valkyrie 3.2.0 → 3.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7439f8be7099e4b54998826abda9222d07226ff7a54fdb868f2ab7d99596fd56
-  data.tar.gz: 866c57941b2e2604a8d7bd8b013e0b31c07dd909404733cfb7043c61412389ad
+  metadata.gz: 3bd57af32d07a78e2ac9a9a1c6217b8b601018413c1a08a67762565f92da91cd
+  data.tar.gz: 58e0073f9ef70694e77d016cf08869d90b920f2fca3a7c42e7141289a8b7121d
 SHA512:
-  metadata.gz: 89d1857a2f40c0d8c32c6a31a9ff1247e9b8c32aa84e917c77c3d89e6a0cb52c7548802852914ddcf11d9805f3cf0b6784331f937cf9004b18e63ecb27449971
-  data.tar.gz: 5b143107a9ca53b86c83c5ba884635fb869919a7641c1373324193f9ac918684557177cf2c2519f5d8eeb0420bd48404037da5fb46c84d5a17bda0e4381b72f7
+  metadata.gz: 939dbb151e2cb770e9bfbac0cedbfc78a160ace0f3f1c482560659f1a6cd16ad0984ae1eb15caf49d4ef5926f3a01721e032f7c8363ea815e935064eb547f7c0
+  data.tar.gz: d2b295e1881cc0d8d45e1c69e6255fab7b5be4d3e277d3e01eaa53498ca7346131ac1c2180d73f675c93b9b4a03d73316b9f6a8256b6b40a7492dda8c088d16a

data/CHANGELOG.md

@@ -1,3 +1,9 @@
+# v3.3.0 2024-07-03
+
+## Changes since last release
+
+* Add support for configurable pairtree IDs in Fedora 6. ([randalldfloyd](https://github.com/randalldfloyd))
+
 # v3.2.0 2024-04-10
 
 ## Changes since last release

data/README.md

@@ -234,6 +234,39 @@ config.  If Fedora requires auth, you can also include that in the URL, e.g.:
      fedora_version: 5
    )
    ```
+#### Pairtree paths in Fedora 4/6
+Fedora 4 and 6.5+ support automatic creation of "pairtree" paths, which means that identifiers are stored in
+a hashed directory structure to avoid excessive child nodes directly in the root node.
+
+Unlike Fedora 4 where this is the default behavior, it has to be enabled via options in Fedora 6.5. This is 
+done by configuring both the count and length of the pairtree algorithm, so it needs to be configured
+in the Fedora adapters so that both sides match.
+
+In the following example, Fedora 6.5 is being started with options to create pairtree paths consisting of 
+4 segments, 2 characters in length:  
+
+```angular2html
+CATALINA_OPTS=-Dfcrepo.home=/fcrepo-home ...
+...
+-Dfcrepo.pid.minter.length=2 -Dfcrepo.pid.minter.count=4
+```
+For the Fedora metadata/storage adapters to correctly translate identifiers into URI's, they should be initialized
+like:
+
+```angular2html
+Valkyrie::Persistence::Fedora::MetadataAdapter.new(
+    connection: ::Ldp::Client.new("http://localhost:8080/fcrepo/rest"),
+    fedora_version: 6.5,
+    fedora_pairtree_count: 4,
+    fedora_pairtree_length: 2
+  )
+```
+
+In the configuration above, an ID of `AaBbCcDd` will be created in Fedora under the root node as `/Aa/Bb/Cc/Dd/AaBbCcDd`. 
+If count and length correctly match in the Fedora adapters, that same path will be calculated and returned.
+
+Note that the configuration above is not required for pairtree paths to work correctly in Fedora 4, and automatic
+pairtree creation is not supported under Fedora 5 or Fedora 6 < 6.5. 
 
