module MBox
class Loader < Source
- def initialize uri_or_fp, start_offset=nil, usual=true, archived=false, id=nil
- super
+ yaml_properties :uri, :cur_offset, :usual, :archived, :id, :labels
+ ## uri_or_fp is horrific. need to refactor.
+ def initialize uri_or_fp, start_offset=nil, usual=true, archived=false, id=nil, labels=[]
@mutex = Mutex.new
- @labels = [:unread]
+ @labels = ((labels || []) - LabelManager::RESERVED_LABELS).uniq.freeze
case uri_or_fp
when String
- raise ArgumentError, "not an mbox uri" unless uri_or_fp =~ %r!mbox://!
-
- fn = URI(uri_or_fp).path
- ## heuristic: use the filename as a label, unless the file
- ## has a path that probably represents an inbox.
- @labels << File.basename(fn).intern unless File.dirname(fn) =~ /\b(var|usr|spool)\b/
- @f = File.open fn
+ uri = URI(Source.expand_filesystem_uri(uri_or_fp))
+ raise ArgumentError, "not an mbox uri" unless uri.scheme == "mbox"
+ raise ArgumentError, "mbox URI ('#{uri}') cannot have a host: #{uri.host}" if uri.host
+ raise ArgumentError, "mbox URI must have a path component" unless uri.path
+ @f = File.open uri.path
+ @path = uri.path
else
@f = uri_or_fp
+ @path = uri_or_fp.path
+ end
+
+ super uri_or_fp, start_offset, usual, archived, id
+ end
+
+ def file_path; @path end
+ def is_source_for? uri; super || (self.uri.is_a?(String) && (URI(Source.expand_filesystem_uri(uri)) == URI(Source.expand_filesystem_uri(self.uri)))) end
+
+ def self.suggest_labels_for path
+ ## heuristic: use the filename as a label, unless the file
+ ## has a path that probably represents an inbox.
+ if File.dirname(path) =~ /\b(var|usr|spool)\b/
+ []
+ else
+ [File.basename(path).downcase.intern]
end
end
def check
- if cur_offset > end_offset
+ if (cur_offset ||= start_offset) > end_offset
raise OutOfSyncSourceError, "mbox file is smaller than last recorded message offset. Messages have probably been deleted by another client."
end
end
@f.seek offset
begin
RMail::Mailbox::MBoxReader.new(@f).each_message do |input|
- return RMail::Parser.read(input)
+ m = RMail::Parser.read(input)
+ if m.body && m.body.is_a?(String)
+ m.body.gsub!(/^>From /, "From ")
+ end
+ return m
end
rescue RMail::Parser::Error => e
raise FatalSourceError, "error parsing mbox file: #{e.message}"
ret
end
- def raw_full_message offset
+ def raw_message offset
ret = ""
+ each_raw_message_line(offset) { |l| ret += l }
+ ret
+ 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.
+ ##
+ ## i hoped never to have to move shit around on disk but
+ ## sup-sync-back has to do it.
+ def each_raw_message_line offset
@mutex.synchronize do
@f.seek offset
- @f.gets # skip mbox header
+ yield @f.gets
until @f.eof? || (l = @f.gets) =~ BREAK_RE
- ret += l
+ yield l
end
end
- ret
end
def next
next_offset = @f.tell
end
end
- rescue SystemCallError => e
+ rescue SystemCallError, IOError => e
raise FatalSourceError, "Error reading #{@f.path}: #{e.message}"
end
self.cur_offset = next_offset
- [returned_offset, @labels.clone]
+ [returned_offset, (self.labels + [:unread]).uniq]
end
end
-Redwood::register_yaml(Loader, %w(uri cur_offset usual archived id))
-
end
end