exe/httpspell in httpspell-1.4.1 vs exe/httpspell in httpspell-1.5.0

@@ -8,12 +8,12 @@
 
 personal_dictionary_path = nil
 force_language = nil
 tracing = nil
 verbose = nil
-whitelist = nil
-blacklist = []
+included = nil
+excluded = []
 
 begin
   OptionParser.new do |parser|
     parser.banner.prepend <<~BANNER
       Spellchecks a website via HTTP.
@@ -27,71 +27,71 @@
 
     parser.on('-l', '--language=LANGUAGE', 'override LANGUAGE of content') do |l|
       force_language = l
     end
 
-    parser.on('-w', '--whitelist=EXPRESSION', 'when recursively retrieving URLs, allow only those matching the given regular EXPRESSION') do |w|
-      whitelist ||= []
-      whitelist << Regexp.new(w)
+    parser.on('-i', '--include=EXPRESSION', 'when recursively retrieving URLs, allow only those matching the given regular EXPRESSION') do |w|
+      included ||= []
+      included << Regexp.new(w)
     end
 
     parser.on('-t', '--trace', 'enable error tracing') do
       tracing = true
     end
 
     parser.on('-V', '--verbose', "explain what's happening") do
       verbose = true
     end
 
-    parser.on('-b', '--blacklist=EXPRESSION', 'blacklist (ignore) URLs matching the given regular EXPRESSION') do |b|
-      blacklist << Regexp.new(b)
+    parser.on('-e', '--exclude=EXPRESSION', 'exclude URLs matching the given regular EXPRESSION') do |b|
+      excluded << Regexp.new(b)
     end
 
     # TODO: --recursive, defaults to false
     # TODO wget has some additional options for recursive behavior that should be reviewed
   end.parse!
 rescue StandardError
-  warn "Error - #{$ERROR_INFO}"
+  warn "Error: #{$ERROR_INFO}"
   exit 1
 end
 
 if ARGV.size != 1
   warn "Expected exactly one argument, but received #{ARGV.size}."
   exit 1
 end
 
-def check(doc, lang, personal_dictionary_path, verbose)
+def check(url, doc, lang, personal_dictionary_path, verbose)
   unknown_words = HttpSpell::SpellChecker.new(personal_dictionary_path, verbose:).check(doc, lang)
 
   if unknown_words.empty?
-    warn 'No unknown words.' if verbose
+    warn "#{url} (lang=#{lang}): No unknown words" if verbose
+    false
   else
-    warn "#{unknown_words.size} unknown words:" if verbose
+    warn "#{url} (lang=#{lang}): #{unknown_words.size} unknown words:" if verbose
     puts unknown_words
     true
   end
 end
 
 has_unknown_words = false
 
-spider_success = HttpSpell::Spider.new(ARGV.first, whitelist:, blacklist:, verbose:, tracing:).start do |url, doc|
+spider_success = HttpSpell::Spider.new(ARGV.first, included:, excluded:, verbose:, tracing:).start do |url, doc|
   lang = force_language || doc.root['lang'] || ENV.fetch('LANGUAGE', nil)
-  warn "Checking #{url} as #{lang}" if verbose
 
   # Remove elements that are not to be spellchecked
   doc.css('pre').each(&:unlink)
   doc.css('code').each(&:unlink)
+  doc.css('iframe').each(&:unlink)
   doc.css('[spellcheck=false]').each(&:unlink)
 
   # Handle elements with a different lang attribute separately
   doc.css(%([lang]:not([lang="#{lang}"]))).each do |element|
-    warn "Handling #{element.name} with lang #{element['lang']}:" if verbose
-    has_unknown_words |= check(element.to_s, element['lang'], personal_dictionary_path, verbose)
+    has_unknown_words |= check("#{url} => #{element.name} with", element.to_s, element['lang'], personal_dictionary_path, verbose)
     element.unlink
   end
 
   # Everything else
-  has_unknown_words |= check(doc.to_s, lang, personal_dictionary_path, verbose)
+  has_unknown_words |= check("#{url} => document with", doc.to_s, lang, personal_dictionary_path, verbose)
 end
 
 exit 2 unless spider_success
 exit 1 if has_unknown_words