cr.rb 4.1.4 → 4.1.5

data/lib/cr.rb

@@ -1,36 +1,546 @@
 # ------------------------------------------------------
 # File          : cr.rb
-# Authors       : ccmywish <ccmywish@qq.com>
+# Authors       : Aoran Zeng <ccmywish@qq.com>
 # Created on    : <2022-04-15>
-# Last modified : <2023-03-22>
+# Last modified : <2023-05-09>
 #
 # cr:
 #
 #   This file is the lib of `cr.rb``
 # ------------------------------------------------------
 
-require 'cr/version'
-# require_relative "#{__dir__}/cr/version"
+require 'colorator'
+require_relative 'cr/version'
 
-module CrypticResolver
+class String
+  def underline = "\e[4m#{self}\e[0m"
+end
+
+class CrypticResolver::Resolver
+
+  require_relative 'cr/counter'
+
+  # Notice that we only update the Default library, not Extra library
+  def update_dicts
+
+    @counter.count_def_lib(display: false)
+    old_wc = @counter.word_count_of_def_lib
+
+    puts "cr: Updating all dicts in Default library..."
+
+    begin
+    Dir.chdir DEFAULT_LIB_PATH do
+      if Gem.win_platform?
+        # Windows doesn't have fork
+        Dir.children(DEFAULT_LIB_PATH).each do |dict|
+          next if File.file? dict
+          puts "cr: Wait to update #{dict}..."
+          `git -C ./#{dict} pull -q`
+        end
+      else
+        # *nix
+        Dir.children(DEFAULT_LIB_PATH).each do |dict|
+          next if File.file? dict
+          fork do
+            puts "cr: Wait to update #{dict}..."
+          `git -C ./#{dict} pull -q`
+          end
+        end
+        Process.waitall
+      end # end if/else
+    end
+
+    rescue Interrupt
+    abort "cr: Cancel update"
+    end
+
+    puts "cr: Update done"
+
+    # clear
+    @counter.word_count_of_def_lib = 0
+    # recount
+    @counter.count_def_lib(display: false)
+
+    new_wc = @counter.word_count_of_def_lib
+    puts ; puts "#{new_wc - old_wc} words added in Default library"
+  end
+
+
+  def add_dict(repo)
+    if repo.nil?
+      abort "cr: Need an argument!".bold.red
+    end
+
+    # Simplify adding dictionary
+    if !repo.start_with?("https://") and !repo.start_with?("git@")
+      if repo.include?('/')
+        repo = "https://github.com/#{repo}.git"
+      else
+        repo = "https://github.com/cryptic-resolver/cryptic_#{repo}.git"
+      end
+    end
+
+    begin
+    puts "cr: Adding new dictionary..."
+    `git -C #{DEFAULT_LIB_PATH} clone #{repo} -q`
+    rescue Interrupt
+    abort "cr: Cancel add dict"
+    end
+
+    puts "cr: Add new dictionary done"
+
+    # github/com/ccmywish/ruby_knowledge(.git)
+    dict = repo.split('/')[-1].delete_suffix('.git')
+    wc = @counter.count_dict_words(DEFAULT_LIB_PATH ,dict)
+    puts ; puts "#{wc} words added"
+  end
+
+
+  def del_dict(repo)
+    if repo.nil?
+      abort "cr: Need an argument!".bold.red
+    end
+    Dir.chdir DEFAULT_LIB_PATH do
+      begin
+      # Dir.rmdir repo        # Can't rm a filled dir
+      # FileUtils.rmdir repo  # Can't rm a filled dir
+      FileUtils.rm_rf repo
+      puts "cr: Delete dictionary #{repo.bold.green} done"
+      rescue Exception => e
+      puts "cr: #{e}".bold.red
+      list_dicts
+      end
+    end
+  end
+
+
+  def load_sheet(library, dict, sheet_name)
+    file = library + "/#{dict}/#{sheet_name}.toml"
+    if File.exist? file
+      return Tomlrb.load_file file # gem 'tomlrb'
+      # return TOML.load_file file # gem 'toml'
+    else nil end
+  end
+
+
+=begin
+Pretty print the info of the given word
+
+A info looks like this
+  emacs = {
+    name = "Emacs"
+    desc = "edit macros"
+    more = "a feature-rich editor"
+    see  = ["Vim"]
+  }
+
+@param info [Hash] the information of the given word (mapped to a keyword in TOML)
+=end
+  def pp_info(info)
+    name = info['name'] || "No name!".red  # keyword `or` is invalid here in Ruby
+
+    desc = info['desc']
+    more = info['more']
+
+    if desc
+      puts "\n  #{name}: #{desc}"
+      print "\n  ",more,"\n" if more
+    else
+      puts "\n  #{name}"
+      print "\n  ",more,"\n" if more
+    end
+
+    if see_also = info['see']
+      print "\n", "SEE ALSO ".magenta
+      if see_also.is_a?(Array)
+        last_ndx = see_also.size - 1
+        see_also.each_with_index do |x,i|
+          if last_ndx == i
+            print x.underline  # Last word doesn't show space
+          else
+            print x.underline, ' '
+          end
+        end
+      else
+        print see_also.underline
+      end
+      puts
+    end
+    puts
+  end
+
+
+  # Print default cryptic_ dicts
+  def pp_dict(dict)
+    puts "From: #{dict}".green
+  end
+
+
+=begin
+  Used for synonym jump
+  Because we absolutely jump to a must-have word
+  So we can directly lookup to it
+
+  Notice that, we must jump to a specific word definition
+  So in the toml file, you must specify the precise word.
+  If it has multiple meanings, for example
+
+  [blah]
+  same = "xdg"  # this is wrong, because xdg has multiple
+                # definitions, and all of them specify a
+                # category
+
+  [blah]
+  same = "XDG downloader =>xdg.Download" # this is correct
+
+  [blah]
+  name = "BlaH" # You may want to display a better name first
+  same = "XDG downloader =>xdg.Download" # this is correct
+=end
+  def pp_same_info(library, dict, word, cache_content, same_key, own_name)
+
+    # If it's a synonym for anther word,
+    # we should lookup into this dict again, but maybe with a different file
+
+    # file name
+    x = word.chr.downcase
+
+=begin
+    dictionary maintainer must obey the rule for xxx.yyy word:
+    xxx should be lower case
+    yyy can be any case
+
+    Because yyy should clearly explain the category info, IBM is better than ibm
+    Implementation should not be too simple if we want to stress the function we
+    expect.
+
+    'jump to' will output to user, so this is important not only inside our sheet.
+
+      same = "XDM downloader=>xdm.Download"
+
+    We split 'same' key into two parts via spaceship symbol `=>`, first part will
+    output to user, the second part is for internal jump.
+=end
+    jump_to, same = same_key.split("=>")
+    same =  jump_to if same.nil?
+
+    unless own_name
+      own_name = word
+    end
+    puts own_name.bold.blue + ' redirects to ' + jump_to.bold.blue
+
+=begin
+    As '.' is used to delimit a word and a category, what if
+    we jump to a dotted word?
+
+      [eg]
+      same = "e.g." # this must lead to a wrong resolution to
+                    # word 'e', category 'g'
+
+      All you need is to be like this:
+
+      [eg]
+      same = "'e.g.'" # cr will notice the single quote
+=end
+    if same =~ /^'(.*)'$/
+      same, category = $1, nil
+    else
+      same, category = same.split('.')
+    end
+
+    if same.chr == x
+      # No need to load another dictionary if match
+      sheet_content = cache_content
+    else
+      sheet_content = load_sheet(library, dict, same.chr.downcase)
+    end
+
+    if category.nil?
+      info = sheet_content[same]
+    else
+      info = sheet_content[same][category]
+    end
+
+    if info.nil?
+      puts "WARN: Synonym jumps to the wrong place `#{same}`,
+      Please consider fixing this in `#{x}.toml` of the dictionary `#{dict}`".red
+      # exit
+      return false
+    # double or more jumps
+    elsif same_key = info['same']
+      own_name = info['name']
+      return pp_same_info(library, dict, same, cache_content, same_key, own_name)
+    else
+      pp_info(info)
+      return true
+    end
+  end
+
+
+
+  #
+  # Lookup the given word in a sheet (a toml file) and also print.
+  # The core idea is that:
+  #
+  # 1. if the word is `same` with another synonym, it will directly jump to
+  #   a word in this dictionary, but maybe a different sheet.
+  #
+  # 2. load the toml file and check the given word
+  #   2.1 with no category specifier
+  #       [abcd]
+  #   2.2 with category specifier
+  #       [abcd.tYPe]
+  #
+  def lookup(library, dict, file, word)
+    sheet_content = load_sheet(library, dict, file)
+    return false if sheet_content.nil?
+
+    info = sheet_content[word]
+    return false if info.nil?
+
+    # Warn if the info is empty. For example:
+    #   emacs = { }
+    if info.size == 0
+      abort "WARN: Lack of everything of the given word
+      Please consider fixing this in the dict `#{dict}`".red
+    end
+
+    # Word with no category specifier
+    # We call this meaning as type 1
+    type_1_exist_flag = false
+
+    # if already jump, don't check the word itself
+    is_jump = false
+
+    # synonym info print
+    if same_key = info['same']
+      own_name = info['name']
+      pp_dict(dict)
+      pp_same_info(library, dict, word, sheet_content, same_key, own_name)
+      # It's also a type 1
+      type_1_exist_flag = true
+      is_jump = true
+    end
 
