middleman-spellcheck 0.7.5 → 0.8.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: cc78078d67f2333e29a7b05751aa5bbf08d313d2
-  data.tar.gz: cfd3822f1fccf0007a1336f94b4befe3c48cf794
+  metadata.gz: 8608b5d6fd9049afe93af7ef3f005566bb10e537
+  data.tar.gz: 00b148f8992c92b178992c485bea7a4275b80ea8
 SHA512:
-  metadata.gz: 4d9077b0513f76faffc0d14f8a43548b86b53732520d457d6c60096307b4e317f2299ccdce3625e7dd7db3567e15e9765acf54b79af030338d8d2e9d0e9680ce
-  data.tar.gz: 38d36abde02d9af11b31d6ad9e1fe62226be035e00097dde0ad57f5babba28a9922a88e351045e95bca54fcc552431f2e1417ac00b5d87346ae2984f1e1a829e
+  metadata.gz: d86956a715506d4924060b035a3c8edd1f595b154d54b7a6e40ac4ebbb39b72f23a40d31939246e2a561d0f90a05e358975ce929ac0eec14a9c6bd1f8ab21701
+  data.tar.gz: 376ba57130d2b77c563a2ee040adc564012e790ba964451d832d9ebe627e762bfcb0af4538da72bd70eff1496dcdb92e770302020e3dc36ee7e3dacf417225d0

data/README.md

@@ -44,6 +44,39 @@ like to skip:
 activate :spellcheck, ignored_exts: [".xml", ".png"]
 ```
 
+To select a dictionary used by a spellchecker, use lang: option. For
+example, to use Polish dictionary, use:
+
+```ruby
+activate :spellcheck, lang: "pl"
+```
+
+Middleman-spellcheck can issue many warnings if you run it over a new
+content. If you want to give yourself a chance to fix mistakes gradually and
+not fail each time you build, use :dontfail flag:
+
+```ruby
+activate :spellcheck, lang: en, dontfail: 1
+```
+
+Advanced users wishing to invoke Middleman-spellcheck backend (Aspell) with
+a custom command line may find cmdargs: useful. Please note that "-a" is a
+mandatory flag which one must specify in order for middleman-spellcheck to
+work. Other flags are up to the user. See Aspell's man page for more
+details.
+
+```ruby
+activate :spellcheck, cmdargs: "-a -l pl"
+```
+
+For developers interested in extending Middleman-spellcheck and for those
+who encountered issues, useful might be debug: option, which will turn on
+extensive amount of debugging.
+
+```ruby
+activate :spellcheck, debug: 1
+```
+
 ## Contributing
 
 1. Fork it

data/lib/middleman-spellcheck/extension.rb

@@ -9,14 +9,20 @@ module Middleman
       option :tags, [], "Run spellcheck only on some tags from the output"
       option :allow, [], "Allow specific words to be misspelled"
       option :ignored_exts, [], "Ignore specific extensions (ex: '.xml')"
+      option :lang, "en", "Language for spellchecking"
+      option :cmdargs, "", "Pass alternative command line arguments"
+      option :debug, 0, "Enable debugging (for developers only)"
+      option :dontfail, 0, "Don't fail when misspelled words are found"
 
       def after_build(builder)
+        Spellchecker.cmdargs=(options.cmdargs)
+        Spellchecker.debug_enabled=(options.debug)
         filtered = filter_resources(app, options.page)
         total_misspelled = []
 
         filtered.each do |resource|
           builder.say_status :spellcheck, "Running spell checker for #{resource.url}", :blue
-          current_misspelled = run_check(select_content(resource))
+          current_misspelled = run_check(select_content(resource), options.lang)
           current_misspelled.each do |misspell|
             builder.say_status :misspell, error_message(misspell), :red
           end
@@ -24,7 +30,13 @@ module Middleman
         end
 
         unless total_misspelled.empty?
-          raise Thor::Error, "Build failed. There are spelling errors."
+	  estr = "Build failed. There are spelling errors."
