X-Git-Url: https://git.cworth.org/git?a=blobdiff_plain;f=lib%2Fsup%2Fmessage.rb;h=c06a03fa547a3d6c96dd19945fae0934b907c925;hb=073e1f40024d7721ada08bce7c38d87883ec62f6;hp=ca7efad4cd0d4d25d734978abe61d6414decca80;hpb=b675cd3d9c8674390fd23c54ecc8f1ff001a6999;p=sup

diff --git a/lib/sup/message.rb b/lib/sup/message.rb
index ca7efad..c06a03f 100644
--- a/lib/sup/message.rb
+++ b/lib/sup/message.rb
@@ -1,4 +1,3 @@
-require 'tempfile'
 require 'time'
 require 'iconv'
 
@@ -13,6 +12,10 @@ class MessageFormatError < StandardError; end
 ## i would like, for example, to be able to add in a ruby-talk
 ## specific module that would detect and link to /ruby-talk:\d+/
 ## sequences in the text of an email. (how sweet would that be?)
+##
+## this class cathces all source exceptions. if the underlying source throws
+## an error, it is caught and handled.
+
 class Message
   SNIPPET_LEN = 80
   RE_PATTERN = /^((re|re[\[\(]\d[\]\)]):\s*)+/i
@@ -26,7 +29,7 @@ class Message
 
   QUOTE_PATTERN = /^\s{0,4}[>|\}]/
   BLOCK_QUOTE_PATTERN = /^-----\s*Original Message\s*----+$/
-  QUOTE_START_PATTERN = /(^\s*Excerpts from)|(^\s*In message )|(^\s*In article )|(^\s*Quoting )|((wrote|writes|said|says)\s*:\s*$)/
+  QUOTE_START_PATTERN = /\w.*:$/
   SIG_PATTERN = /(^-- ?$)|(^\s*----------+\s*$)|(^\s*_________+\s*$)|(^\s*--~--~-)|(^\s*--\+\+\*\*==)/
 
   MAX_SIG_DISTANCE = 15 # lines from the end
@@ -35,45 +38,54 @@ class Message
 
   attr_reader :id, :date, :from, :subj, :refs, :replytos, :to, :source,
               :cc, :bcc, :labels, :list_address, :recipient_email, :replyto,
-              :source_info, :chunks, :list_subscribe, :list_unsubscribe
+              :source_info, :list_subscribe, :list_unsubscribe
 
-  bool_reader :dirty, :source_marked_read
+  bool_reader :dirty, :source_marked_read, :snippet_contains_encrypted_content
 
   ## if you specify a :header, will use values from that. otherwise,
   ## will try and load the header from the source.
   def initialize opts
     @source = opts[:source] or raise ArgumentError, "source can't be nil"
     @source_info = opts[:source_info] or raise ArgumentError, "source_info can't be nil"
-    @snippet = opts[:snippet] || ""
-    @have_snippet = !opts[:snippet].nil?
+    @snippet = opts[:snippet]
+    @snippet_contains_encrypted_content = false
+    @have_snippet = !(opts[:snippet].nil? || opts[:snippet].empty?)
     @labels = [] + (opts[:labels] || [])
     @dirty = false
+    @encrypted = false
     @chunks = nil
 
+    ## we need to initialize this. see comments in parse_header as to
+    ## why.
+    @refs = []
+
     parse_header(opts[:header] || @source.load_header(@source_info))
   end
 
   def parse_header header
     header.each { |k, v| header[k.downcase] = v }
-    
+
+    fakeid = nil
+    fakename = nil
+
     @id =
       if header["message-id"]
         sanitize_message_id header["message-id"]
       else
-        returning("sup-faked-" + Digest::MD5.hexdigest(raw_header)) do |id|
-          Redwood::log "faking message-id for message from #@from: #{id}"
-        end
+        fakeid = "sup-faked-" + Digest::MD5.hexdigest(raw_header)
       end
     
     @from =
       if header["from"]
         PersonManager.person_for header["from"]
       else
-        name = "Sup Auto-generated Fake Sender <sup@fake.sender.example.com>"
-        Redwood::log "faking from for message #@id: #{name}"
-        PersonManager.person_for name
+        fakename = "Sup Auto-generated Fake Sender <sup@fake.sender.example.com>"
+        PersonManager.person_for fakename
       end
 
+    Redwood::log "faking message-id for message from #@from: #{id}" if fakeid
+    Redwood::log "faking from for message #@id: #{fakename}" if fakename
+
     date = header["date"]
     @date =
       case date
@@ -83,7 +95,8 @@ class Message
         begin
           Time.parse date
         rescue ArgumentError => e
-          raise MessageFormatError, "unparsable date #{header['date']}: #{e.message}"
+          Redwood::log "faking date header for #{@id} due to error parsing date #{header['date'].inspect}: #{e.message}"
+          Time.now
         end
       else
         Redwood::log "faking date header for #{@id}"
@@ -94,7 +107,13 @@ class Message
     @to = PersonManager.people_for header["to"]
     @cc = PersonManager.people_for header["cc"]
     @bcc = PersonManager.people_for header["bcc"]
-    @refs = (header["references"] || "").scan(/<(.+?)>/).map { |x| sanitize_message_id x.first }
+
+    ## before loading our full header from the source, we can actually
+    ## have some extra refs set by the UI. (this happens when the user
+    ## joins threads manually). so we will merge the current refs values
+    ## in here.
+    refs = (header["references"] || "").scan(/<(.+?)>/).map { |x| sanitize_message_id x.first }
+    @refs = (@refs + refs).uniq
     @replytos = (header["in-reply-to"] || "").scan(/<(.+?)>/).map { |x| sanitize_message_id x.first }
 
     @replyto = PersonManager.person_for header["reply-to"]
@@ -112,7 +131,16 @@ class Message
   end
   private :parse_header
 
-  def snippet; @snippet || chunks && @snippet; end
+  def add_ref ref
+    @refs << ref
+    @dirty = true
+  end
+
+  def remove_ref ref
+    @dirty = true if @refs.delete ref
+  end
+
+  def snippet; @snippet || (chunks && @snippet); end
   def is_list_message?; !@list_address.nil?; end
   def is_draft?; @source.is_a? DraftLoader; end
   def draft_filename
@@ -123,8 +151,10 @@ class Message
   def sanitize_message_id mid; mid.gsub(/\s/, "") end
 
   def save index
-    index.sync_message self if @dirty
+    return unless @dirty
+    index.sync_message self
     @dirty = false
+    true
   end
 
   def has_label? t; @labels.member? t; end
@@ -148,11 +178,16 @@ class Message
     @dirty = true
   end
 
+  def chunks
+    load_from_source!
+    @chunks
+  end
+
   ## this is called when the message body needs to actually be loaded.
   def load_from_source!
     @chunks ||=
       if @source.has_errors?
-        [Chunk::Text.new(error_message(@source.error.message.split("\n")))]
+        [Chunk::Text.new(error_message(@source.error.message).split("\n"))]
       else
         begin
           ## we need to re-read the header because it contains information
@@ -169,8 +204,9 @@ class Message
           Redwood::log "problem getting messages from #{@source}: #{e.message}"
           ## we need force_to_top here otherwise this window will cover
           ## up the error message one
+          @source.error ||= e
           Redwood::report_broken_sources :force_to_top => true
-          [Chunk::Text.new(error_message(e.message))]
+          [Chunk::Text.new(error_message(e.message).split("\n"))]
         end
       end
   end
@@ -194,11 +230,14 @@ The error message was:
 EOS
   end
 
+  ## wrap any source methods that might throw sourceerrors
   def with_source_errors_handled
     begin
       yield
     rescue SourceError => e
       Redwood::log "problem getting messages from #{@source}: #{e.message}"
+      @source.error ||= e
+      Redwood::report_broken_sources :force_to_top => true
       error_message e.message
     end
   end
@@ -216,23 +255,24 @@ EOS
     with_source_errors_handled { @source.each_raw_message_line(@source_info, &b) }
   end
 