-  module Color
-    def bold(str)       "\e[1m#{str}\e[0m" end
-    def underline(str)  "\e[4m#{str}\e[0m" end
-    def red(str)        "\e[31m#{str}\e[0m" end
-    def green(str)      "\e[32m#{str}\e[0m" end
-    def yellow(str)     "\e[33m#{str}\e[0m" end
-    def blue(str)       "\e[34m#{str}\e[0m" end
-    def purple(str)     "\e[35m#{str}\e[0m" end
-    def cyan(str)       "\e[36m#{str}\e[0m" end
+    # normal info print
+    # To developer:
+    #   The word should at least has one of `desc` and `more`
+    #   But when none exists, this may not be considered wrong,
+    #   Because the type2 make the case too.
+    #
+    #   So, just ignore it, even if it's really a mistake(insignificant)
+    #   by dictionary maintainers.
+    #
+    if !is_jump && (info.has_key?('desc') || info.has_key?('more'))
+      pp_dict(dict)
+      pp_info(info)
+      type_1_exist_flag = true
+    end
+
+    # Meanings with category specifier
+    # We call this meaning as type 2
+    categories = info.keys - ["name", "desc", "more", "same", "see"]
+
+    if !categories.empty?
+
+      if type_1_exist_flag
+        print "OR".bold.blue, "\n"
+      else
+        pp_dict(dict)
+      end
+
+      categories.each do |meaning|
+        info0 = sheet_content[word][meaning]
+        if same_key = info0['same']
+          own_name = info0['name']
+          pp_same_info(library, dict, word, sheet_content, same_key, own_name)
+        else
+          pp_info(info0)
+        end
+
+        # last meaning doesn't show this separate line
+        print "OR".bold.blue, "\n" unless categories.last == meaning
+      end
+      return true
+    elsif type_1_exist_flag then return true
+    else return false end
+  end
+
+
+  #
+  # The main procedure of `cr`
+  #
+  #   1. Search the default library first
+  #   2. Search the extra library if it does exist
+  #
+  # The `search` is done via the `lookup` function. It will print
+  # the info while finding. If `lookup` always return false then
+  # means lacking of this word in our dicts. So a welcomed
+  # contribution is printed on the screen.
+  #
+  def resolve_word(word)
+
+    word = word.downcase # downcase! would lead to frozen error in Ruby 2.7.2
+    # The index is the toml file we'll look into
+    index = word.chr
+    case index
+    when '0'..'9'
+      index = '0-9'
+    end
+
+    # cache lookup's results
+    results = []
+
+    # First consider the default library
+    default = Dir.children(DEFAULT_LIB_PATH)
+    default.each do |dict|
+      results << lookup(DEFAULT_LIB_PATH,dict,index,word)
+    end
+
+    # Then is the extra library
+    if @extra_lib_path
+      extra = Dir.children(@extra_lib_path)
+      extra.each do |dict|
+        results << lookup(@extra_lib_path,dict,index,word)
+      end
+    end
+
+    unless results.include? true
+      puts <<~NotFound
+      cr: Not found anything about '#{word}'. You could try
+
+      #{"case 1: Update all dictionaries".blue}
+       #{"$ cr -u".yellow}
