mailcvt 0.2.29 → 0.2.30

data/lib/encoded_word.rb

@@ -63,6 +63,12 @@ class EncodedWord
         end
     end
 
+    def concat_one_line(words)
+        line = words.join('')
+        parts = mysplit(line, "\n")
+        parts.join('')
+    end
+
     def mysplit(line, sep = "\t")
         return [] unless line
         return [] unless line.length > 0
@@ -72,7 +78,7 @@ class EncodedWord
         pos1 = -1
         while true do
             pos1 += 1
-            pos2 = line.index("\t", pos1)
+            pos2 = line.index(sep, pos1)
             if pos2
                 parts << line[pos1...pos2]
                 pos1 = pos2
@@ -94,9 +100,13 @@ class EncodedWord
 
     def decode_subject(sub)
         return '' unless sub and sub.length > 0
-        sub = sub.encode('utf-8', @input_enc, :undef=>:replace, :invalid=>:replace)
-        one = sub.gsub("\a", '')
-        plain = mime_decode(one).gsub("\n", '')
+        parts = mysplit(sub, "\a")
+        words = []
+        parts.each do |p|
+            wd = word_decode(p) 
+            words << wd
+        end
+        concat_one_line(words)
     end
 
     def trim_emails(emails)
@@ -136,18 +146,31 @@ class EncodedWord
             attaches << decode_attach(parts[i])
         end
         return '' unless attaches.length > 0
-        attaches.join("\a").gsub("\n", '')
+        attaches.join("\a")
     end
 
     def decode_attach(attach)
         return '' unless attach and attach.length > 0
-        attach = attach.encode('utf-8', @input_enc, :undef=>:replace, :invalid=>:replace)
         parts = mysplit(attach, "\a")
-        newparts = []
+        words = []
         parts.each do |p|
-            newparts << mime_decode(p)
+            wd = word_decode(p) 
+            words << wd
         end
-        newparts.join('')
+        concat_one_line(words)
+    end
+
+    def word_decode(input, out_charset = 'utf-8')
+        u8 = input.encode('utf-8', @u8_enc, :undef=>:replace, :invalid=>:replace)
+        pos1 = u8.index('=?')
+        return input unless pos1
+        pos2 = u8.index('?=', pos1+1)
+        trim = u8[pos1..pos2+1]
+        parts = trim.scan(/=\?([A-Za-z0-9_-]+)\?([BQbq])\?([^\?]+)\?=/).first
+        charset = parts[0]
+        enc = parts[1].upcase
+        wd = parts[2].unpack({ "B"=>"m*", "Q"=>"M*" }[enc]).first
+        wd.encode(out_charset, charset, :undef=>:replace, :invalid=>:replace)
     end
 
     def mime_decode(input, out_charset = 'utf-8')
@@ -161,7 +184,8 @@ class EncodedWord
                 word.encode(out_charset, charset, :undef=>:replace, :invalid=>:replace)
             }
             return ret ? mime_decode(input) : input
-        rescue
+        rescue => e
+            puts e
             return input
         end
    end

data/lib/mailcvt/version.rb

@@ -1,3 +1,3 @@
 module Mailcvt
-  VERSION = '0.2.29'
+  VERSION = '0.2.30'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: mailcvt
 version: !ruby/object:Gem::Version
-  version: 0.2.29
+  version: 0.2.30
   prerelease: 
 platform: ruby
 authors: