citesight 0.1.2 → 1.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: 7e11117f9c3c2e1eacd42126bc7a97ed7841ee28
-  data.tar.gz: a0dfd14a313ec7ece65bed421eb335fdda9427de
+SHA256:
+  metadata.gz: f8bbbcea21b92c660093f84144abd1c113ea3632d766759d3fd323b7b3dc5cf7
+  data.tar.gz: fd1b56b1ab356e63d2519844deccd8685b967edfc1fa8995c039e71a8c8bbef9
 SHA512:
-  metadata.gz: ace1dc22e05ee5eebe3383d075d8965ab503209bbd14dd8d72ca042aff3608c7497404fd1d5acebe5317c35b5e6c973604f527665f7425cbfa26503436da5b3e
-  data.tar.gz: 49ab665d7ea38767d508681707a15c0c6063dad04df974354e352a859311e4bf6b2b00b9c673391e0b5a6498102a777ac76c91a3fe4106ce4ea78f58bc15bba2
+  metadata.gz: b638261f8e8ec4fbfc3e685fca5e223d5949078c13d62cd240a2b0819eaca0be500f1748c0f9638fc00923b11a2158f48aedc5ff0ae6d862b418883067fb0a1b
+  data.tar.gz: '079afeb512af5a7eff458a2cfa23d50892a871380058c3e5dcdc8218ee8326250fd6b620a4565cd4e503f623c7c37e725972bdbe266b2c20958a8ad987e005f8'

data/.github/workflows/ruby.yml

@@ -0,0 +1,39 @@
+# This workflow uses actions that are not certified by GitHub.
+# They are provided by a third-party and are governed by
+# separate terms of service, privacy policy, and support
+# documentation.
+# This workflow will download a prebuilt Ruby version, install dependencies and run tests with Rake
+# For more information see: https://github.com/marketplace/actions/setup-ruby-jruby-and-truffleruby
+
+name: Ruby
+
+on:
+  push:
+    branches: [ "main", "develop" ]
+  pull_request:
+    branches: '**'
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+
+    strategy:
+      matrix:
+        os: [ubuntu, macos]
+        ruby-version: ['3.0', '3.1', '3.2']
+    runs-on: ${{ matrix.os }}-latest
+
+    steps:
+    - uses: actions/checkout@v3
+    - name: Set up Ruby
+    # To automatically get bug fixes and new Ruby versions for ruby/setup-ruby,
+    # change this to (see https://github.com/ruby/setup-ruby#versioning):
+    # uses: ruby/setup-ruby@v1
+      uses: ruby/setup-ruby@55283cc23133118229fd3f97f9336ee23a179fcf # v1.146.0
+      with:
+        ruby-version: ${{ matrix.ruby-version }}
+        bundler-cache: true # runs 'bundle install' and caches installed gems automatically
+    - name: Run tests
+      run: bundle exec rake spec

data/README.md

@@ -1,4 +1,5 @@
-#citesight
+# citesight
+
 [![Gem Version](https://badge.fury.io/rb/citesight.svg)](http://badge.fury.io/rb/citesight)
 [![Build Status](https://travis-ci.org/soumyaray/citesight.svg?branch=master)](https://travis-ci.org/soumyaray/citesight)
 
@@ -6,17 +7,24 @@ home: [https://github.com/soumyaray/citesight](https://github.com/soumyaray/cite
 Gem to extract and report on citations in an academic text
 
 ## Usage
+
 This gem may be used as a command line utility or called from code
 
-### CLI:
+<!-- TODO: installation instructions -->
+
+### CLI
+
     citesight mydoc.txt
 
-### Code example:
+<!-- TODO: update usage instructions -->
+### Code example
+
     require 'citesight'
     require 'pp'
 
     contents = File.read("spec/testfiles/large_test.txt", :encoding => "UTF-8")
-    cites = PaperCitations.unique_cites(contents)
+    paper = PaperCitations.new(contents)
+    cites = paper.unique_cites
 
     puts "\nTotal unique citations: #{cites.count}"
     PP.pp(Hash[cites])
@@ -24,5 +32,5 @@ This gem may be used as a command line utility or called from code
     top_cite = cites.sort_by { |_c, count| count}.reverse.first[0]
     puts "\nYour top citation: #{top_cite}"
 
-    top_cite_indexes = PaperCitations.index_of_cite(contents, top_cite)
+    top_cite_indexes = paper.index_of_cite(top_cite)
     puts "It was cited at locations: #{top_cite_indexes.join(', ')}"

data/bin/citesight

@@ -5,12 +5,13 @@ require 'citesight'
 require 'pp'
 
