research_metadata_announcement 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 9cc9974c6b361bb579f8454ffcf2602d78a62746
+  data.tar.gz: 33da4d070192919b91c45317d7a026f6c8f19f20
+SHA512:
+  metadata.gz: 5db2978940ff75facb1fba4cf2d53f9a3abd462c3131b78cb1c518bbf77335f61116f7d993c213af663c7e7598c4baa4292fa8bfee4f4d2d9fbf4bb00e9142f0
+  data.tar.gz: bd79b981ed25b340755a255f609789415c695afbdea235d1e75cb9f9da6650ee39204e235273d0053e4fe6de9a7504501c5830655adecc59128bf4577c088875

data/.gitignore

@@ -0,0 +1,18 @@
+*.gem
+*.rbc
+.bundle
+.config
+.yardoc
+Gemfile.lock
+InstalledFiles
+_yardoc
+coverage
+doc/
+lib/bundler/man
+pkg
+rdoc
+spec/reports
+test/tmp
+test/version_tmp
+tmp
+.idea

data/CHANGELOG.md

@@ -0,0 +1,9 @@
+# Change Log
+All notable changes to this project will be documented in this file.
+This project adheres to [Semantic Versioning](http://semver.org/).
+
+## Unreleased
+
+## 0.1.0 - 2017-09-26
+### Added
+- Working product supports datasets.

data/Gemfile

@@ -0,0 +1,4 @@
+source "https://rubygems.org"
+
+# Specify your gem's dependencies in research_metadata_announcement.gemspec
+gemspec

data/README.md

@@ -0,0 +1,89 @@
+# ResearchMetadataAnnouncement
+
+Metadata extraction from the Pure Research Information System and transformation of the metadata into an announcement.
+
+## Status
+
+[![Gem Version](https://badge.fury.io/rb/research_metadata_announcement.svg)](https://badge.fury.io/rb/research_metadata_announcement)
+[![Build Status](https://semaphoreci.com/api/v1/aalbinclark/research_metadata_announcement/branches/master/badge.svg)](https://semaphoreci.com/aalbinclark/research_metadata_announcement)
+[![Code Climate](https://codeclimate.com/github/lulibrary/research_metadata_announcement/badges/gpa.svg)](https://codeclimate.com/github/lulibrary/research_metadata_announcement)
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+    gem 'research_metadata_announcement'
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install research_metadata_announcement
+
+## Usage
+
+### Configuration
+
+Create a hash for passing to a transformer.
+
+```ruby
+# Pure host with authentication.
+config = {
+  url:      ENV['PURE_URL'],
+  username: ENV['PURE_USERNAME'],
+  password: ENV['PURE_PASSWORD'],
+}
+```
+
+```ruby
+# Pure host without authentication.
+config = {
+  url: ENV['PURE_URL']
+}
+```
+
+### Transformation
+
+Create a metadata transformer for a Pure dataset.
+
+```ruby
+transformer = ResearchMetadataAnnouncement::Transformer::Dataset.new config
+```
+
+Give it a Pure identifier and extract the metadata.
+
+```ruby
+transformer.extract uuid: 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx'
+```
+
+An announcement can be obtained in various formats, assuming the metadata is
+available and the announcement length does not exceed the optional max_length
+argument. Each example uses a different dataset for illustrative purposes.
+
+```ruby
+
+transformer.title_uri
+#=> "Operating Nanobeams in a Quantum Fluid. dx.doi.org/10.17635/lancaster/researchdata/139."
+
+transformer.title_uri max_length: 140
+#=> "Ruthenium Volatilisation from Reprocessed Spent Nuclear Fuel – Studying the Baseline Therm... dx.doi.org/10.17635/lancaster/researchdata/14."
+
+transformer.uri_title
+#=> "dx.doi.org/10.17635/lancaster/researchdata/29. Herpes simplex virus 1 (HSV-1) evolution."
+
+transformer.keywords_uri
+#=> "smart cities, sustainability. dx.doi.org/10.17635/lancaster/researchdata/35."
+
+transformer.hashtags_uri
+#=> "#treatedhypertension #microvascularbloodflow. dx.doi.org/10.17635/lancaster/researchdata/148."
+
+transformer.uri_keywords
+#=> "dx.doi.org/10.17635/lancaster/researchdata/134. metagenomics, deep sequencing."
+
+transformer.uri_hashtags max_descriptors: 4
+#=> "dx.doi.org/10.17635/lancaster/researchdata/111. #influenza #nasopharynx #virology #virus."
+
+```
+

