flnews_post_proc 1.7

data/lib/body.rb

@@ -0,0 +1,364 @@
+#encoding: UTF-8
+
+=begin
+  /***************************************************************************
+   *   2023-2024, Michael Uplawski <michael.uplawski@uplawski.eu>            *
+   *   This program is free software; you can redistribute it and/or modify  *
+   *   it under the terms of the WTFPL 2.0 or later, see                     * 
+   *   http://www.wtfpl.net/about/                                           *
+   *                                                                         *
+   *   This program is distributed in the hope that it will be useful,       *
+   *   but WITHOUT ANY WARRANTY; without even the implied warranty of        *
+   *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.                  *
+   *                                                                         *
+   ***************************************************************************/
+=end
+
+require_relative 'basic_logging'
+  require_relative 'configuration'
+
+WRAP_LENGTH = 65
+# Endow the String class with a wrap function.
+# This is not yet applicable to the message body, itself.
+class String
+  include BasicLogging
+  public
+  # wraps at length, returns the result
+  def wrap! (length = WRAP_LENGTH, indent = 0)
+    max = length
+
+    line = 0
+    out = [""]
+
+    self.gsub!(/^\r\n/, "[rnrn]")
+    self.gsub!("\r\n", " ")
+    self.gsub!("  ", " ")
+
+    words = self.split(" ")
+
+    while !words.empty?
+      word = words.shift.strip
+      break if not word
+      if out[line].length + word.length > max
+        out[line].squeeze!(" ")
+        line += 1
+        out[line] = (line > 0 ? " " * indent : "")
+      end
+      out[line] << word + " "
+    end
+    out = out.join("\r\n") 
+    out.gsub!("[rnrn]", "\r\n\r\n")
+    self.replace( out)
+  end
+
+end
+
+# An object of this class represents the body of a news-article.
+# It processes the original text and changes some details:
+#    intro-lines
+#    signatures
+#    URI formats
+#    Footnotes (and/or a list of references) are created
+class Body
+  # a class-level configuration instance.
+  @@config = Configuration.instance
+  include BasicLogging  
+
+  # reads the body text of the article
+  def initialize(article_text)
+    debug 'body intialize'
+    # for simplicity.
+    line = nil
+    #  transform the article into an array.
+    line_array = article_text.split($LN)
+    # keep only from the first after an empty line  ''
+    start_index = line_array.index('')
+
+    # ... to the end of the current array (all that follows '').
+    @lines = line_array.slice(start_index + 1, line_array.size)
+    debug('initialize(): body lines are ' << @lines.inspect)
+  end
+
+  # If so configured, replace an eventual followup-intro by the one
+  # configured for a group. This may depend on other conditions and 
+  # must be triggered explicitly.
+  def set_intro(intro)
+    return if !intro || intro.empty? || @@config.no_intro
+
+    # name of the previous poster
+    fup_name = nil
+    # the current newsgroup 
+    fup_group = nil
+
+    debug('FUP_NAME is ' << @@config.FUP_NAME)
+    debug('FUP_GROUP is ' << @@config.FUP_GROUP)
+    # The expressions which allow the identification of both
+    # in the current article.
+    fn = @@config.FUP_NAME
+    fg = @@config.FUP_GROUP
+
+    # Okay, this is called parsing, when it is well done.
+    # I just try and am happy when it works.
+    @lines.each_with_index do |line, i|
+      # find the name in the intro-line
+      if !fn.strip.empty? && !line.strip.empty? && !fup_name
+        # match a name
+        fup_name = line.match(Regexp.new(fn) ) do |md|
+          # debug("\tmatch: " << md.to_s)
+          md.length == 2 ? md[1] : md[0]
+        end
+        debug("\tfup_name: " << fup_name.to_s) 
+
+        if !fg.strip.empty? && !fup_group
+          # match a group
+          fup_group = line.match(Regexp.new(fg) ) do |md| 
+            debug("\tmatch: " << md.to_s)
+            md.length == 2 ? md[1] : nil 
+          end 
+        end
+        debug "group is " << fup_group.to_s
+
+        # All that follows depends on the presence of a name
+        # in the intro-string.
+        if fup_name && !fup_name.strip.empty?
