-require 'tempfile'
require 'time'
require 'iconv'
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
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, :snippet_contains_encrypted_content
@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
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}"
@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"]
end
private :parse_header
+ 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 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
@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
## 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
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 " "
## otherwise, it's body text
else
- body = Message.convert_from m.decode, m.charset
- text_to_chunks (body || "").normalize_whitespace.split("\n"), encrypted
+ 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
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
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