 # executable requirements: (1) env shebang above; (2) file mode 0755
+# TODO: Return with error message instead of failing
 
 fail ArgumentError, "Usage:  get_citations [filename]\n" if ARGV.count == 0
 
 contents = File.open(ARGV[0], 'r').read
 
-results = CiteSight::PaperCitations.unique_cites(contents).sort_by do |c, _|
+results = CiteSight::PaperCitations.new(contents).unique_cites.sort_by do |c, _|
   c.downcase
 end
 

data/citesight.gemspec

@@ -7,17 +7,17 @@ Gem::Specification.new do |s|
   s.date        =  CiteSight::DATE
 
   s.executables << 'citesight'
-  s.add_development_dependency 'minitest'
-  s.add_development_dependency 'minitest-rg'
+  s.add_development_dependency 'minitest', '~> 5.20'
+  s.add_development_dependency 'minitest-rg', '~> 5.3'
 
   s.summary     =  'Citation extractor and analyzer'
-  s.description =  'Extract and analyze citations from APA style text'
+  s.description =  'Extract and analyze citations from MISQ and APA style text'
   s.authors     =  ['Soumya Ray']
   s.email       =  'soumya.ray@gmail.com'
 
   s.files       =  `git ls-files`.split("\n")
   s.test_files  =  `git ls-files -- {test,spec,features}/*`.split("\n")
-  
+
   s.homepage    =  'https://github.com/soumyaray/citesight'
   s.license     =  'MIT'
 end

data/lib/citesight/paper_citations.rb

@@ -1,16 +1,8 @@
-# This class extracts and counts APA-style citations in a paper.
+# This class extracts and counts MISQ or APA style citations in a paper.
 # The unique_cites method returns a hash of citations and counts
 # in the order in which they were encountered.
 module CiteSight
   class PaperCitations
-    def self.unique_cites(contents)
-      new(contents).unique_cites
-    end
-
-    def self.index_of_cite(contents, cite)
-      new(contents).index_of_cite(cite)
-    end
-
     def initialize(contents)
       @contents = contents
     end
@@ -39,19 +31,23 @@ module CiteSight
 
     private
 
+    def apostrophe() "\'\u2019" end
     def prefix() '(([dD]e|[vV]an[ ]?[dD]er)[ ]?)' end
-    def author() "(#{prefix}?[A-Z][[:alpha:]\'\u2019\-]+)" end
-    def other_authors() "([ ]and[ ]#{author} | ([ ]et[ ]al.){1})" end
-    def possessive() "([\'\u2019]s|[\'\u2019])" end
+    def author() "(#{prefix}?[A-Z][[:alpha:]#{apostrophe}\-]+)" end
+    def other_authors() "([ ](and|\&)[ ]#{author} | ([ ]et[ ]al.){1})" end
+    def possessive() "([#{apostrophe}]s|[#{apostrophe}])" end
     def year_literal() "[1-2][0-9]{3}[a-z]?" end
-    def year(yr) "([ ][\(]?#{yr}[,\)\;])" end
+    def year(yr) "([,]?[ ][\(]?#{yr}[,\)\;])" end
 
     def cite_match
       /( #{author}{1}#{other_authors}?#{possessive}?#{year(year_literal)} )/x
     end
 
     def remove_punctuation(cite)
-      cite.gsub(/[\(\),;]|([\'\u2019]s)/, '').gsub(/[\'\u2019]\s/, ' ')
+      cite
+        .gsub(/[\(\),;]|([#{apostrophe}]s)/, '')
+        .gsub(/[#{apostrophe}]\s/, ' ')
+        .gsub(/\&/, 'and')
     end
   end
 end

data/lib/citesight/version.rb

@@ -1,4 +1,4 @@
 module CiteSight
-  VERSION = '0.1.2'
-  DATE    = '2016-05-10'
+  VERSION = '1.0.0'
+  DATE    = '2023-12-27'
 end

data/spec/citesight_spec.rb

@@ -1,23 +1,22 @@
 require 'minitest/autorun'
 require 'minitest/rg'
-require './spec/minitest_helper.rb'
+require './spec/spec_helper.rb'
 
 describe 'Paper', 'A text document' do
 
   describe 'when there are citations in the text' do
-    before do
-      @results = CiteSight::PaperCitations.unique_cites(TEST_CONTENTS)
-    end
-
-    it 'should fine the right citations' do
-      @results.must_equal TEST_CITES
+    TEST_RESULTS.each do |cite, count|
+      it "should find the right citation count for #{cite}" do
+        _(TEST_CITES[cite]).must_equal count
+      end
     end
   end
 
   describe 'when there are no citations to be found' do
     it 'should return an empty hash' do
       no_cites_txt = 'these are not the citations you are looking for'
-      # TODO: check return of empty hash
+      _(CiteSight::PaperCitations.new(no_cites_txt).unique_cites.any?)
+        .must_equal(false)
     end
   end
 end
@@ -25,24 +24,17 @@ end
 describe 'Cases', 'Test different citation cases' do
   TEST_CASES.keys.each do |k|
     it "can detect #{k}" do
-      CiteSight::PaperCitations.unique_cites(TEST_CASES[k]["case"]).to_a\
+      _(CiteSight::PaperCitations.new(TEST_CASES[k]["case"]).unique_cites.to_a)
         .must_equal(TEST_CASES[k]["result"])
     end
   end
 end
 
