lib/sup/message.rb in sup-0.7 vs lib/sup/message.rb in sup-0.8

- old
+ new

@@ -1,22 +1,19 @@ require 'time' -require 'iconv' module Redwood -class MessageFormatError < StandardError; end - ## a Message is what's threaded. ## ## it is also where the parsing for quotes and signatures is done, but ## that should be moved out to a separate class at some point (because ## 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. +## this class catches 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 @@ -48,91 +45,90 @@ @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] @snippet_contains_encrypted_content = false @have_snippet = !(opts[:snippet].nil? || opts[:snippet].empty?) - @labels = [] + (opts[:labels] || []) + @labels = (opts[:labels] || []).to_set_of_symbols @dirty = false @encrypted = false @chunks = nil @attachments = [] ## we need to initialize this. see comments in parse_header as to ## why. @refs = [] - parse_header(opts[:header] || @source.load_header(@source_info)) + #parse_header(opts[:header] || @source.load_header(@source_info)) end def parse_header header - header.keys.each { |k| header[k.downcase] = header[k] } # canonicalize + ## forcibly decode these headers from and to the current encoding, + ## which serves to strip out characters that aren't displayable + ## (and which would otherwise be screwing up the display) + %w(from to subject cc bcc).each do |f| + header[f] = Iconv.easy_decode($encoding, $encoding, header[f]) if header[f] + end - fakeid = nil - fakename = nil + @id = if header["message-id"] + mid = header["message-id"] =~ /<(.+?)>/ ? $1 : header["message-id"] + sanitize_message_id mid + else + id = "sup-faked-" + Digest::MD5.hexdigest(raw_header) + from = header["from"] + #Redwood::log "faking non-existent message-id for message from #{from}: #{id}" + id + end - @id = - if header["message-id"] - sanitize_message_id header["message-id"] - else - fakeid = "sup-faked-" + Digest::MD5.hexdigest(raw_header) - end - - @from = - if header["from"] - PersonManager.person_for header["from"] - else - fakename = "Sup Auto-generated Fake Sender <sup@fake.sender.example.com>" - PersonManager.person_for fakename - end + @from = Person.from_address(if header["from"] + header["from"] + else + name = "Sup Auto-generated Fake Sender <sup@fake.sender.example.com>" + #Redwood::log "faking non-existent sender for message #@id: #{name}" + name + end) - Redwood::log "faking message-id for message from #@from: #{id}" if fakeid - Redwood::log "faking from for message #@id: #{fakename}" if fakename - - date = header["date"] - @date = - case date - when Time - date - when String - begin - Time.parse date - rescue ArgumentError => e - Redwood::log "faking date header for #{@id} due to error parsing date #{header['date'].inspect}: #{e.message}" - Time.now - end - else - Redwood::log "faking date header for #{@id}" + @date = case(date = header["date"]) + when Time + date + when String + begin + Time.parse date + rescue ArgumentError => e + #Redwood::log "faking mangled date header for #{@id} (orig #{header['date'].inspect} gave error: #{e.message})" Time.now end + else + #Redwood::log "faking non-existent date header for #{@id}" + Time.now + end @subj = header.member?("subject") ? header["subject"].gsub(/\s+/, " ").gsub(/\s+$/, "") : DEFAULT_SUBJECT - @to = PersonManager.people_for header["to"] - @cc = PersonManager.people_for header["cc"] - @bcc = PersonManager.people_for header["bcc"] + @to = Person.from_address_list header["to"] + @cc = Person.from_address_list header["cc"] + @bcc = Person.from_address_list header["bcc"] ## 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"] + @replyto = Person.from_address header["reply-to"] @list_address = if header["list-post"] - @list_address = PersonManager.person_for header["list-post"].gsub(/^<mailto:|>$/, "") + @list_address = Person.from_address header["list-post"].gsub(/^<mailto:|>$/, "") else nil end @recipient_email = header["envelope-to"] || header["x-original-to"] || header["delivered-to"] @source_marked_read = header["status"] == "RO" @list_subscribe = header["list-subscribe"] @list_unsubscribe = header["list-unsubscribe"] end - private :parse_header def add_ref ref @refs << ref @dirty = true end @@ -170,11 +166,11 @@ end def has_label? t; @labels.member? t; end def add_label t return if @labels.member? t - @labels.push t + @labels = (@labels + [t]).to_set_of_symbols @dirty = true end def remove_label t return unless @labels.member? t @labels.delete t @@ -184,11 +180,11 @@ def recipients @to + @cc + @bcc end def labels= l - @labels = l + @labels = l.to_set_of_symbols @dirty = true end def chunks load_from_source! @@ -196,11 +192,11 @@ end ## this is called when the message body needs to actually be loaded. def load_from_source! @chunks ||= - if @source.has_errors? + if @source.respond_to?(:has_errors?) && @source.has_errors? [Chunk::Text.new(error_message(@source.error.message).split("\n"))] else begin ## we need to re-read the header because it contains information ## that we don't store in the index. actually i think it's just @@ -210,11 +206,11 @@ ## bloat the index. ## actually, it's also the differentiation between to/cc/bcc, ## so i will keep this. parse_header @source.load_header(@source_info) message_to_chunks @source.load_message(@source_info) - rescue SourceError, SocketError, MessageFormatError => e + rescue SourceError, SocketError => e 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 @@ -370,10 +366,11 @@ decryptedm, sig, notice = CryptoManager.decrypt payload children = message_to_chunks(decryptedm, true) if decryptedm [notice, sig, children].flatten.compact end + ## takes a RMail::Message, breaks it into Chunk:: classes. def message_to_chunks m, encrypted=false, sibling_types=[] if m.multipart? chunks = case m.header.content_type when "multipart/signed" @@ -389,11 +386,11 @@ chunks elsif m.header.content_type == "message/rfc822" payload = RMail::Parser.read(m.body) from = payload.header.from.first - from_person = from ? PersonManager.person_for(from.format) : nil + from_person = from ? Person.from_address(from.format) : nil [Chunk::EnclosedMessage.new(from_person, payload.to_s)] + message_to_chunks(payload, encrypted) else filename = ## first, paw through the headers looking for a filename @@ -421,28 +418,20 @@ # add this to the attachments list if its not a generated html # attachment (should we allow images with generated names?). # Lowercase the filename because searches are easier that way @attachments.push filename.downcase unless filename =~ /^sup-attachment-/ add_label :attachment unless filename =~ /^sup-attachment-/ - [Chunk::Attachment.new(m.header.content_type, filename, m, sibling_types)] + content_type = m.header.content_type || "application/unknown" # sometimes RubyMail gives us nil + [Chunk::Attachment.new(content_type, filename, m, sibling_types)] ## otherwise, it's body text else - body = Message.convert_from m.decode, m.charset if m.body + ## if there's no charset, use the current encoding as the charset. + ## this ensures that the body is normalized to avoid non-displayable + ## characters + body = Iconv.easy_decode($encoding, m.charset || $encoding, m.decode) if m.body text_to_chunks((body || "").normalize_whitespace.split("\n"), encrypted) end - end - end - - def self.convert_from body, charset - begin - raise MessageFormatError, "RubyMail decode returned a null body" unless body - return body unless charset - Iconv.easy_decode($encoding, charset, body) - 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(File.join(BASE_DIR,"unable-to-decode.txt"), "w") { |f| f.write body } - body end end ## parse the lines of text into chunk objects. the heuristics here ## need tweaking in some nice manner. TODO: move these heuristics