mailcvt 0.2.20 → 0.2.21

data/lib/encoded_word.rb

@@ -16,14 +16,14 @@ class EncodedWord
             puts f
             File.open(f + '.plain', 'w') do |out|
             File.readlines(f).each do |line|
-                parts = line.split("\t")
+                parts = mysplit(line)
                 newparts = []
                 newparts << format_date(parts[0]) #date
                 newparts << parts[1] #message-id
                 newparts << decode_subject(parts[2]) #subject
-                newparts << trim_email(parts[3]) #from
-                newparts << trim_email(parts[4]) #to
-                newparts << trim_email(parts[5]) #cc
+                newparts << trim_emails(parts[3]) #from
+                newparts << trim_emails(parts[4]) #to
+                newparts << trim_emails(parts[5]) #cc
                 newparts << decode_attachments(parts[6]) unless parts.length == 6 #attachment
                 out.puts newparts.join("\t")
             end
@@ -31,38 +31,72 @@ class EncodedWord
         end
     end
 
+    def mysplit(line, sep = "\t")
+        return [] unless line
+        return [] unless line.length > 0
+
+        parts = []
+        pos1 = -1
+        while true do
+            pos1 += 1
+            pos2 = line.index("\t", pos1)
+            if pos2
+                parts << line[pos1...pos2]
+                pos1 = pos2
+            else
+                parts << line[pos1..-1]
+                break
+            end
+        end
+        return parts
+    end
+
     def format_date(engdate)
+        return '' unless engdate and engdate.length > 0
         dt = DateTime.parse(engdate)
         dt = dt.new_offset('+0900')
         dt.strftime("%Y/%m/%d %H:%M:%S")
     end
 
     def decode_subject(sub)
+        return '' unless sub and sub.length > 0
         one = sub.gsub("\a", '')
         mime_decode(one)
     end
 
     def trim_emails(emails)
-        parts = emails.split("\a")
+        return '' unless emails and emails.length > 0
+        pos1 = -1
         newparts = []
-        parts.each do |p|
-            newparts << trim_email(p)
+        while true do
+            pos1 += 1
+            pos2 = emails.index('@', pos1)
+            break unless pos2
+            newparts << getmail(emails, pos2)
+            pos1 = pos2
         end
         newparts.join("\a")
     end
 
-    def trim_email(email)
-        pos1 = email.index('<')
-        if pos1
-            pos1 += 1
-            pos2 = email.index('>', pos1)
-            return email[pos1...pos2]
+    def getmail(line, at)
+        pos1 = 0
+        pos2 = line.length - 1
+        (at-1).step(0, -1) do |i|
+            next if line[i] =~ /[\._a-zA-Z0-9-]/
+            pos1 = i + 1
+            break
+        end
+        (at+1).upto(line.length) do |i|
+            next if line[i] =~ /[\.a-zA-Z0-9-]/
+            pos2 = i - 1
+            break
         end
-        email.strip
+        line[pos1..pos2]
     end
 
     def decode_attachments(attachments)
-        parts = attachments.split("\a")
+        return '' unless attachments and attachments.length > 0
+        parts = mysplit(attachments, "\a")
         newparts = []
         parts.each do |p|
             newparts << mime_decode(p)
@@ -71,6 +105,7 @@ class EncodedWord
     end
 
     def mime_decode(input, out_charset = 'utf-8')
+        return '' unless input and input.length > 0
         ret = input.sub!(/=\?([A-Za-z0-9_-]+)\?([BQbq])\?([^\?]+)\?=/) {
             charset = $1
             enc = $2.upcase

data/lib/mailcvt/version.rb

@@ -1,3 +1,3 @@
 module Mailcvt
-  VERSION = '0.2.20'
+  VERSION = '0.2.21'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: mailcvt
 version: !ruby/object:Gem::Version
-  version: 0.2.20
+  version: 0.2.21
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-10-03 00:00:00.000000000 Z
+date: 2013-10-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake