- bitmessage could end up having no known nodes and then it would
freeze. Now it shouldn't freeze, however it can still end up without
known nodes until a restart in some cases (e.g. when suspending the
computer for more then 3 days while BM is running)
- you can now use SMTP to send messages
- uses bmaddr.lan domain
- runs on 127.0.0.8425 if you set "smtpd" to True
- mandatory authentication with smtpdusername and smtpdpassword
- handles old dialog versions better if using curses
- can spawn SMTP delivery thread if configured (only when in daemon
mode)
- daemonized mode now works more like it's properly supposed to on unix
(double fork etc). You may have to adjust your init scripts, when
when using upstart for example you should now use "expect daemon"
- daemon mode now cleanly shuts down when TERM/INT signal is received
- PyBitmessage only used to quit on disk full when running in daemon
mode. When this happened with the QT-GUI, it would end up in a
half-frozen status instead. Quitting is a safer choice
Fixes#572
- helper classes for encoding/decoding messages
- includes both old as well as new extended one (msgpack+zlib)
- the classes are unused yet and are supposed to be for experimenting
- when running a hidden service, the IP of the tor relay was a part of
the verack message. In setups where it's not 127.0.0.1 it may leak
info about network topology
- thanks for an anonymous bug report
- will send the correct combination of hostname and port
- if proxyhostname is a hostname and an IP address, it will now allow
multiple parallel connections for hidden service
- PyBitmessage can now run as a hidden service on Tor
- three new variables in keys.dat: onionhostname, onionport, onionbindip
- you need to manually add a hidden service to tor
- bitmsghash should now build and run on BSD (thanks for
FreeBSD/Dragonfly maintainers for assistance)
- if it cannot detect the number of cores, will default to one thread
(previously it broke)
Two file merge conflicts, __init__.py and upnp.py, were not resolved
correctly by the automatic resolving (probably because the affected code
was written by other people and I merged them into mailchuck fork). This
changes it to the same code that is in the mailchuck fork)