 ## Installing a Development environment
 

data/lib/valkyrie/persistence/fedora/metadata_adapter.rb

@@ -9,17 +9,23 @@ module Valkyrie::Persistence::Fedora
   #     schema: Valkyrie::Persistence::Fedora::PermissiveSchema.new(title: RDF::URI("http://example.com/title"))
   #   )
   class MetadataAdapter
-    attr_reader :connection, :base_path, :schema, :fedora_version
+    attr_reader :connection, :base_path, :schema, :fedora_version, :pairtree_count, :pairtree_length
 
     # @param [Ldp::Client] connection
     # @param [String] base_path
     # @param [Valkyrie::Persistence::Fedora::PermissiveSchema] schema
     # @param [Integer] fedora_version
-    def initialize(connection:, base_path: "/", schema: Valkyrie::Persistence::Fedora::PermissiveSchema.new, fedora_version: Valkyrie::Persistence::Fedora::DEFAULT_FEDORA_VERSION)
+    # @param [Integer] fedora_pairtree_count
+    # @param [Integer] fedora_pairtree_length
+    def initialize(connection:, base_path: "/", schema: Valkyrie::Persistence::Fedora::PermissiveSchema.new, # rubocop:disable Metrics/ParameterLists
+                   fedora_version: Valkyrie::Persistence::Fedora::DEFAULT_FEDORA_VERSION,
+                   fedora_pairtree_count: 0, fedora_pairtree_length: 0)
       @connection = connection
       @base_path = base_path
       @schema = schema
       @fedora_version = fedora_version
+      @pairtree_count = fedora_pairtree_count
+      @pairtree_length = fedora_pairtree_length
     end
 
     # Construct the query service object using this adapter
@@ -58,7 +64,8 @@ module Valkyrie::Persistence::Fedora
     # @param [RDF::URI] id the Valkyrie ID
     # @return [RDF::URI]
     def id_to_uri(id)
-      prefix = [5, 6].include?(fedora_version) ? "" : "#{pair_path(id)}/"
+      prefix = ""
+      prefix = "#{pair_path(id)}/" if fedora_version == 4 || (fedora_version >= 6.5 && (pairtree_count * pairtree_length).positive?)
       RDF::URI("#{connection_prefix}/#{prefix}#{CGI.escape(id.to_s)}")
     end
 
@@ -68,7 +75,16 @@ module Valkyrie::Persistence::Fedora
     # @param [Valkyrie::ID] id the Valkyrie ID
     # @return [Array<String>]
     def pair_path(id)
-      id.to_s.split(/[-\/]/).first.split("").each_slice(2).map(&:join).join("/")
+      if fedora_version >= 6.5
+        # When configurable, pair the part up to count * length, but only up to a slash
+        pair_part = id.to_s[0, pairtree_count * pairtree_length].split(/[\/]/).first
+        slice_length = pairtree_length
+      else
+        # When not configurable, pair the full string, but only up to a dash or slash
+        pair_part = id.to_s.split(/[-\/]/).first
+        slice_length = 2
+      end
+      pair_part.split("").each_slice(slice_length).map(&:join).join("/")
     end
 
     def url_prefix

data/lib/valkyrie/storage/fedora.rb

@@ -2,15 +2,26 @@
 module Valkyrie::Storage
   # Implements the DataMapper Pattern to store binary data in fedora
   class Fedora
-    attr_reader :connection, :base_path, :fedora_version
+    attr_reader :connection, :base_path, :fedora_version, :pairtree_count, :pairtree_length, :uri_transformer
     PROTOCOL = 'fedora://'
     SLASH = '/'
 
     # @param [Ldp::Client] connection
-    def initialize(connection:, base_path: "/", fedora_version: Valkyrie::Persistence::Fedora::DEFAULT_FEDORA_VERSION)
+    # @param [Integer] fedora_pairtree_count
+    # @param [Integer] fedora_pairtree_length
+    def initialize(connection:, base_path: "/", fedora_version: Valkyrie::Persistence::Fedora::DEFAULT_FEDORA_VERSION,
+                   fedora_pairtree_count: 0, fedora_pairtree_length: 0)
       @connection = connection
       @base_path = base_path
       @fedora_version = fedora_version
+      @pairtree_count = fedora_pairtree_count
+      @pairtree_length = fedora_pairtree_length
+
+      @uri_transformer = if fedora_version >= 6.5 && (pairtree_count * pairtree_length).positive?
+                           pairtree_resource_uri_transformer
+                         else
+                           default_resource_uri_transformer
+                         end
     end
 
     # @param id [Valkyrie::ID]
@@ -24,7 +35,7 @@ module Valkyrie::Storage
     def supports?(feature)
       return true if feature == :versions
       # Fedora 6 auto versions and you can't delete versions.
-      return true if feature == :version_deletion && fedora_version != 6
+      return true if feature == :version_deletion && fedora_version < 6
       false
     end
 
@@ -44,7 +55,7 @@ module Valkyrie::Storage
     # @param extra_arguments [Hash] additional arguments which may be passed to other adapters
     # @return [Valkyrie::StorageAdapter::StreamFile]
     def upload(file:, original_filename:, resource:, content_type: "application/octet-stream", # rubocop:disable Metrics/ParameterLists
-               resource_uri_transformer: default_resource_uri_transformer, **_extra_arguments)
+               resource_uri_transformer: uri_transformer, **_extra_arguments)
       identifier = resource_uri_transformer.call(resource, base_url) + '/original'
       upload_file(fedora_uri: identifier, io: file, content_type: content_type, original_filename: original_filename)
       # Fedora 6 auto versions, so check to see if there's a version for this
@@ -60,7 +71,7 @@ module Valkyrie::Storage
       uri = fedora_identifier(id: id)
       # Fedora 6 has auto versioning, so have to sleep if it's too soon after last
       # upload.
-      if fedora_version == 6 && current_version_id(id: id).to_s.split("/").last == Time.current.utc.strftime("%Y%m%d%H%M%S")
+      if fedora_version >= 6 && current_version_id(id: id).to_s.split("/").last == Time.current.utc.strftime("%Y%m%d%H%M%S")
         sleep(0.5)
         return upload_version(id: id, file: file)
       end
@@ -102,7 +113,7 @@ module Valkyrie::Storage
     end
 
     def upload_file(fedora_uri:, io:, content_type: "application/octet-stream", original_filename: "default")
-      sha1 = [5, 6].include?(fedora_version) ? "sha" : "sha1"
+      sha1 = fedora_version >= 5 ? "sha" : "sha1"
       connection.http.put do |request|
         request.url fedora_uri
         request.headers['Content-Type'] = content_type
@@ -141,7 +152,7 @@ module Valkyrie::Storage
       version_id ||= id if id != current_id
       # No version got passed and we're asking for a current_id, gotta get the
       # version ID
-      return perform_find(id: current_id, version_id: (current_version_id(id: id) || :empty)) if version_id.nil?
+      return perform_find(id: current_id, version_id: current_version_id(id: id) || :empty) if version_id.nil?
       Valkyrie::StorageAdapter::StreamFile.new(id: current_id, io: response(id: id), version_id: version_id)
     end
 
@@ -211,6 +222,18 @@ module Valkyrie::Storage
       end
     end
 
+    def pairtree_resource_uri_transformer
+      lambda do |resource, base_url|
+        id = CGI.escape(resource.id.to_s)
+        return RDF::URI.new(base_url + id) unless (pairtree_count * pairtree_length).positive?
+
+        pair_part = id.to_s[0, pairtree_count * pairtree_length].split(/[\/]/).first
+        slice_length = pairtree_length
+        prefix = pair_part.split("").each_slice(slice_length).map(&:join).join("/")
+        RDF::URI.new(base_url + "#{prefix}/" + id)
+      end
+    end
+
     def base_url
       pre_divider = base_path.starts_with?(SLASH) ? '' : SLASH
       post_divider = base_path.ends_with?(SLASH) ? '' : SLASH

data/lib/valkyrie/version.rb

@@ -1,4 +1,4 @@
 # frozen_string_literal: true
 module Valkyrie
-  VERSION = "3.2.0"
+  VERSION = "3.3.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: valkyrie
 version: !ruby/object:Gem::Version
-  version: 3.2.0
+  version: 3.3.0
 platform: ruby
 authors:
 - Trey Pendragon
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-05-07 00:00:00.000000000 Z
+date: 2024-07-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: dry-struct