-  def content
+  ## returns all the content from a message that will be indexed
+  def indexable_content
     load_from_source!
     [
-      from && "#{from.name} #{from.email}",
-      to.map { |p| "#{p.name} #{p.email}" },
-      cc.map { |p| "#{p.name} #{p.email}" },
-      bcc.map { |p| "#{p.name} #{p.email}" },
+      from && from.indexable_content,
+      to.map { |p| p.indexable_content },
+      cc.map { |p| p.indexable_content },
+      bcc.map { |p| p.indexable_content },
       chunks.select { |c| c.is_a? Chunk::Text }.map { |c| c.lines },
       Message.normalize_subj(subj),
     ].flatten.compact.join " "
   end
 
-  def basic_body_lines
-    chunks.find_all { |c| c.is_a?(Chunk::Text) || c.is_a?(Chunk::Quote) }.map { |c| c.lines }.flatten
+  def quotable_body_lines
+    chunks.find_all { |c| c.quotable? }.map { |c| c.lines }.flatten
   end
 
-  def basic_header_lines
+  def quotable_header_lines
     ["From: #{@from.full_address}"] +
       (@to.empty? ? [] : ["To: " + @to.map { |p| p.full_address }.join(", ")]) +
       (@cc.empty? ? [] : ["Cc: " + @cc.map { |p| p.full_address }.join(", ")]) +
@@ -267,7 +307,6 @@ private
   ## product.
 
   def multipart_signed_to_chunks m
-#    Redwood::log ">> multipart SIGNED: #{m.header['Content-Type']}: #{m.body.size}"
     if m.body.size != 2
       Redwood::log "warning: multipart/signed with #{m.body.size} parts (expecting 2)"
       return
@@ -279,13 +318,15 @@ private
       return
     end
 
+    ## this probably will never happen
     if payload.header.content_type == "application/pgp-signature"
       Redwood::log "warning: multipart/signed with payload content type #{payload.header.content_type}"
       return
     end
 
     if signature.header.content_type != "application/pgp-signature"
-      Redwood::log "warning: multipart/signed with signature content type #{signature.header.content_type}"
+      ## unknown signature type; just ignore.
+      #Redwood::log "warning: multipart/signed with signature content type #{signature.header.content_type}"
       return
     end
 
@@ -293,7 +334,6 @@ private
   end
 
   def multipart_encrypted_to_chunks m
-    Redwood::log ">> multipart ENCRYPTED: #{m.header['Content-Type']}: #{m.body.size}"
     if m.body.size != 2
       Redwood::log "warning: multipart/encrypted with #{m.body.size} parts (expecting 2)"
       return
@@ -316,11 +356,11 @@ private
     end
 
     decryptedm, sig, notice = CryptoManager.decrypt payload
-    children = message_to_chunks(decryptedm) if decryptedm
+    children = message_to_chunks(decryptedm, true) if decryptedm
     [notice, sig, children].flatten.compact
   end
 
-  def message_to_chunks m, sibling_types=[]
+  def message_to_chunks m, encrypted=false, sibling_types=[]
     if m.multipart?
       chunks =
         case m.header.content_type
@@ -332,7 +372,7 @@ private
 
       unless chunks
         sibling_types = m.body.map { |p| p.header.content_type }
-        chunks = m.body.map { |p| message_to_chunks p, sibling_types }.flatten.compact
+        chunks = m.body.map { |p| message_to_chunks p, encrypted, sibling_types }.flatten.compact
       end
 
       chunks
@@ -344,17 +384,23 @@ private
     else
       filename =
         ## first, paw through the headers looking for a filename
-        if m.header["Content-Disposition"] &&
-            m.header["Content-Disposition"] =~ /filename="?(.*?[^\\])("|;|$)/
+        if m.header["Content-Disposition"] && m.header["Content-Disposition"] =~ /filename="?(.*?[^\\])("|;|$)/
           $1
-        elsif m.header["Content-Type"] &&
-            m.header["Content-Type"] =~ /name=(.*?)(;|$)/
+        elsif m.header["Content-Type"] && m.header["Content-Type"] =~ /name="?(.*?[^\\])("|;|$)/
           $1
 
