get_pomo 0.7.1 → 0.8.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: bb1e35bef67b57c020e28e8a0695678104b80ed1
-  data.tar.gz: 5e489f0e04fab3e8510b7bc97ed698d7f20bc280
+  metadata.gz: 4a7082c11274ecb41c557230f20679ab1afc2b47
+  data.tar.gz: 96a578a70bf2e2690624d6b8f630da7237b55dc5
 SHA512:
-  metadata.gz: 4dcbfcd7b5c29ff12b612c85b3edb06c7e7280be55c07f1d75c6fefbebcc9c91021087e6fc32df37fc6736cb33c4e713336b341da26b077b84620eade298e4b7
-  data.tar.gz: 41da61f15f4bc18050a7cf0532ad39453f58064ec527e16264c6632ebb154eef2100d764e4c9f9e955d8e28fe0cec840f4f8efb2ccb5e17283b4c568118e72a2
+  metadata.gz: 435a39400d98fe210781d43bbf80db4a56fbfc7e747e23ba872b107fc6b04c44e60fa6e68c0967a5bf5ebc6d8edb1fcc763230ae09b836da306b5e0d2fd33200
+  data.tar.gz: cd614be59a4bbfb31a2b487f7c17bbf4721c14e29c7674cbb6a091fe940c986c74394f61b45a39581764088fd0a663988aa62bde84917255d48713c93a556192

data/lib/get_pomo/po_file.rb

@@ -3,13 +3,17 @@ require 'get_pomo/translation'
 
 module GetPomo
   class PoFile
-    def self.parse(text)
-      PoFile.new.add_translations_from_text(text)
+    def self.parse(text, options = {})
+      default_options = {:parse_obsoletes => false}
+      options = default_options.merge(options)
+      PoFile.new.add_translations_from_text(text, options)
     end
 
-    def self.to_text(translations)
+    def self.to_text(translations, options = {})
+      default_options = {:merge => false}
+      options = default_options.merge(options)
       p = PoFile.new(:translations=>translations)
-      p.to_text
+      p.to_text(options)
     end
 
     attr_reader :translations
@@ -21,7 +25,9 @@ module GetPomo
     #the text is split into lines and then converted into logical translations
     #each translation consists of comments(that come before a translation)
     #and a msgid / msgstr
-    def add_translations_from_text(text)
+    def add_translations_from_text(text, options = {})
+      default_options = {:parse_obsoletes => false}
+      options = default_options.merge(options)
       start_new_translation
       text.gsub!(/^#{"\357\273\277"}/, "") #remove boom
       text.split(/$/).each_with_index do |line,index|
@@ -35,12 +41,14 @@ module GetPomo
           add_string line
         end
       end
-      start_new_translation #instance_variable has to be overwritten or errors can occur on next add
+      start_new_translation(options[:parse_obsoletes]) #instance_variable has to be overwritten or errors can occur on next add
       translations
     end
 
-    def to_text
-      GetPomo.unique_translations(translations).map do |translation|
+    def to_text(options = {})
+      default_options = {:merge => false}
+      options = default_options.merge(options)
+      GetPomo.unique_translations(translations, options[:merge]).map do |translation|
         comment = translation.comment.to_s.split(/\n|\r\n/).map{|line|"#{line}\n"}*''
 
         msgctxt = if translation.msgctxt
@@ -61,7 +69,7 @@ module GetPomo
 
           msgids + (msgstrs*"\n")
         else
-          %Q(msgid "#{translation.msgid}"\n)+
+          translation.obsolete? ? "" : %Q(msgid "#{translation.msgid}"\n)+
           %Q(msgstr "#{translation.msgstr}")
         end
 
@@ -90,7 +98,6 @@ module GetPomo
     def parse_method_call(line)
       method, string = line.match(/^\s*([a-z0-9_\[\]]+)(.*)/)[1..2]
       raise "no method found" unless method
-
       start_new_translation if %W(msgid msgctxt msgctxt).include? method and translation_complete?
       @last_method = method.to_sym
       add_string(string)
@@ -110,12 +117,15 @@ module GetPomo
       @current_translation.complete?
     end
 
-    def store_translation
-      @translations += [@current_translation] if @current_translation.complete?
+    def store_translation(parse_obsoletes = false)
+      if translation_complete? && (parse_obsoletes || !@current_translation.obsolete?)
+       @translations += [@current_translation]
+      end
+
     end
 