+          # keep the current intro for later
+          ointro = line
+          line = ''
+          while line.strip.empty?
+            i = i.next
+            line = @lines[i]
+          end
+          # check if there is a quote, at all
+          if(line.start_with?('>'))
+            debug("\tfound intro " << ointro)
+            # variables are part of the $intro.
+            # Do substitutions.
+            intro.sub!('%fup_name%', fup_name) if fup_name 
+            intro.sub!('%fup_group%', fup_group)  if fup_group 
+            debug("\tsetting intro " << intro.to_s)
+
+            # exchange original intro-line against the new one
+            @lines[@lines.index(ointro)] = intro.strip 
+            # looked complicated because it is.
+
+            # keep this line for reference.
+            @intro = intro
+          else
+            wmsg = 'Text following the intro is not a citation!'
+            wmsg << "\n\tIntro will NOT be modified: \"#{ointro}\"!"
+            warn wmsg
+          end
+        end
+      end # fn.strip.empty?
+    end # lines.each_with_index
+  end
+
+  def set_signature(signature)
+    # unless no changes requested.
+    if signature && !signature.empty? 
+      # remove any signature(s) from 
+      # the current article
+      sigpos = @lines.index('-- ')
+      debug('found signature in line ' << sigpos.to_s) if sigpos
+      @lines = @lines.slice(0, sigpos ) if sigpos
+      debug('setting signature ' << signature) if signature
+      @lines << "-- " << signature if signature
+    end
+  end
+
+  def join 
+    new_body = @lines.join($LN) << $LN
+    # replace ellipsis 
+    new_body.force_encoding('utf-8').gsub!('...', '…')
+    return new_body
+  end
+
+  # ------->
+  # TODO : Concentrate all URL/URI munging functionality in 1 or 2 helper
+  # classes, e.g. Body::News and Body::Http
+  # <------ 
+
+  # Verify and possibly correct links in the post.
+  # Simple.
+  def handle_uris() 
+    # Determine here or elsewhere if URLs shall be verified.
+    # Default is no. nil or '' do qualify as default.
+    debug 'verify URLs ? ' << @@config.VFY_URLS.to_s
+    if @@config.VFY_URLS
+      debug 'verifying URLs'
+      @lines.each_with_index do | l, i | 
+        # leave cited lines as they are.
+        if !l.start_with?( '>') 
+=begin Currently Unused
+          # IMPORTANT --------------------------------->
+          # IT IS HENCEFORTH PROHIBITED TO WRITE AN EMAIL-ADDRESS
+          # IN THE BODY OF A NEWS-POST AND TO NOT PREPEND IT WITH
+          # mailto:
+          #  ... Because I do not know what to do in these cases,
+          #  and I am the boss!
+          #  <----------------------------
+
+          # BUGFIX : Urls with @
+          # BUGFIX : email address in the intro 
+          new_line = handle_news(l) if !@intro || l.strip != @intro.strip
+=end
+          # http(s)
+          if l.include?('http')
+            new_line = handle_http(l)
+          end #if http
+          @lines[i] = new_line if new_line
+        end # if '>'
+      end # @lines.each_with_index
+    end # if verify
+  end
+  #
+  # extract URL or other stuff, if configured for footnotes, 
+  def handle_references()
+    # a symbol or string to mark the beginning an ending of a future footnote.
+    ref_delim = @@config.REFERENCES_DELIMITER
+    debug('references delimiter is ' << ref_delim)
+    references = Array.new
+    body = @lines.join($LN)
+    if ref_delim && !ref_delim.strip.empty? 
+      unless ref_delim == ref_delim.reverse
+        ref_delim.strip!
+        ref_rx = Regexp.new(ref_delim.dup << ".*?" << ref_delim.reverse, Regexp::MULTILINE) 
+        debug('ref_rx is ' << ref_rx.to_s)
+        index = 0
+        # I cannot work with an array, here, and apply the pattern
+        # to the whole body, over multiple lines, if need be.
+        begin
+          ref = body.match(ref_rx )
+          debug("found reference " << ref.to_s << " (length: " << (ref ? ref.to_s.size.to_s : '0') << ")") if ref && ref.length > 0
+          if ref 
+            #  ... This is some presentation thing and I think
+            #  it works, too.
+            r = ref[0].gsub(/[ \t]+/, ' ').strip
+            r.gsub!("\n", "\n   ") 
+            references << r  
+            index += 1
+            body.gsub!(ref[0], format(@@config.REFERENCE_FORMAT, index.to_s ))
+          end
+        end until ref == nil
+        debug("all references found:\n" << references.join('\n')) if !references.empty?
+        # re-wrap body
+        # ATTN! Does not work!
+        # body.wrap!(WRAP_LENGTH)
+
+      else
+        msg = 'The References Delimiter is the same in its reversed form.'
+        msg << "#{$LN}Cannot handle references or footnotes!"
+        error(msg)
+      end
+
+      if(references && !references.empty?) 
+        # a line, separating the footnotes from the body of the article
+        body << $LN << @@config.REFERENCES_SEPARATOR << $LN
+        references.each_with_index do |r, i| 
+          r = r.gsub(ref_delim, '').gsub(ref_delim.reverse,'')
+          num = (i + 1).to_s << ") "
+          r.strip!
+          debug 'reference ' << r
+          body << num << r <<  $LN
+        end
+      else
+        debug('no refences found')
+      end
+    end
+    @lines = body.split($LN)
+    @encoding = body.encoding
+  end
+  private 
+
+  # handle <news:..@...>
+  def handle_news(l)
+    # ... “Do not trust nobody!”
+    if l.include?('@') && ! l.include?('http:') && ! l.include?('https:') 
+      # And I forgot how this works. It does.
+      l.split.collect do |ele| 
+        # angular brackets are there alright. Hurra. 
+        url = ele.match(/\<(.*)\>/)
+        unless !url
+          debug(' with angular brackets: ' << url.inspect)
+          l = l.sub(url[1], 'news:' << url[1]) if !url[1].start_with?('mailto') && !url[1].start_with?('news')
+        else
+          url = ele.match(/\S+@(\.|[^\p{P}]+)\S+/)
+          if url
+            debug(' withOUT angular brackets: ' << url.inspect)
+            if !url[0].start_with?('mailto') && !url[0].start_with?('news')
+              l.sub!(url[0], 'news:' << url[0] )          
+            end
+            l.sub!('news:', '<news:')
+            l.sub!(url[0], url[0] << '>')
+          end
+        end
+      end # url_strs = collect
+      debug(' news: line is now ' << l)
+      return l
+    end
+  end
+
+  def handle_http(l)
+    debug('handle http')
+    l_array = l.split
+    url_strs = l_array.collect do |ele| 
+      if ele.strip.include?('http') && ele.match(/https?:\S+/)
+        ele.lstrip! 
+        # ensure last character is alphanumeric or '>' or '/'
+        until ele.match(/([[:alnum:]]|>|\/)$/) do 
+          ele.chop!
+        end
+        ele
+      end
+    end.compact
+    debug('url_strs: ' << url_strs.to_s)
+    url_strs.each do |str|
+      # keep str for final replacement
+      nstr = str.dup
+      debug( 'try to match http-link')
+      begin
+        # missing slashes ?
+        match = str.match(/(https?):(\/*)/)
+        if (!match || match.length != 3 || match[2] != '//')
+          debug 'cannot find (enough) slashes in the URL'
+          if str.match(/https:/)
+            nstr = str.sub(/https:\/*/, "https://")
+          elsif str.match(/http:/)
+            nstr = str.sub(/http:\/*/, "http://") 
+          end
+          debug ' with slashes: ' << nstr
+        end
+        # missing angular brackets
+        nstr.sub!(/http/, "<http") if !nstr.include?( "<http") 
+        nstr << '>' if !nstr.end_with? '>'
+        debug ' matching nstr ' << nstr
+        match = nstr.match(/\<?(https?:\/\/[^\>]*)\>?/)
+        # get the darned URL already!
+        url = match[1] if match
+        debug 'url is ' << url
+        #  nstr = with_angular_brackets(nstr, url)
+        l.sub!(str, nstr)
+      rescue Exception => ex
+        # this had looked intelligent, once.