         ## haven't found one, but it's a non-text message. fake
         ## it.
+        ##
+        ## TODO: make this less lame.
         elsif m.header["Content-Type"] && m.header["Content-Type"] !~ /^text\/plain/
-          "sup-attachment-#{Time.now.to_i}-#{rand 10000}"
+          extension =
+            case m.header["Content-Type"]
+            when /text\/html/: "html"
+            when /image\/(.*)/: $1
+            end
+
+          ["sup-attachment-#{Time.now.to_i}-#{rand 10000}", extension].join(".")
         end
 
       ## if there's a filename, we'll treat it as an attachment.
@@ -363,17 +409,18 @@ private
 
       ## otherwise, it's body text
       else
-        body = Message.convert_from m.decode, m.charset
-        text_to_chunks body.normalize_whitespace.split("\n")
+        body = Message.convert_from m.decode, m.charset if m.body
+        text_to_chunks((body || "").normalize_whitespace.split("\n"), encrypted)
       end
     end
   end
 
   def self.convert_from body, charset
+    charset = "utf-8" if charset =~ /UTF_?8/i
     begin
       raise MessageFormatError, "RubyMail decode returned a null body" unless body
       return body unless charset
-      Iconv.iconv($encoding, charset, body).join
+      Iconv.iconv($encoding + "//IGNORE", charset, body + " ").join[0 .. -2]
     rescue Errno::EINVAL, Iconv::InvalidEncoding, Iconv::IllegalSequence, MessageFormatError => e
       Redwood::log "warning: error (#{e.class.name}) decoding message body from #{charset}: #{e.message}"
       File.open("sup-unable-to-decode.txt", "w") { |f| f.write body }
@@ -384,7 +431,7 @@ private
   ## parse the lines of text into chunk objects.  the heuristics here
   ## need tweaking in some nice manner. TODO: move these heuristics
   ## into the classes themselves.
-  def text_to_chunks lines
+  def text_to_chunks lines, encrypted
     state = :text # one of :text, :quote, or :sig
     chunks = []
     chunk_lines = []
@@ -396,7 +443,7 @@ private
       when :text
         newstate = nil
 
-        if line =~ QUOTE_PATTERN || (line =~ QUOTE_START_PATTERN && (nextline =~ QUOTE_PATTERN || nextline =~ QUOTE_START_PATTERN))
+        if line =~ QUOTE_PATTERN || (line =~ QUOTE_START_PATTERN && nextline =~ QUOTE_PATTERN)
           newstate = :quote
         elsif line =~ SIG_PATTERN && (lines.length - i) < MAX_SIG_DISTANCE
           newstate = :sig
@@ -415,7 +462,7 @@ private
       when :quote
         newstate = nil
 
-        if line =~ QUOTE_PATTERN || line =~ QUOTE_START_PATTERN #|| line =~ /^\s*$/
+        if line =~ QUOTE_PATTERN || (line =~ /^\s*$/ && nextline =~ QUOTE_PATTERN)
           chunk_lines << line
         elsif line =~ SIG_PATTERN && (lines.length - i) < MAX_SIG_DISTANCE
           newstate = :sig
@@ -436,11 +483,14 @@ private
       when :block_quote, :sig
         chunk_lines << line
       end
- 
+
       if !@have_snippet && state == :text && (@snippet.nil? || @snippet.length < SNIPPET_LEN) && line !~ /[=\*#_-]{3,}/ && line !~ /^\s*$/
+        @snippet ||= ""
         @snippet += " " unless @snippet.empty?
         @snippet += line.gsub(/^\s+/, "").gsub(/[\r\n]/, "").gsub(/\s+/, " ")
         @snippet = @snippet[0 ... SNIPPET_LEN].chomp
+        @dirty = true unless encrypted && $config[:discard_snippets_from_encrypted_messages]
+        @snippet_contains_encrypted_content = true if encrypted
       end
     end