citesight 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 60b2a4cc5aaf0a5c7b4f29aa1f7adc0912e2d83a
+  data.tar.gz: 38770a64ec74e3bd40eddc6bb05ad3911a9ca058
+SHA512:
+  metadata.gz: f8745a2261b85156d191068357b10210c82fabd510a7a6206a3cfb27a9643b325a7265b5eb88aba846c18b4e82179e640be0f46b66d79d413860db9bd80226c1
+  data.tar.gz: 42cb14f320069779e39427bc94dd745e7703a67b33793f32410b456a6dbca0687742e629f15f93e8a2cab64e0cbbfca7264242cceabe2c4afc731c5214c13b05

data/bin/citesight

@@ -0,0 +1,18 @@
+#!/usr/bin/env ruby
+
+require 'citesight'
+require 'pp'
+
+# executable requirements: (1) env shebang above; (2) file mode 0755
+
+fail ArgumentError, "Usage:  get_citations [filename]\n" if ARGV.count == 0
+
+contents = File.open(ARGV[0], 'r').read
+
+results = PaperCitations.unique_cites(contents).sort_by { |c, _| c.downcase }
+
+puts "Top citations:\n"
+PP.pp(Hash[results.sort_by { |_cite, count| count }.reverse.take(5)])
+
+puts "\nTotal unique citations: #{results.count}"
+PP.pp(Hash[results])

data/lib/citesight.rb

@@ -0,0 +1,41 @@
+# This class extracts and counts 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.
+class PaperCitations
+  def self.unique_cites(contents)
+    new(contents).unique_cites
+  end
+
+  def initialize(contents)
+    @contents = contents
+  end
+
+  def unique_cites
+    # clean citations of slashes, commas, semi-colons, possesives
+    clean_cites = @contents.scan(cite_match).map do |c|
+      c[0].gsub(/[\(\),;]|([\'\’]s)/, '').gsub(/[\'\’]\s/, ' ')
+    end
+
+    # create hash of citations (key) with counts (value)
+    Hash[clean_cites.group_by { |c| c }.map { |k, v| [k, v.count] }]
+  end
+
+  private
+
+  def prefix
+    '(([dD]e|[vV]an[ ]?[dD]er)[ ]?)'
+  end
+
+  def name
+    "(#{prefix}?[A-Z][[:alpha:]\'\’\-]+)"      # name: caps, accents, 's
+  end
+
+  def cite_match
+    /(
+      #{name}{1}                                    # first author
+      ([ ]and[ ]#{name} | ([ ]et[ ]al.){1})?        # remaining authors
+      ([\'\’]s|s[\'\’])?                            # possessive form
+      ([ ][\(]?[1-2][0-9]{3}[a-z]?[,\)\;])          # year
+    )/x
+  end
+end

metadata

@@ -0,0 +1,74 @@
+--- !ruby/object:Gem::Specification
+name: citesight
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Soumya Ray
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-05-27 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: minitest
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: minitest-rg
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  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 paper
+email: soumya.ray@gmail.com
+executables:
+- citesight
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/citesight.rb
+- bin/citesight
+homepage: https://github.com/soumyaray/citesight
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.1.11
+signing_key: 
+specification_version: 4
+summary: Citation extractor and analyzer
+test_files: []