X-Git-Url: https://git.cworth.org/git?a=blobdiff_plain;f=lib%2Fsup%2Fmbox%2Floader.rb;h=a11bf9541e6efed049754cf85442c6cdbd714c8b;hb=4efc8adaf8c63ff054c71149de6c12fe8194ff03;hp=4acc0645d90f99dd7ac5f77768a1b3e26c30e4a5;hpb=f562596aa5bc0f459b55408f87976ccee1db9eb5;p=sup

diff --git a/lib/sup/mbox/loader.rb b/lib/sup/mbox/loader.rb
index 4acc064..a11bf95 100644
--- a/lib/sup/mbox/loader.rb
+++ b/lib/sup/mbox/loader.rb
@@ -1,16 +1,20 @@
 require 'rmail'
 require 'uri'
+require 'set'
 
 module Redwood
 module MBox
 
 class Loader < Source
+  include SerializeLabelsNicely
   yaml_properties :uri, :cur_offset, :usual, :archived, :id, :labels
 
+  attr_reader :labels
+
   ## uri_or_fp is horrific. need to refactor.
-  def initialize uri_or_fp, start_offset=nil, usual=true, archived=false, id=nil, labels=[]
+  def initialize uri_or_fp, start_offset=0, usual=true, archived=false, id=nil, labels=nil
     @mutex = Mutex.new
-    @labels = ((labels || []) - LabelManager::RESERVED_LABELS).uniq.freeze
+    @labels = Set.new((labels || []) - LabelManager::RESERVED_LABELS)
 
     case uri_or_fp
     when String
@@ -46,7 +50,7 @@ class Loader < Source
       raise OutOfSyncSourceError, "mbox file is smaller than last recorded message offset. Messages have probably been deleted by another client."
     end
   end
-    
+
   def start_offset; 0; end
   def end_offset; File.size @f; end
 
@@ -55,10 +59,10 @@ class Loader < Source
     @mutex.synchronize do
       @f.seek offset
       l = @f.gets
-      unless l =~ BREAK_RE
+      unless MBox::is_break_line? l
         raise OutOfSyncSourceError, "mismatch in mbox file offset #{offset.inspect}: #{l.inspect}." 
       end
-      header = MBox::read_header @f
+      header = parse_raw_email_header @f
     end
     header
   end
@@ -67,25 +71,36 @@ class Loader < Source
     @mutex.synchronize do
       @f.seek offset
       begin
-        RMail::Mailbox::MBoxReader.new(@f).each_message do |input|
-          m = RMail::Parser.read(input)
-          if m.body && m.body.is_a?(String)
-            m.body.gsub!(/^>From /, "From ")
-          end
-          return m
-        end
+        ## don't use RMail::Mailbox::MBoxReader because it doesn't properly ignore
+        ## "From" at the start of a message body line.
+        string = ""
+        l = @f.gets
+        string << l until @f.eof? || MBox::is_break_line?(l = @f.gets)
+        RMail::Parser.read string
       rescue RMail::Parser::Error => e
         raise FatalSourceError, "error parsing mbox file: #{e.message}"
       end
     end
   end
 
+  ## scan forward until we're at the valid start of a message
+  def correct_offset!
+    @mutex.synchronize do
+      @f.seek cur_offset
+      string = ""
+      until @f.eof? || MBox::is_break_line?(l = @f.gets)
+        string << l
+      end
+      self.cur_offset += string.length
+    end
+  end
+
   def raw_header offset
     ret = ""
     @mutex.synchronize do
       @f.seek offset
-      until @f.eof? || (l = @f.gets) =~ /^$/
-        ret += l
+      until @f.eof? || (l = @f.gets) =~ /^\r*$/
+        ret << l
       end
     end
     ret
@@ -93,10 +108,19 @@ class Loader < Source
 
   def raw_message offset
     ret = ""
-    each_raw_message_line(offset) { |l| ret += l }
+    each_raw_message_line(offset) { |l| ret << l }
     ret
   end
 
+  def store_message date, from_email, &block
+    need_blank = File.exists?(@filename) && !File.zero?(@filename)
+    File.open(@filename, "a") do |f|
+      f.puts if need_blank
+      f.puts "From #{from_email} #{date.utc}"
+      yield f
+    end
+  end
+
   ## apparently it's a million times faster to call this directly if
   ## we're just moving messages around on disk, than reading things
   ## into memory with raw_message.
@@ -107,7 +131,7 @@ class Loader < Source
     @mutex.synchronize do
       @f.seek offset
       yield @f.gets
-      until @f.eof? || (l = @f.gets) =~ BREAK_RE
+      until @f.eof? || MBox::is_break_line?(l = @f.gets)
         yield l
       end
     end
@@ -128,7 +152,7 @@ class Loader < Source
         ## 2. at the beginning of an mbox separator (in all other
         ##    cases).
 
-        l = @f.gets or raise "next while at EOF"
+        l = @f.gets or return nil
         if l =~ /^\s*$/ # case 1
           returned_offset = @f.tell
           @f.gets # now we're at a BREAK_RE, so skip past it
@@ -138,7 +162,7 @@ class Loader < Source
         end
 
         while(line = @f.gets)
-          break if line =~ BREAK_RE
+          break if MBox::is_break_line? line
           next_offset = @f.tell
         end
       end
@@ -147,7 +171,7 @@ class Loader < Source
     end
 
     self.cur_offset = next_offset
-    [returned_offset, (self.labels + [:unread]).uniq]
+    [returned_offset, (labels + [:unread])]
   end
 end