 describe 'Indexes', 'Accurately find index of different citations' do
-  it "can find the right index for all citations" do
-    TEST_CITES.map do |cite, _count|
-      TEST_INDEXES[cite].must_equal \
-                CiteSight::PaperCitations.index_of_cite(TEST_CONTENTS, cite)
+  TEST_CITES.map do |cite, _count|
+    it "can find the right index for #{cite}" do
+        _(TEST_INDEXES[cite]).must_equal \
+          CiteSight::PaperCitations.new(TEST_CONTENTS).index_of_cite(cite)
     end
   end
 end
-
-## Produce hash of all cite indexes:
-# Hash[
-#   TEST_CITES.map do |cite, count|
-#     [cite, PaperCitations.index_cite(doc, cite)]
-#   end
-# ]

data/spec/{minitest_helper.rb → spec_helper.rb}

@@ -8,10 +8,10 @@ TEST_CITES =
     "Charlton 2002" => 1,
     "Griffiths 2000" => 1,
     "Peters et al. 2007" => 2,
-    "Ma and Agarwal 2007" => 1,
-    "Hur 2007" => 2,
+    "Ma and Agarwal 2007" => 3,
+    "Hur 2007" => 3,
     "Oreg 1995b" => 1,
-    "Ray 2000" => 2,
+    "Ray 2000" => 4,
     "Ma et al. 2002" => 3,
     "Griffins 2000" => 1,
     "O'Hern 2010" => 1,
@@ -27,9 +27,9 @@ TEST_INDEXES =
     "Griffiths 2000"=>[97],
     "Peters et al. 2007"=>[219, 500],
     "Ma and Agarwal 2007"=>[174],
-    "Hur 2007"=>[146, 490],
+    "Hur 2007"=>[146, 490, 828],
     "Oreg 1995b"=>[318],
-    "Ray 2000"=>[346, 432],
+    "Ray 2000"=>[346, 432, 781, 817],
     "Ma et al. 2002"=>[272, 377, 407],
     "Griffins 2000"=>[522],
     "O'Hern 2010"=>[566],
@@ -42,3 +42,5 @@ TEST_INDEXES =
 TEST_CASES = File.open('./spec/testfiles/test_cases.json', 'r') do |f|
   JSON.load(f)
 end
+
+TEST_RESULTS = CiteSight::PaperCitations.new(TEST_CONTENTS).unique_cites

data/spec/testfiles/test.txt

@@ -11,3 +11,6 @@ Multiple cites: in (Hur 2007, Peters et al.’s 2007, Griffins' 2000) from
 Apostrophes in names: (O'Hern 2010) and Wa'el et al. (1993) or
 Capitals within names: from McDonald (2003) we gather that
 Multiword last names: even from van der Aalst (2004) or others (De Boor 1980).
+APA commas: according to literature (Ray, 2000) the
+APA multiple cites: (Ray, 2000; Hur, 2007)
+APA two authors: Ma & Agarwal (2007) agree with this (Ma & Agarwal 2007)

metadata

@@ -1,44 +1,44 @@
 --- !ruby/object:Gem::Specification
 name: citesight
 version: !ruby/object:Gem::Version
-  version: 0.1.2
+  version: 1.0.0
 platform: ruby
 authors:
 - Soumya Ray
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2016-05-10 00:00:00.000000000 Z
+date: 2023-12-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: minitest
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '5.20'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '5.20'
 - !ruby/object:Gem::Dependency
   name: minitest-rg
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '5.3'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
-description: Extract and analyze citations from APA style text
+        version: '5.3'
+description: Extract and analyze citations from MISQ and APA style text
 email: soumya.ray@gmail.com
 executables:
 - citesight
@@ -46,8 +46,8 @@ extensions: []
 extra_rdoc_files: []
 files:
 - ".bundle/config"
+- ".github/workflows/ruby.yml"
 - ".gitignore"
-- ".travis.yml"
 - Gemfile
 - LICENSE
 - README.md
@@ -58,7 +58,7 @@ files:
 - lib/citesight/paper_citations.rb
 - lib/citesight/version.rb
 - spec/citesight_spec.rb
-- spec/minitest_helper.rb
+- spec/spec_helper.rb
 - spec/testfiles/large_test.txt
 - spec/testfiles/large_test_results.txt
 - spec/testfiles/no_match.txt
@@ -68,7 +68,7 @@ homepage: https://github.com/soumyaray/citesight
 licenses:
 - MIT
 metadata: {}
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -83,14 +83,13 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.5.1
-signing_key: 
+rubygems_version: 3.4.10
+signing_key:
 specification_version: 4
 summary: Citation extractor and analyzer
 test_files:
 - spec/citesight_spec.rb
-- spec/minitest_helper.rb
+- spec/spec_helper.rb
 - spec/testfiles/large_test.txt
 - spec/testfiles/large_test_results.txt
 - spec/testfiles/no_match.txt

data/.travis.yml

@@ -1,10 +0,0 @@
-language: ruby
-rvm:
-  - ruby-head
-  - 2.1.0
-
-branches:
-  only:
-    - master
-    - develop
-    - /^rel-.*$/