data/Rakefile

@@ -0,0 +1,2 @@
+require "bundler/gem_tasks"
+task :default => :spec

data/lib/research_metadata_announcement.rb

@@ -0,0 +1,11 @@
+require 'puree'
+
+require 'research_metadata_announcement/transformer/base'
+require 'research_metadata_announcement/transformer/dataset'
+require 'research_metadata_announcement/version'
+
+# Metadata extraction from the Pure Research Information System and
+# transformation of the metadata into an announcement
+#
+module ResearchMetadataAnnouncement
+end

data/lib/research_metadata_announcement/transformer/base.rb

@@ -0,0 +1,56 @@
+module ResearchMetadataAnnouncement
+  module Transformer
+
+    # Base transformer
+    #
+    class Base
+
+      # @param config [Hash]
+      # @option config [String] :url The URL of the Pure host.
+      # @option config [String] :username The username of the Pure host account.
+      # @option config [String] :password The password of the Pure host account.
+
+      def initialize(config)
+        @config = config
+      end
+
+      # Extract metadata from Pure
+      #
+      # @param id [String]
+      # @param uuid [String]
+      def extract(uuid: nil, id: nil)
+        if !uuid.nil?
+          @resource = @resource_extractor.find uuid: uuid
+        else
+          @resource = @resource_extractor.find id: id
+        end
+      end
+
+      private
+
+      def append_sentence(str, str_to_append)
+        if str_to_append && str_to_append.size > 0
+          if str
+            if str.size > 0
+              "#{str}. #{str_to_append}."
+            else
+              "#{str_to_append}."
+            end
+          end
+        else
+          str
+        end
+      end
+
+      def strip_uri_scheme(uri)
+        uri.sub /^.+\/\//, ''
+      end
+
+      def length_constrained?(max_length)
+        max_length && max_length > 0
+      end
+
+    end
+  end
+
+end

data/lib/research_metadata_announcement/transformer/dataset.rb

@@ -0,0 +1,222 @@
+module ResearchMetadataAnnouncement
+
+  module Transformer
+
+    # Extracts dataset metadata from the Pure Research Information System and
+    # converts it into an announcement
+    #
+    class Dataset < ResearchMetadataAnnouncement::Transformer::Base
+
+      # @param config [Hash]
+      # @option config [String] :url The URL of the Pure host.
+      # @option config [String] :username The username of the Pure host account.
+      # @option config [String] :password The password of the Pure host account.
+      def initialize(config)
+        super
+        @resource_extractor = Puree::Extractor::Dataset.new config
+      end
+
+      # Keywords followed by uri format
+      #
+      # @param max_length [Fixnum] Maximum length of announcement.
+      # @param max_descriptors [Fixnum] Maximum number of descriptors (common name for keywords, tags, hashtags).
+      # @return [String, nil] Announcement returned if the metadata is available and the announcement length does not exceed the max_length argument.
+      def keywords_uri(max_length: nil, max_descriptors: 2)
+        return nil if !@resource
+        keywords = @resource.keywords
+        uri = prepare_uri
+        if uri
+          return build_keywords_uri(keywords: keywords,
+                                    uri: uri,
+                                    max_length: max_length,
+                                    max_descriptors: max_descriptors)
+        end
+        nil
+      end
+
+      # Uri followed by keywords format
+      #
+      # @see #keywords_uri
+      def uri_keywords(max_length: nil, max_descriptors: 2)
+        return nil if !@resource
+        keywords = @resource.keywords
+        uri = prepare_uri
+        if uri
+          return build_uri_keywords(keywords: keywords,
+                                    uri: uri,
+                                    max_length: max_length,
+                                    max_descriptors: max_descriptors)
+        end
+        nil
+      end
+
+      # Uri followed by hashtags format
+      #
+      # @see #keywords_uri
+      def uri_hashtags(max_length: nil, max_descriptors: 2)
+        return nil if !@resource
+        keywords = @resource.keywords
+        uri = prepare_uri
+        if uri
+          return build_uri_hashtags(keywords: keywords,
+                                    uri: uri,
+                                    max_length: max_length,
+                                    max_descriptors: max_descriptors)
+        end
+        nil
+      end
+
+      # Hashtags followed by uri format
+      #
+      # @see #keywords_uri
+      def hashtags_uri(max_length: nil, max_descriptors: 2)
+        return nil if !@resource
+        keywords = @resource.keywords
+        uri = prepare_uri
+        if uri
+          return build_hashtags_uri(keywords: keywords,
+                                    uri: uri,
+                                    max_length: max_length,
+                                    max_descriptors: max_descriptors)
+        end
+        nil
+      end
+
+      # Title followed by uri format
+      #
+      # @param max_length [Fixnum]
+      # @return [String, nil] Announcement returned if the metadata is available and the announcement length does not exceed the max_length argument.
+      def title_uri(max_length: nil)
+        return nil if !@resource
+        title = @resource.title
+        uri = prepare_uri
+        if uri
+          return build_title_uri(uri: uri, title: title, max_length: max_length)
+        end
+        nil
+      end
+
+      # Uri followed by title format
+      #
+      # @see #title_uri
+      def uri_title(max_length: nil)
+        return nil if !@resource
+        title = @resource.title
+        uri = prepare_uri
+        if uri
+          return build_uri_title(uri: uri, title: title, max_length: max_length)
+        end
+        nil
+      end
+
+      private
+
+      def prepare_uri
+        if @resource && @resource.doi
+          return strip_uri_scheme @resource.doi
+        end
+        nil
+      end
+
+      def build_keywords_uri(keywords:, uri:, max_length:, max_descriptors:)
+        if !keywords.empty?
+          str = append_sentence(build_keywords(keywords, max_descriptors), uri)
+          if length_constrained? max_length
+            return str if str.size <= max_length
+          else
+            return str
+          end
+        end
+        nil
+      end
+
+      def build_uri_keywords(keywords:, uri:, max_length:, max_descriptors:)
+        if !keywords.empty?
+          str = append_sentence(uri, build_keywords(keywords, max_descriptors))
+          if length_constrained? max_length
+            return str if str.size <= max_length
+          else
+            return str
+          end
+        end
+        nil
+      end
+
+      def build_uri_hashtags(keywords:, uri:, max_length:, max_descriptors:)
+        if !keywords.empty?
+          str = append_sentence(uri, build_hashtags(keywords, max_descriptors))
+          if length_constrained? max_length
+            return str if str.size <= max_length
+          else
+            return str
+          end
+        end
+        nil
+      end
+
+      def build_hashtags_uri(keywords:, uri:, max_length:, max_descriptors:)
+        if !keywords.empty?
+          str = append_sentence(build_hashtags(keywords, max_descriptors), uri)
+          if length_constrained? max_length
+            return str if str.size <= max_length
+          else
+            return str
+          end
+        end
+        nil
+      end
+
+      def build_title_uri(title:, uri:, max_length:)
+        if length_constrained? max_length
+          available_chars = max_length - (uri.size + 3)
+          available_chars = 0 if available_chars < 0
+          if title.size <= available_chars
+            return append_sentence title, uri
+          end
+          if available_chars-3 > 0
+            truncated_title = title[0..available_chars-3].strip + '...'
+            return "#{truncated_title} #{uri}."
+          end
+        else
+          return append_sentence title, uri
+        end
+      end
+
+      def build_uri_title(uri:, title:, max_length:)
+        if length_constrained? max_length
+          available_chars = max_length - (uri.size + 3)
+          available_chars = 0 if available_chars < 0
+          if title.size <= available_chars
+            return append_sentence uri, title
+          end
+          if available_chars-3 > 0
+            truncated_title = title[0..available_chars-3].strip + '...'
+            return "#{uri}. #{truncated_title}"
+          end
+        else
+          return append_sentence uri, title
+        end
+      end
+
+      def build_keywords(keywords, max)
+        return keywords[0..max-1].join ', ' if keywords
+        nil
+      end
+
+      def build_hashtags(keywords, max)
+        a = keywords[0..max-1].map { |i| i.downcase }
+        a = a.map { |i| i.gsub(/[^a-zA-Z0-9]/,'') }
+        a = a.map { |i| i.gsub(/\s+/, '')  }
+        a = a.map { |i| "##{i}" }
+        if a.size > 0
+          return a.join ' '
+        else
+          return nil
+        end
+      end
+
+    end
+
+  end
+
+end

data/lib/research_metadata_announcement/transformer/transformer.rb

@@ -0,0 +1,7 @@
+module ResearchMetadataAnnouncement
+
+  # Transformer
+  #
+  module Transformer
+  end
+end

data/lib/research_metadata_announcement/version.rb

@@ -0,0 +1,5 @@
+module ResearchMetadataAnnouncement
+  # Semantic version number
+  #
+  VERSION = "0.1.0"
+end

data/research_metadata_announcement.gemspec

@@ -0,0 +1,29 @@
+# coding: utf-8
+lib = File.expand_path("../lib", __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require "research_metadata_announcement/version"
+
+Gem::Specification.new do |spec|
+  spec.name          = "research_metadata_announcement"
+  spec.version       = ResearchMetadataAnnouncement::VERSION
+  spec.authors       = ["Adrian Albin-Clark"]
+  spec.email         = ["a.albin-clark@lancaster.ac.uk"]
+
+  spec.summary       = %q{Metadata extraction from the Pure Research Information System and transformation of the metadata into an announcement.}
+
+  spec.license       = 'MIT'
+
+  spec.homepage      = "https://github.com/lulibrary/research_metadata_announcement"
+
+  spec.files         = `git ls-files -z`.split("\x0").reject do |f|
+    f.match(%r{^(test|spec|features)/})
+  end
+
+  spec.require_paths = ["lib"]
+
+  spec.required_ruby_version = '~> 2.1'
+
+  spec.add_runtime_dependency "puree", "~> 1.4"
+
+  spec.add_development_dependency "minitest-reporters", "~> 1.1"
+end

metadata

@@ -0,0 +1,84 @@
+--- !ruby/object:Gem::Specification
+name: research_metadata_announcement
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Adrian Albin-Clark
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2017-09-26 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: puree
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.4'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.4'
+- !ruby/object:Gem::Dependency
+  name: minitest-reporters
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.1'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.1'
+description: 
+email:
+- a.albin-clark@lancaster.ac.uk
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- CHANGELOG.md
+- Gemfile
+- README.md
+- Rakefile
+- lib/research_metadata_announcement.rb
+- lib/research_metadata_announcement/transformer/base.rb
+- lib/research_metadata_announcement/transformer/dataset.rb
+- lib/research_metadata_announcement/transformer/transformer.rb
+- lib/research_metadata_announcement/version.rb
+- research_metadata_announcement.gemspec
+homepage: https://github.com/lulibrary/research_metadata_announcement
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - "~>"
+    - !ruby/object:Gem::Version
+      version: '2.1'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.2.2
+signing_key: 
+specification_version: 4
+summary: Metadata extraction from the Pure Research Information System and transformation
+  of the metadata into an announcement.
+test_files: []