bug with '>From'

  • NeoM

    NeoM - 2003-07-25

    in any message with

    X-Spam-Status: No, hits=0.000014 required=0.900000

    last line added by bmf (0.9.4; openbsd pkg)
    X is anything

    ideas ?

    • Anonymous - 2003-10-11

      yeah... i have same problem.

      bmf reads all incoming mail with:

      :0 fw
      | bmf -p -k 50

      randomly, when an incoming email has a:

      >From whatever whatever:
      bla blah balh

      i get an X-Spam-Status: message like you saw, which is a duplicate of the one inserted in the headers.

      i've been saving these mails for a little while and have fallen short in visually analyzing them to find what trips bmf off like this.  also, i've c/p'd some of the messages and resent them to myself at the scanned address; once a resend gave me the duplicate X-Spam-Status, but then i started modifying it to see if it would continue to happen, and now even the original message, saved from mutt and passed to bmf via commandline, doesn't give the X-Spam-Status duplicate.

      so far it seems like the only thing i have in common is the '>From' lines all have nospace between > and F, the F is capital, has at least one punctuation ( "," "." "-" "_" "/" "\" "?" "'" "(" ")" ":"  ) -- most have two, but one just has a terminating comma, and the '>From' line is not the last line of the email ( ie, content follows it, but not necessarily on the next line. ).

      the messages i have saved into my ">From" bugfolder are all from misc@openbsd, so i could point anyone to the postings at an archive.

      i just took one of the messages with the duplicate ">From" and in mutt 'e'dited the message, deleted the extra X-Spam-Status line, and did a 'bmf -tp < mboxfile' and it did not create the extra X-Spam-Status line; but if i take the actual bodytext of the message, and physically c/p that out of the message and into a file ( ie no email headers, just the message from the person ), and then 'bmf -tp < justtextfile', it creates the extra X-Spam-Status every time.

      while i still have 264 chiptunes in the xmp -R * queue, i'm gonna have another go to see if i can make a good analysis of it  ( ie, duplicatable and predictable ).


    • Anonymous - 2003-10-11

      oh wow, nevermind.

      looks like i'm utterly superseded by the bmf-users mailing list.

      tee hee


Log in to post a comment.

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:

No, thanks