On 3 Sep 2014, at 12:47, Stevan Bajić <ste...@bajic.ch> wrote: > I don't know how to intelligently catch this issue in the code (it has been a > while since I looked or touched the code). IMHO piping data to the DSPAM > binary should not suffer from this dot stuffing. If you would send that > directly to LMTP/DLMTP socket then I would say that it is clearly your > problem but piping? I don't know.
Obviously, if the protocol is between the dspamc and dspamd programs, those should be fixed so that the pipe through the client is transparent to the user. But if LMTP is employed by the user directly, he is responsible for dot-stuffing. Nowadays every MTA implements LMTP correctly and this is always preferred. Thanks for letting us know, anyway—I planned to use SQLite for my next project, which would have meant invoking dspam, directly. So I will continue to do that for now. IMHO the penalty of fork() these days is greatly overstated, but yes this should be properly fixed. ------------------------------------------------------------------------------ Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/ _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