+	  if options.dontfail
+	    print "== :dontfail set! Will issue warning only, but not fail.\n"
+	    print estr, "\n"
+	  else
+            raise Thor::Error, estr
+	  end
         end
       end
 
@@ -60,8 +72,8 @@ module Middleman
                              .reject { |resource| option_ignored_exts.include? resource.ext }
       end
 
-      def run_check(text, dictionary="en")
-        results = Spellchecker.check(text, dictionary)
+      def run_check(text, lang)
+        results = Spellchecker.check(text, lang)
         results = exclude_allowed(results)
         results.reject { |entry| entry[:correct] }
       end

data/lib/middleman-spellcheck/spellchecker.rb

@@ -1,5 +1,7 @@
 class Spellchecker
   @@aspell_path = "aspell"
+  @@aspell_cmdargs = ""
+  @@debug_enabled = 0
 
   def self.aspell_path=(path)
     @@aspell_path = path
@@ -9,28 +11,79 @@ class Spellchecker
     @@aspell_path
   end
 
-  def self.query(text, lang='en')
-    result = `echo "#{text}" | #{@@aspell_path} -a -l #{lang}`
+  def self.cmdargs=(args)
+    @@aspell_cmdargs = args
+  end
+
+  def self.cmdargs
+    @@aspell_cmdargs
+  end
+
+  def self.debug_enabled=(args)
+    @@debug_enabled = args
+  end
+
+  def self.debug_enabled
+    @@debug_enabled
+  end
+
+  def self.sdbg(*args)
+    if @@debug_enabled <= 0
+      return
+    end
+    print "# DBG ", *args, "\n"
+  end
+
+  def self.query(words, lang)
+    args = "-a -l #{lang}"
+    if @@aspell_cmdargs != ""
+      args = @@aspell_cmdargs
+    end
+    cmd = %Q<#{@@aspell_path} #{args}>
+    result = []
+
+    sdbg "Starting aspell"
+    IO.popen(cmd, "a+", :err=>[:child, :out]) { |f|
+      val = f.gets.strip()	# skip the Aspell's intro
+      sdbg "Expected Aspell intro, got #{val}"
+      words.each do |word|
+        sdbg "-> Writing word '#{word}'"
+        f.write(word + "\n")
+        f.flush
+
+        # should be * or &
+        word_check_res = f.gets.strip()
+        sdbg "<- got result '#{word_check_res}'"
+
+        # skip the empty line
+        val = f.gets()
+	sdbg "Expected empty line, got '#{val}'"
+
+        result << word_check_res
+      end
+    }
     raise 'Aspell command not found' unless result
-    new_result = result.split("\n")
-    new_result[1..-1] || []
+    result || []
   end
 
   def self.correct?(result_string)
     result_string == "*"
   end
 
-  def self.check(text, lang='en')
-    # join then re-split the word list to get a consistent word count,
-    # because sometimes there's a "" (blank) word in the array that gets lost,
-    # which makes the maps not equal, leading to an off by one type issue, where
-    # the reported mispelled word is actually a correct word.
-    # see: https://github.com/minivan/middleman-spellcheck/issues/7
-    words   = text.split(/[^A-Za-z’']+/).join(" ")
+  def self.check(text, lang)
+    # do ’ -> ' for aspell. Otherwise 's' is passed as a word to aspell.
+    text.gsub! '’', '\''
+    sdbg "self.check got raw text:\n#{text}\n"
+
+    words = text.split(/[^A-Za-z']+/).select { |s|
+       s != "" and s != "'s" and s != "'"
+    }
+    sdbg "self.check word array:\n#{words}\n"
+
     results = query(words, lang).map do |query_result|
       correct?(query_result)
     end
 
-    words.split(" ").zip(results).map {|word, correctness| { word: word, correct: correctness } }
+    words.zip(results).map {|word, correctness| { word: word, correct: correctness } }
   end
 end

data/lib/middleman-spellcheck/version.rb

@@ -1,5 +1,5 @@
 module Middleman
   module Spellcheck
-    VERSION = "0.7.5"
+    VERSION = "0.8.0"
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: middleman-spellcheck
 version: !ruby/object:Gem::Version
-  version: 0.7.5
+  version: 0.8.0
 platform: ruby
 authors:
 - Ivan Zarea
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-10-25 00:00:00.000000000 Z
+date: 2015-08-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: middleman-core