+        line = __LINE__
+        msg = "Line #{line}: Cannot match the url " << str << ' against the Regexp (' << ex.message << ')'
+        msg << "\nAborting. Bye"
+        fatal msg
+        STDERR.puts msg
+        exit false
+      end
+    end
+
+    return l
+  end
+end
+# ------- TEST --------
+
+puts "eins zwei drei vier fünf sechs sieben acht neun zehn\n
+eins zwei drei vier fünf sechs sieben acht neun zehn\n
+\n
+eins zwei drei vier fünf sechs sieben acht neun zehn\n
+eins zwei drei vier fünf sechs sieben acht neun zehn".wrap!(20)
+
+# Ω
+

data/lib/color_output.rb

@@ -0,0 +1,65 @@
+#encoding: UTF-8
+=begin
+/***************************************************************************
+ *   2023-2024, Michael Uplawski <michael.uplawski@uplawski.eu>            *
+ *   This program is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the WTFPL 2.0 or later, see                     * 
+ *   http://www.wtfpl.net/about/                                           *
+ *                                                                         *
+ *   This program is distributed in the hope that it will be useful,       *
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of        *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.                  *
+ *                                                                         *
+ ***************************************************************************/
+=end
+
+# Functions to apply colors to terminal output.
+# This is stolen from the Internet and I have lost track of my own additions
+# and modifications.
+
+COLORS = {:default => 9, :black => 0, :red => 1, :green => 2, :yellow => 3, :blue => 4, :purple => 5, :cyan => 6, :white => 7 }
+
+BG = 4
+FG = 3
+REGULAR = 0
+BOLD = 1
+UNDERLINE = 4
+BLINK = 5
+SWAP = 7
+NEUTRAL = 0
+
+STYLES = {:regular => REGULAR, :bold => BOLD, :underline => UNDERLINE, :blink => BLINK, :swap => SWAP, :neutral => NEUTRAL}
+
+# Colorizes the given text. Color-code is either an escape-sequence or one of
+# the symbols representing color-names in the COLORS hash.
+def colorize(text, color_code)
+  if (COLORS.keys.include?(color_code) )
+    "\033[3#{COLORS[color_code]}m#{text}\033[0m"
+  else
+    "#{color_code}#{text}\033[0m"
+  end
+end
+
+def style(text, style_code)
+  "#{style_code}#{text}\033[0m"
+end
+# a function which allows to manipulate every known aspect of the ansi-output.
+def colored_output(output_text, fg_color = :default, bg_color = :default, style = :regular , mode = :neutral )
+  "\033[%i;%i;%i%i;%i%im%s\033[0m" %[STYLES[mode.to_sym], STYLES[style.to_sym], FG, COLORS[fg_color.to_sym], BG, COLORS[bg_color.to_sym], output_text]
+end
+
+# convenience functions
+def red(text); colorize(text, "\033[31m"); end
+def green(text); colorize(text, "\033[32m"); end
+def yellow(text); colorize(text, "\033[33m"); end
+def purple(text); colorize(text, "\033[35m"); end
+def cyan(text); colorize(text, "\033[36m"); end
+def blue(text); colorize(text, "\033[34m"); end
+def white(text); colorize(text, "\033[37m"); end
+
+def black_on_white(text); colorize(colorize(text, "\033[30m"), "\033[47m");end
+def white_on_black(text); colorize(colorize(text, "\033[37m"), "\033[40m");end
+
+def bold(text); style(text, "\033[01m");end
+def underline(text); style(text, "\033[04m");end
+

data/lib/configuration.rb

@@ -0,0 +1,143 @@
+#encoding: UTF-8
+=begin
+/***************************************************************************
+ *   2023-2024, Michael Uplawski <michael.uplawski@uplawski.eu>            *
+ *   This program is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the WTFPL 2.0 or later, see                     * 
+ *   http://www.wtfpl.net/about/                                           *
+ *                                                                         *
+ *   This program is distributed in the hope that it will be useful,       *
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of        *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.                  *
+ *                                                                         *
+ ***************************************************************************/
+=end
+
+require_relative 'basic_logging'
+require_relative 'version'
+require 'yaml'
+require 'singleton'
+
+# An object of this class represents the configuration of the program.
+# The parameters are set in the configuration-file
+
+class Configuration
+  include Singleton
+  include BasicLogging
+
+
+  def initialize() 
+    debug 'Configuration::initialize()'
+    confname = PROGNAME << '.conf' 
+    # try to open user-configuration
+    @config_file = ENV['HOME'].dup << File::Separator << '.' << confname
+    # if user-configuration does not exist, copy installed version.
+    installed_config = File::dirname(File::absolute_path(__FILE__)) << File::Separator << confname
+    debug 'installed configuration is in ' << installed_config
+    if !File.exist?(@config_file)
+      begin
+        File.open(@config_file, 'w') do |cf|
+          cf.write(File.read(installed_config ) )
+        end
+      rescue => ex
+        STDERR.puts('Cannot write user-configuration to ' << @config_file << '! (' << ex.message << ')' )
+      end
+    end
+    # read the config-file.
+    read_config
+    # updates the user-configuration *ONLY*
+    # if necessary
+    update_config(installed_config)    
+  end
+
+  # What this object does not have, may still be in the configuration.
+  def method_missing(method, args = nil)
+    if @conf
+      v = @conf[method]
+      debug('method_missing returns value for ' << method.to_s << ': |' << v.to_s << '|') 
+      return v
+    else
+      error("user-version of the configuration (#{@conf}) is not accessible")
+    end
+  end
+
+  # set a configuration option.
+  def set(key, value)
+    if @conf 
+      @conf[key] = value
+      if key == 'DEBUG_LOG' 
+        set_target value
+      end
+    end
+  end
+
+  attr_reader :log_target, :log_level
+  private 
+
+  # updates the configuration file, if need be.
+  def update_config(i_config)
+    if @conf
+    conf_version = @conf[:CONFIG_VERSION]
+    if !conf_version || conf_version < PROGVERSION.to_f
+      info "configuration has an older version number, looking for changes"
+      i_conf = YAML::load_file(i_config)
+      i_conf.transform_keys!{|k| k.to_sym}
+      i_keys = i_conf.keys
+      i_keys.each do |k|
+        if !@conf[k]
+          info('new configuration option: ' << k.to_s)
+          @conf[k] = ''
+        end
+      end
+      @conf[:CONFIG_VERSION] = PROGVERSION.to_f
+      bak_conf = @config_file.dup << '_' << conf_version.to_s
+      bak_conf.gsub!('.' << PROGNAME,  'bak_' << PROGNAME)
+      info('Old configuration is saved to ' << bak_conf ) 
+      info('New configuration is saved to ' << @config_file ) 
+      begin
+        File::write(bak_conf, File.read(@config_file))
+        File::write(@config_file, @conf.to_yaml)
+      rescue Exception => ex
+        msg = "Cannot write altered configuration to " << @config_file << "!\n\t" << ex.message
+        STDERR.puts(msg << "\nAborting, bye\n")
+        error msg
+        exit false
+      end
+    end
+    else
+      error "User-version of the configuration (#{@conf}) inaccessibble."
+      exit false
+    end
+  end
+
+  # reads the configuration and sets some properties.
+  def read_config
+    if File::exist?(@config_file) && File::readable?(@config_file)
+      begin
+        @conf = YAML::load_file(@config_file)
+
+        @conf.transform_keys!{|k| k.to_sym}
+        @keys = @conf.keys
+        clevel = @conf[:LOG_LEVEL]
+        set_level clevel
+        target_string = @conf[:DEBUG_LOG]
+        set_target (target_string == 'STDOUT' ? STDOUT : target_string)
+        # clear the log file.
+        clear_log
+        debug('log target and -level set: ' << @@log_level.to_s << ', ' << target_string)
+      rescue Exception => ex
+        msg = "Cannot use the configuration-file (" << @config_file << ")"
+        msg << "\n\t" << ex.message
+        # No log.
+        # error msg
+        STDERR.puts(msg << "\nAborting, bye\n") 
+        exit false
+      end
+    else
+      STDERR.puts("Cannot read the configuration-file (" << @config_file << ")") 
+      exit false
+    end
+  end
+end
+# EOF
+