+      #{"case 2: List available official and feature dictionaries".blue}
+       #{"$ cr -l".yellow}
+       #{"$ cr -a repo".yellow} (Add a specific dict to default lib)
+      #{"case 3: Contribute to theses dictionaries".blue}
+       Visit: https://github.com/cryptic-resolver
+
+      NotFound
+    else return end
+  end
+
+
+  #
+  # Delegate to `search_word_internal`
+  #
+  def search_word(pattern)
+    found_or_not1 = false
+    found_or_not2 = false
+
+    found_or_not1 = search_word_internal(pattern, DEFAULT_LIB_PATH)
+    if @extra_lib_path
+      found_or_not2 = search_word_internal(pattern, @extra_lib_path)
+    end
+
+    if (found_or_not1 == false) && (found_or_not2 == false)
+      puts "cr: No words match with #{pattern.inspect}".red ; puts
+    end
+  end
+
+  #
+  # This `search_word_internal` routine is quite like `resolve_word`
+  # Notice:
+  #   We handle two cases
+  #
+  #   1. the 'pattern' is the regexp itself
+  #   2. the 'pattern' is like '/blah/'
+  #
+  #  The second is what Ruby and Perl users like to do, handle it!
+  #
+  def search_word_internal(pattern, library)
+
+    if pattern.nil?
+      abort "cr: Need an argument!".bold.red
+    end
+
+    if pattern =~ /^\/(.*)\/$/
+      regexp = %r[#$1]
+    else
+      regexp = %r[#{pattern}]
+    end
+
+    found_or_not = false
+
+    # Try to match every word in all dicts
+    Dir.children(library).each do |dict|
+
+      path = File.join(library, dict)
+      next if File.file? path
+      sheets = Dir.children(path).select do
+        _1.end_with?('.toml')
+      end
+
+      similar_words_in_a_dict = []
+
+      sheets.each do |sheet|
+        sheet_content = load_sheet(library, dict, File.basename(sheet,'.toml'))
+
+        sheet_content.keys.each do
+          if _1 =~ regexp
+            found_or_not = true
+            similar_words_in_a_dict << _1
+          end
+        end
+      # end of each sheet in a dict
+      end
+
+      unless similar_words_in_a_dict.empty?
+        pp_dict(dict)
+        require 'ls_table'
+        LsTable.ls(similar_words_in_a_dict) do |e|
+          puts e.blue
+        end
+        puts
+      end
+    end
+    return found_or_not
+  end
+
+
+  # 1. List Default library's dicts
+  # 2. List Extra   library's dicts
+  # 3. List official dicts
+  def list_dicts
+
+    def _do_the_same_thing
+      count = 0
+      Dir.children(".").each do |dict|
+        next if File.file? dict
+        count += 1
+        prefix = (count.to_s + '.').ljust 4
+        puts "  #{prefix}#{dict}"
+      end
+    end
+
+    Dir.chdir DEFAULT_LIB_PATH do
+      puts "Default library: #{DEFAULT_LIB_PATH}".bold.green
+      _do_the_same_thing
+    end
+
+    if @extra_lib_path
+      puts
+      Dir.chdir @extra_lib_path do
+        puts "Extra library: #{@extra_lib_path}".bold.green
+        _do_the_same_thing
+      end
+    end
+
+    puts ; puts "Official dictionaries: (Add it by 'cr -a xxx')".bold.green
+    puts RECOMMENDED_DICTS
+  end
+
+
+  def word_count
+    @counter.count!(display: true)
   end
 
 end
 
 
-class CrypticResolver::Resolver
 
-  include CrypticResolver::Color
+class CrypticResolver::Resolver
 
   require 'tomlrb'
   require 'fileutils'
@@ -47,21 +557,19 @@ class CrypticResolver::Resolver
     "https://github.com/cryptic-resolver/cryptic_technology"
   ]
 
-  extend CrypticResolver::Color
-
   RECOMMENDED_DICTS = <<~EOF
-  #{yellow("Default:")}
+  #{"Default:".yellow}
     common       computer    windows
     linux       technology
 
-  #{yellow("Field:")}
+  #{"Field:".yellow}
     economy     medicine   electronics
     science       math
 
-  #{yellow("Specific:")}
+  #{"Specific:".yellow}
       x86        signal
 
-  #{yellow("Feature:")}
+  #{"Feature:".yellow}
     ccmywish/CRuby-Source-Code-Dictionary
 
   EOF
@@ -164,99 +672,4 @@ class CrypticResolver::Resolver
     return false
   end
 
-
-end
-
-
-class CrypticResolver::Resolver::Counter
-
-  include CrypticResolver::Color
-
-  attr_accessor :word_count_of_two_libs,  # def_lib + extra_lib
-                :word_count_of_def_lib,
-                :word_count_of_extra_lib,
-
-                :resolver
-
-  def initialize(resolver)
-    @word_count_of_two_libs = 0
-    @word_count_of_def_lib = 0
-    @word_count_of_extra_lib = 0
-    @resolver = resolver
-  end
-
-  # a.toml
-  # b.toml
-  # ...
-  def count_dict_words(library, dict)
-    dict_dir = library + "/#{dict}"
-    wc = 0
-    Dir.children(dict_dir).each do |entry|
-      next unless entry.end_with?('.toml')
-      next if File.directory? dict_dir + "/#{entry}"
-      sheet_content = @resolver.load_sheet(library, dict, entry.delete_suffix('.toml'))
-      count = sheet_content.keys.count
-
-      wc = wc + count
-    end
-    return wc
-  end
-
-
-  # Count default library
-  def count_def_lib(display: )
-    path = CrypticResolver::Resolver::DEFAULT_LIB_PATH
-    default_lib = Dir.children path
-    unless default_lib.empty?
-      puts bold(green("Default library: "))  if display
-      default_lib.each do |s|
-        next if File.file? path + "/#{s}"
-        wc = count_dict_words(path,s)
-        @word_count_of_def_lib += wc
-        # With color, ljust not works, so we disable color
-        puts("#{wc.to_s.rjust(5)}  #{s}")   if display
-      end
-    end
-    return @word_count_of_def_lib
-  end
-
-
-  # Count extra library
-  def count_extra_lib(display: )
-    if path = @resolver.extra_lib_path
-      extra_lib = Dir.children path
-      unless extra_lib.empty?
-        puts bold(green("\nExtra library:"))  if display
-        extra_lib.each do |s|
-          next if File.file? path + "/#{s}"
-          wc = count_dict_words(path,s)
-          @word_count_of_extra_lib += wc
-          puts("#{wc.to_s.rjust(5)}  #{s}")  if display
-        end
-      end
-    end
-    return @word_count_of_extra_lib
-  end
-
-
-  def count!(display: )
-    count_def_lib(display: display)
-    count_extra_lib(display: display)
-    @word_count_of_two_libs = @word_count_of_def_lib + @word_count_of_extra_lib
-
-    if display
-    puts
-    puts "#{@word_count_of_def_lib.to_s.rjust(4)} words in Default library"
-    puts "#{@word_count_of_extra_lib.to_s.rjust(4)  } words in  Extra  library"
-    puts "#{@word_count_of_two_libs.to_s.rjust(4)    } words altogether"
-    end
-  end
-
-
-  def reset!
-    @word_count_of_two_libs = 0
-    @word_count_of_def_lib = 0
-    @word_count_of_extra_lib = 0
-  end
-
 end