if you have created a subclass of ROCrate::Crate that you want to use. (defaults to ROCrate::Crate).
|
152
|
+
# @param context [nil, String, Array, Hash] A custom JSON-LD @context (parsed), or nil to use default.
|
143
153
|
# @return [Crate] The RO-Crate.
|
144
|
-
def self.build_crate(entity_hash, source)
|
145
|
-
|
154
|
+
def self.build_crate(entity_hash, source, crate_class: ROCrate::Crate, context:)
|
155
|
+
crate = initialize_crate(entity_hash, source, crate_class: crate_class, context: context)
|
156
|
+
|
157
|
+
extract_data_entities(crate, source, entity_hash).each do |entity|
|
158
|
+
crate.add_data_entity(entity)
|
159
|
+
end
|
160
|
+
|
161
|
+
# The remaining entities in the hash must be contextual.
|
162
|
+
extract_contextual_entities(crate, entity_hash).each do |entity|
|
163
|
+
crate.add_contextual_entity(entity)
|
164
|
+
end
|
165
|
+
|
166
|
+
crate
|
167
|
+
end
|
168
|
+
|
169
|
+
##
|
170
|
+
# Initialize a crate from the given set of entities.
|
171
|
+
#
|
172
|
+
# @param entity_hash [Hash{String => Hash}] A Hash containing all the entities in the @graph, mapped by their @id.
|
173
|
+
# @param source [String, ::File, Pathname] The location of the RO-Crate being read.
|
174
|
+
# @param crate_class [Class] The class to use to instantiate the crate,
|
175
|
+
# useful if you have created a subclass of ROCrate::Crate that you want to use. (defaults to ROCrate::Crate).
|
176
|
+
# @param context [nil, String, Array, Hash] A custom JSON-LD @context (parsed), or nil to use default.
|
177
|
+
# @return [Crate] The RO-Crate.
|
178
|
+
def self.initialize_crate(entity_hash, source, crate_class: ROCrate::Crate, context:)
|
179
|
+
crate_class.new.tap do |crate|
|
146
180
|
crate.properties = entity_hash.delete(ROCrate::Crate::IDENTIFIER)
|
147
181
|
crate.metadata.properties = entity_hash.delete(ROCrate::Metadata::IDENTIFIER)
|
182
|
+
crate.metadata.context = context
|
148
183
|
preview_properties = entity_hash.delete(ROCrate::Preview::IDENTIFIER)
|
149
|
-
|
150
|
-
|
151
|
-
|
152
|
-
crate.
|
153
|
-
end
|
154
|
-
# The remaining entities in the hash must be contextual.
|
155
|
-
extract_contextual_entities(crate, entity_hash).each do |entity|
|
156
|
-
crate.add_contextual_entity(entity)
|
184
|
+
preview_path = ::File.join(source, ROCrate::Preview::IDENTIFIER)
|
185
|
+
preview_path = ::File.exists?(preview_path) ? Pathname.new(preview_path) : nil
|
186
|
+
if preview_properties || preview_path
|
187
|
+
crate.preview = ROCrate::Preview.new(crate, preview_path, preview_properties || {})
|
157
188
|
end
|
189
|
+
crate.add_all(source, false)
|
158
190
|
end
|
159
191
|
end
|
160
192
|
|
@@ -226,8 +258,8 @@ module ROCrate
|
|
226
258
|
##
|
227
259
|
# Extract the metadata entity from the entity hash, according to the rules defined here:
|
228
260
|
# https://www.researchobject.org/ro-crate/1.1/root-data-entity.html#finding-the-root-data-entity
|
229
|
-
# @return [Hash{String => Hash}] A Hash containing (hopefully) one value, the metadata entity's properties
|
230
|
-
# mapped by its @id.
|
261
|
+
# @return [nil, Hash{String => Hash}] A Hash containing (hopefully) one value, the metadata entity's properties
|
262
|
+
# mapped by its @id, or nil if nothing is found.
|
231
263
|
def self.extract_metadata_entity(entities)
|
232
264
|
key = entities.detect do |_, props|
|
233
265
|
props.dig('conformsTo', '@id')&.start_with?(ROCrate::Metadata::RO_CRATE_BASE)
|
@@ -242,6 +274,13 @@ module ROCrate
|
|
242
274
|
entities.delete(ROCrate::Metadata::IDENTIFIER_1_0))
|
243
275
|
end
|
244
276
|
|
277
|
+
##
|
278
|
+
# Extract the ro-crate-preview entity from the entity hash.
|
279
|
+
# @return [Hash{String => Hash}] A Hash containing the preview entity's properties mapped by its @id, or nil if nothing is found.
|
280
|
+
def self.extract_preview_entity(entities)
|
281
|
+
entities.delete("./#{ROCrate::Preview::IDENTIFIER}") || entities.delete(ROCrate::Preview::IDENTIFIER)
|
282
|
+
end
|
283
|
+
|
245
284
|
##
|
246
285
|
# Extract the root entity from the entity hash, according to the rules defined here:
|
247
286
|
# https://www.researchobject.org/ro-crate/1.1/root-data-entity.html#finding-the-root-data-entity
|
@@ -252,5 +291,23 @@ module ROCrate
|
|
252
291
|
raise "Metadata entity does not reference any root entity" unless root_id
|
253
292
|
entities.delete(root_id)
|
254
293
|
end
|
294
|
+
|
295
|
+
##
|
296
|
+
# Finds an RO-Crate's root directory (where `ro-crate-metdata.json` is located) within a given directory.
|
297
|
+
#
|
298
|
+
# @param source [String, ::File, Pathname] The location of the directory.
|
299
|
+
# @return [Pathname, nil] The path to the root, or nil if not found.
|
300
|
+
def self.detect_root_directory(source)
|
301
|
+
Pathname(source).find do |entry|
|
302
|
+
if entry.file?
|
303
|
+
name = entry.basename.to_s
|
304
|
+
if name == ROCrate::Metadata::IDENTIFIER || name == ROCrate::Metadata::IDENTIFIER_1_0
|
305
|
+
return entry.parent
|
306
|
+
end
|
307
|
+
end
|
308
|
+
end
|
309
|
+
|
310
|
+
nil
|
311
|
+
end
|
255
312
|
end
|
256
313
|
end
|