bibletools 0.1.0 → 0.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b39f3947e9bf8e827ce070ad544f2f03a5b28e43554588ec99e6511aaf7c7109
-  data.tar.gz: 6d88f41a49b461adc25993c65e2abbeffc6616b9d727bab0bcc93c3570bcc824
+  metadata.gz: 5085703f655b244826701beeba91b5bbfc535a8d34f2d4fbbc842c6229ced5fa
+  data.tar.gz: 27639f898a9c016fdbaf47cbaddeae392c93500145492ea0ba51f1a2ebc82e1b
 SHA512:
-  metadata.gz: '09532d8e8b4d88ae66ffdcb7f62aec4c7887cb9b87497deb5b04206f6b3c5cbcbcffa618e2ded85a683a57f43907e8f5e30dcb0d62885481fc0a95540240cad6'
-  data.tar.gz: ec3bc05ded3b431231eaea8f2e00cde75a4da60bc8cc4fbb6fda1cd3d8312fabd1d50aa7e096742f7beb05b36e306b853b4cd92d446a9bb060333ce544614645
+  metadata.gz: a70124fe758ee62e6874e6126db903c56eb0edf8bd35f5ea7a80dae1917cc30c225191dc6eec93f35ac1f63b77dbaef461f25398a5a0db90684ece938aa427c8
+  data.tar.gz: 861ed48ae15f6d7516d98fac70e76b2b2bc6633c90f56193b7ca4f5f8fafbeee07e12ba8adbef23cb57119ef7c0a0c9ff31e7ec49a91383f695c19955dbfc9fc

data/lib/bibletools.rb

@@ -6,6 +6,85 @@ require 'rexle'
 require 'yawc'
 require 'nokorexi'
 
+require 'cerecvoice2019'
+require 'english_spellchecker'
+
+
+BOOKS = [
+  "Genesis",
+  "Exodus",
+  "Leviticus",                                                                  
+  "Numbers",
+  "Deuteronomy",
+  "Joshua",
+  "Judges",
+  "Ruth", 
+  "1 Samuel",
+  "2 Samuel",
+  "1 Kings",
+  "2 Kings",
+  "1 Chronicles",
+  "2 Chronicles",
+  "Ezra",
+  "Nehemiah",
+  "Tobit",
+  "Judith",
+  "Esther",
+  "1 Maccabees",
+  "2 Maccabees",
+  "Job",
+  "Psalms",
+  "The Proverbs",
+  "Ecclesiastes",
+  "The Song of Songs",
+  "Wisdom",
+  "Ecclesiasticus / Sirach",
+  "Isaiah",
+  "Jeremiah",
+  "Lamentations",
+  "Baruch",
+  "Ezekiel",
+  "Daniel",
+  "Hosea",
+  "Joel",
+  "Amos",
+  "Obadiah",
+  "Jonah",
+  "Micah",
+  "Nahum",
+  "Habakkuk",
+  "Zephaniah",
+  "Haggai",
+  "Zechariah",
+  "Malachi",
+  "Matthew",                                                                   
+  "Mark",  
+  "Luke",
+  "John",
+  "Acts of Apostles",
+  "Romans",
+  "1 Corinthians", 
+  "2 Corinthians",
+  "Galatians", 
+  "Ephesians",
+  "Philippians",
+  "Colossians",
+  "1 Thessalonians",
+  "2 Thessalonians",
+  "1 Timothy",
+  "2 Timothy",
+  "Titus",
+  "Philemon",
+  "Hebrews",
+  "James",
+  "1 Peter",
+  "2 Peter",
+  "1 John",
+  "2 John",
+  "3 John",
+  "Jude",
+  "Revelation"
+]
 
 module BibleTools
 
@@ -20,94 +99,17 @@ module BibleTools
       
       baseurl = url
 
-      books = ["Genesis",
-       "Exodus",
-       "Leviticus",                                                                  
-       "Numbers",
-       "Deuteronomy",
-       "Joshua",
-       "Judges",
-       "Ruth", 
-       "1 Samuel",
-       "2 Samuel",
-       "1 Kings",
-       "2 Kings",
-       "1 Chronicles",
-       "2 Chronicles",
-       "Ezra",
-       "Nehemiah",
-       "Tobit",
-       "Judith",
-       "Esther",
-       "1 Maccabees",
-       "2 Maccabees",
-       "Job",
-       "Psalms",
-       "The Proverbs",
-       "Ecclesiastes",
-       "The Song of Songs",
-       "Wisdom",
-       "Ecclesiasticus / Sirach",
-       "Isaiah",
-       "Jeremiah",
-       "Lamentations",
-       "Baruch",
-       "Ezekiel",
-       "Daniel",
-       "Hosea",
-       "Joel",
-       "Amos",
-       "Obadiah",
-       "Jonah",
-       "Micah",
-       "Nahum",
-       "Habakkuk",
-       "Zephaniah",
-       "Haggai",
-       "Zechariah",
-       "Malachi",
-       "Matthew",                                                                   
-       "Mark",  
-       "Luke",
-       "John",
-       "Acts of Apostles",
-       "Romans",
-       "1 Corinthians", 
-       "2 Corinthians",
-       "Galatians", 
-       "Ephesians",
-       "Philippians",
-       "Colossians",
-       "1 Thessalonians",
-       "2 Thessalonians",
-       "1 Timothy",
-       "2 Timothy",
-       "Titus",
-       "Philemon",
-       "Hebrews",
-       "James",
-       "1 Peter",
-       "2 Peter",
-       "1 John",
-       "2 John",
-       "3 John",
-       "Jude",
-       "Revelation"]
-
       id = if book.is_a? String then
-        (books.index(book) + 1).to_s
+        (BOOKS.index(book) + 1).to_s
       else
         book.to_s
       end
 
-
-
       url2 = baseurl + '?id=' + id
       doc = Nokorexi.new(url2).to_doc
       e = doc.root.at_css('.pagination')
       endchapter = e.xpath('li')[-2].text('a')
 
-
       book = Rexle.new('<book/>')
 
       (1..endchapter.to_i).each do |chp|
@@ -143,39 +145,128 @@ module BibleTools
     def to_doc
       @doc
     end
+    
+    def save(title)
+      
+      filename = title.downcase.gsub(/\s+/,'-') + '.xml'      
+      File.write filename, @doc.root.xml(pretty: true)
+      puts filename +' saved!'
+      
+    end
 
   end
   
   class Analyse
     
-    def initialize(bible_obj, debug: false)
+    def initialize(bible_obj, tts: nil, debug: false)
       
       @debug = debug
       
       if bible_obj then
+        
         @doc = bible_obj.to_doc
-        @verses = @doc.root.xpath('chapter/verse')
+        @check = EnglishSpellcheck.new debug: false #verbose: false
+        
+        if tts then
+          @cerevoice = Cerecvoice2019.new accountid: tts[:account], 
+              password: tts[:password], voice: tts[:voice] || 'Stuart'
+        end
+        
+      end
+            
+      
+    end
+    
+    #   Automatically select verses; verses are selected based upon word 
+    # frequency 
+    #
+    def asr()
+      
+      r = assoc_r()
+      e = r[1].root.element('chapter/verse')
+      verse_no = e.attributes[:no]
+      chptr_no = e.parent.attributes[:no]
+      
+      # we are backtracking the results to hopefully find a new branch to explore
+      trail = r.first
+
+      a, trail2 = mine_words(r, trail, level: 2)
+      a << [chptr_no, verse_no,  e.text, 1]
+
+      h = a.group_by(&:first)
+      a2 = h.sort_by {|x, _| x.to_i}
+      @verses = a2.map do |chapter, verses|
+        [chapter, verses.sort_by {|_,x| x.to_i}]
       end
+        
+      [@verses, trail2]
+    end    
+    
+    # find the words associated with a given keyword
+    # 
+    def assoc(keyword, doc=@doc)
+      doc2 = self.search keyword, doc
+      [self.words(doc: doc2), doc2]
+    end
+    
+    def assoc_r(keyword=words().keys.first, doc=@doc, list=[], results=[])
+      
+      result = assoc(keyword, doc)
+      results << result
+      h, doc2 = result
+      new_keyword = h.keys.find {|x| not list.include? x}
+      return [list, doc2, results] unless new_keyword
+      
+      list << new_keyword
+      assoc_r(new_keyword, doc2, list, results)
       
     end
     
-    def wordtally()
+    def custom_words(doc: @doc, level: 3)
+            
+      a = wordtally(level: level, doc: doc).keys.map do |word|
+        puts 'word: ' + word.inspect if @debug
+        [word, @check.spelling(word, verbose: false)]
+      end
       
-      return unless @doc
-      Yawc.new(@doc.root.plaintext).to_h
+      a2 = a.select do |w, x| 
+        if x.is_a? String then
+          w != x
+        elsif x.is_a? Array
+          not (w == x.join or w == x.first)
+        else
+          true
+        end
+      end
+      a2.map(&:first)
+      
+    end
+    
+    def wordtally(level: 2, doc: @doc)
+      
+      return unless doc
+      Yawc.new(doc.root.plaintext, level: level).to_h
       
     end
     
     alias words wordtally
     
     def read(chapter, verse)
-      @doc.root.element("chapter[@no='#{chapter}']/verse[@no='#{verse}']")
+      
+      if verse.is_a? Integer then
+        [@doc.root.element("chapter[@no='#{chapter}']/verse[@no='#{verse}']")]
+      elsif verse.is_a? Array
+        verse.map do |x|
+          @doc.root.element("chapter[@no='#{chapter}']/verse[@no='#{x}']")
+        end
+      end
     end
 
 
-    def search(keyword)
+    def search(keyword, doc=@doc)
       
-      a = @verses.select {|x| x.text =~ /#{keyword}/}
+      verses = doc.root.xpath('chapter/verse')
+      a = verses.select {|x| x.text =~ /#{keyword}/i}
       
       a2 = a.map.with_index do |verse, i|
         txt = verse.text
@@ -185,9 +276,10 @@ module BibleTools
         n = 0
         chapter = verse.parent.attributes[:no]
         
-        until txt.rstrip[-1] == '.' or n > 2 do
+        until txt.rstrip[/[\.\?!]\W*$/] or n > 2 do
           n +=1
-          nverse = read(chapter, verse.attributes[:no].to_i + n)
+          nverse = read(chapter, verse.attributes[:no].to_i + n)[0]
+          puts 'nverse: ' + nverse&.text.inspect if @debug
           txt += nverse.text
         end 
         
@@ -199,11 +291,118 @@ module BibleTools
         [verse.parent.attributes[:no], verse.attributes[:no], txt]
       end
 
-      h1 = a2.group_by(&:first)
+      # group by chapter
+      #
+      h = a2.group_by(&:first)      
+      
+      doc = Rexle.new('<book/>')
+
+      h.each do |key, value|
+
+        echapter = Rexle::Element.new('chapter', attributes: {no: key})
+
+        value.each do |_, verse_no, verse|
+          everse = Rexle::Element.new('verse', attributes: {no: verse_no}).add_text verse
+          echapter.add everse
+        end
+
+        doc.root.add echapter
+
+      end
+      
+      return doc
+      
+    end
+    
+    def tts(chapter, verse)
+      
+      verses = read(chapter, verse).map(&:text).join(' ')
+      @cerevoice.tts verses, out: 'tts.ogg'      
+      
+    end
+    
+    def verses(level: nil, html: false, title: nil)                  
+
+      a = @verses.map {|_, body| body}.flatten(1)
+      a1 = level ? a.select {|c,v,t,l| l <= level} : a.map {|c,v,t,l| [c,v,t]}
+      a2 = a1.map {|c,v,t,l| [c,v,t]}.uniq
+      h = a2.group_by(&:first)
+      
+      puts 'html: ' + html.inspect
+      return h unless html == true
+      
+      doc = Rexle.new('<html/>')
+
+
+      head = Rexle::Element.new('head')
+      head.add(Rexle::Element.new('h1').add_text(title) ) if title
+
+      style = Rexle::Element.new('style')
+      style.add_text '     ins {font-size: 0.8em;padding: 0.8em;}'
+      head.add style
+
+      doc.root.add head
+      body = Rexle::Element.new('body')
+
+      h.each do |chptr, verses|
+
+        h2 = Rexle::Element.new('h2').add_text chptr
+        body.add h2
+
+        verses.each do |cno, vno, text|
+
+          ins = Rexle::Element.new('ins').add_text vno
+          para = Rexle::Element.new('p')
+          para.add ins
+          para.add_text text
+          body.add para
+          
+        end
+        
+      end
+      
+      doc.root.add body
+
+      doc.root.xml pretty: true
+      
     end
     
+    private
+
+    def mine_words(r, trail, verses=[], level: 0)
+
+      i = 0
+      found = []
+      (i-=1; found = r.last[i][0].keys - trail) until found.any?
+      r2 = r.last[i][0].keys - trail
+      
+      a3 = r2.map do |x|
+        
+        _, doc_verse = assoc_r x, r.last[i][1]
+        chaptr_no = doc_verse.root.element('chapter').attributes[:no]
+        everse = doc_verse.root.element('chapter/verse')
+        verse_no = everse.attributes[:no]
+        
+        [chaptr_no, verse_no, everse.text, level]
+        
+      end.uniq
+
+      puts 'verses: ' + verses.inspect if @debug
+      
+      verses.concat a3
+      verses.uniq!
+      trail.concat r2
+      
+      puts 'trail.length: ' + trail.length.inspect if @debug
+      puts 'verses.length: ' + verses.length.inspect if @debug
+
+      if verses.length < 30 or trail.length < 100
+        mine_words(r, trail, verses, level: level+1)
+      else
+        [verses, trail]
+      end
+    end    
     
   end
 
 end
-

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: bibletools
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.1
 platform: ruby
 authors:
 - James Robertson
@@ -36,7 +36,7 @@ cert_chain:
   hIlyeyH9vsaJLam1TYsSqh3KUBmRCIvc3XELg2MAlggTrHuCjBPdYXqjIOFvO6HZ
   Z1XDnMf4RhcEPF4AU3Q+Fefsw4qKRWH3YUw=
   -----END CERTIFICATE-----
-date: 2022-11-01 00:00:00.000000000 Z
+date: 2022-11-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokorexi
@@ -78,6 +78,46 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 0.3.0
+- !ruby/object:Gem::Dependency
+  name: cerecvoice2019
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.0.1
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.0.1
+- !ruby/object:Gem::Dependency
+  name: english_spellchecker
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.3'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.3.1
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.3'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.3.1
 description: 
 email: digital.robertson@gmail.com
 executables: []