-    def start_new_translation
-      store_translation if translation_complete?
+    def start_new_translation(parse_obsoletes = false)
+      store_translation(parse_obsoletes) if translation_complete?
       @current_translation = Translation.new
     end
   end

data/lib/get_pomo/translation.rb

@@ -13,9 +13,6 @@ module GetPomo
         self.msgstr ||= []
         msgstr[$1.to_i] = msgstr[$1.to_i].to_s + text
       elsif to.to_sym == :comment && text =~ OBSOLETE_REGEX
-        # initialize msgid and msgstr on obsolete translations
-        self.msgid ||= ""
-        self.msgstr ||= ""
         send("#{to}=",send(to).to_s+text)
       else
         #simple form
@@ -33,7 +30,7 @@ module GetPomo
     end
 
     def complete?
-      not msgid.nil? and not msgstr.nil?
+      (not msgid.nil? and not msgstr.nil?) or self.obsolete?
     end
 
     def fuzzy?

data/lib/get_pomo/version.rb

@@ -1,3 +1,3 @@
 module GetPomo
-  VERSION = "0.7.1"
+  VERSION = "0.8.0"
 end

data/lib/get_pomo.rb

@@ -3,10 +3,85 @@ module GetPomo
   autoload :VERSION, "get_pomo/version"
   
   extend self
-  
-  def self.unique_translations(translations)
-    last_seen_at_index = {}
-    translations.each_with_index {|translation,index|last_seen_at_index[translation.msgid]=index}
-    last_seen_at_index.values.sort.map{|index| translations[index]}
+
+  # A message is unique if the combination of msgid and msgctxt is unique.
+  # An emtpy msgctxt does not mean the same thing as no msgctxt.
+  # If there are multiple msgid's without msgctxt we use the latest one. If at least one has a msgctxt we completly
+  # drop all msgids without msgctxt and do not even mark them for merging. Multiple msgid's with different msgctxt will
+  # be treated as unique translations each.
+  #
+  # @param merge [Bool]  true merges references of non unique translations considering the rules above
+  def self.unique_translations(translations, merge = false)
+    seen_msgids = {}
+    # unique + merge logic
+    # hash with msgid's as keys.
+    # value is array if that msgid got no msgctxt
+    # value is hash if that msgid got a msgctxt
+    obsoletes = translations.select{|t| t.obsolete?}
+    translations.reject{ |t| t.obsolete?}.each_with_index do |translation, index|
+      key = translation.msgid
+      if seen_msgids.has_key?(key) # msgid not unique?
+        if seen_msgids[key].is_a?(Hash) # other translation with same msgid has a msgctxt?
+          if translation.msgctxt # this translation with same msgid has a msgctxt?, if not ignore it
+            if seen_msgids[key].has_key?(translation.msgctxt) # is it the same msgctxt?
+              seen_msgids[key][translation.msgctxt].push(index)
+            else
+              seen_msgids[key][translation.msgctxt] = [index]
+            end
+          end
+        else # other translation with same msgid had no msgctxt
+          if translation.msgctxt # if other translation has no msgctxt but this one has, just save this new one
+            seen_msgids[key] = {translation.msgctxt => [index]}
+          else # otherwise just push the new found translation for merge
+            seen_msgids[key].push(index)
+          end
+        end
+      else #msgid is unique so far
+        if translation.msgctxt # does the msgid got a msgctxt, if so add a nested hash with the msgctxt as key and an index_array as value
+          seen_msgids[key] = {translation.msgctxt => [index]}
+        else # if not, simply add an index array
+          seen_msgids[key] = [index]
+        end
+      end
+    end
+    trans_indexes = {} # maps unique translation ids as keys to mergeable ids as value array
+
+    seen_msgids.values.each do |value|
+      if value.is_a?(Hash)
+        value.values.each do |v|
+          id = v.pop
+          trans_indexes[id] = []
+          v.each do |index|
+             trans_indexes[id].push(index)
+          end
+        end
+      else
+        id = value.pop
+        trans_indexes[id] = []
+        value.each do |index|
+          trans_indexes[id].push(index)
+        end
+      end
+    end
+
+    trans = trans_indexes.keys.sort.map do |index|
+      if merge
+        t = translations[index]
+        trans_indexes[index].each do |id|
+          # sub starting \n# to \r\n as a split on \r\n should be safe
+          # (\n would match \\n in text, \r\n does not match \\r\\n)
+          # this is still monkey patching but should work fine
+          translations[id].comment.gsub(/\n#/, "\r\n#").split(/\r\n/).each do |com|
+            # prepend all references to the existing reference
+            t.comment.sub!("#:", com.chomp) if com.start_with?("#: ") && !t.comment.include?(com.sub("#:", ''))
+          end
+        end
+        t
+      else
+        translations[index]
+      end
+    end
+    trans.concat(obsoletes) unless obsoletes.nil?
+    trans
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: get_pomo
 version: !ruby/object:Gem::Version
-  version: 0.7.1
+  version: 0.8.0
 platform: ruby
 authors:
 - Michael Grosser
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-05-02 00:00:00.000000000 Z
+date: 2014-08-18 00:00:00.000000000 Z
 dependencies: []
 description: 
 email: michael@grosser.it