This file is indexed.

/usr/bin/sb_mboxtrain is in spambayes 1.1b1-1.1.

This file is owned by root:root, with mode 0o755.

The actual contents of the file can be viewed below.

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
#! /usr/bin/python

### Train spambayes on all previously-untrained messages in a mailbox.
###
### This keeps track of messages it's already trained by adding an
### X-Spambayes-Trained: header to each one.  Then, if you move one to
### another folder, it will retrain that message.  You would want to run
### this from a cron job on your server.

"""Usage: %(program)s [OPTIONS] ...

Where OPTIONS is one or more of:
    -h
        show usage and exit
    -d DBNAME
        use the DBM store.  A DBM file is larger than the pickle and
        creating it is slower, but loading it is much faster,
        especially for large word databases.  Recommended for use with
        sb_filter or any procmail-based filter.
    -p DBNAME
        use the pickle store.  A pickle is smaller and faster to create,
        but much slower to load.  Recommended for use with sb_server and
        sb_xmlrpcserver.
    -g PATH
        mbox or directory of known good messages (non-spam) to train on.
        Can be specified more than once.
    -s PATH
        mbox or directory of known spam messages to train on.
        Can be specified more than once.
    -f
        force training, ignoring the trained header.  Use this if you
        need to rebuild your database from scratch.
    -q
        quiet mode; no output

    -n  train mail residing in "new" directory, in addition to "cur"
        directory, which is always trained (Maildir only)

    -r  remove mail which was trained on (Maildir only)

    -o section:option:value
        set [section, option] in the options database to value
"""

import sys, os, getopt, email
import shutil
from spambayes import hammie, storage, mboxutils
from spambayes.Options import options, get_pathname_option

program = sys.argv[0]
loud = True

def get_message(obj):
    """Return an email Message object.

    This works like mboxutils.get_message, except it doesn't junk the
    headers if there's an error.  Doing so would cause a headerless
    message to be written back out!

    """

    if isinstance(obj, email.Message.Message):
        return obj
    # Create an email Message object.
    if hasattr(obj, "read"):
        obj = obj.read()
    try:
        msg = email.message_from_string(obj)
    except email.Errors.MessageParseError:
        msg = None
    return msg

def msg_train(h, msg, is_spam, force):
    """Train bayes with a single message."""

    # XXX: big hack -- why is email.Message unable to represent
    # multipart/alternative?
    try:
        mboxutils.as_string(msg)
    except TypeError:
        # We'll be unable to represent this as text :(
        return False

    if is_spam:
        spamtxt = options["Headers", "header_spam_string"]
    else:
        spamtxt = options["Headers", "header_ham_string"]
    oldtxt = msg.get(options["Headers", "trained_header_name"])
    if force:
        # Train no matter what.
        if oldtxt != None:
            del msg[options["Headers", "trained_header_name"]]
    elif oldtxt == spamtxt:
        # Skip this one, we've already trained with it.
        return False
    elif oldtxt != None:
        # It's been trained, but as something else.  Untrain.
        del msg[options["Headers", "trained_header_name"]]
        h.untrain(msg, not is_spam)
    h.train(msg, is_spam)
    msg.add_header(options["Headers", "trained_header_name"], spamtxt)

    return True

def maildir_train(h, path, is_spam, force, removetrained):
    """Train bayes with all messages from a maildir."""

    if loud:
        print "  Reading %s as Maildir" % (path,)

    import time
    import socket

    pid = os.getpid()
    host = socket.gethostname()
    counter = 0
    trained = 0

    for fn in os.listdir(path):
        cfn = os.path.join(path, fn)
        tfn = os.path.normpath(os.path.join(path, "..", "tmp",
                           "%d.%d_%d.%s" % (time.time(), pid,
                                            counter, host)))
        if (os.path.isdir(cfn)):
            continue
        counter += 1
        if loud and counter % 10 == 0:
            sys.stdout.write("\r%6d" % counter)
            sys.stdout.flush()
        f = file(cfn, "rb")
        msg = get_message(f)
        f.close()
        if not msg:
            print "Malformed message: %s.  Skipping..." % cfn
            continue
        if not msg_train(h, msg, is_spam, force):
            continue
        trained += 1
        if not options["Headers", "include_trained"]:
            continue
        f = file(tfn, "wb")
        f.write(mboxutils.as_string(msg))
        f.close()
        shutil.copystat(cfn, tfn)

        # XXX: This will raise an exception on Windows.  Do any Windows
        # people actually use Maildirs?
        os.rename(tfn, cfn)
        if (removetrained):
            os.unlink(cfn)

    if loud:
        sys.stdout.write("\r%6d" % counter)
        sys.stdout.write("\r  Trained %d out of %d messages\n" %
                         (trained, counter))

def mbox_train(h, path, is_spam, force):
    """Train bayes with a Unix mbox"""

    if loud:
        print "  Reading as Unix mbox"

    import mailbox
    import fcntl

    # Open and lock the mailbox.  Some systems require it be opened for
    # writes in order to assert an exclusive lock.
    f = file(path, "r+b")
    fcntl.flock(f, fcntl.LOCK_EX)
    mbox = mailbox.PortableUnixMailbox(f, get_message)

    outf = os.tmpfile()
    counter = 0
    trained = 0

    for msg in mbox:
        if not msg:
            print "Malformed message number %d.  I can't train on this mbox, sorry." % counter
            return
        counter += 1
        if loud and counter % 10 == 0:
            sys.stdout.write("\r%6d" % counter)
            sys.stdout.flush()
        if msg_train(h, msg, is_spam, force):
            trained += 1
        if options["Headers", "include_trained"]:
            # Write it out with the Unix "From " line
            outf.write(mboxutils.as_string(msg, True))

    if options["Headers", "include_trained"]:
        outf.seek(0)
        try:
            os.ftruncate(f.fileno(), 0)
            f.seek(0)
        except:
            # If anything goes wrong, don't try to write
            print "Problem truncating mbox--nothing written"
            raise
        try:
            for line in outf.xreadlines():
                f.write(line)
        except:
            print >> sys.stderr ("Problem writing mbox!  Sorry, "
                                 "I tried my best, but your mail "
                                 "may be corrupted.")
            raise

    fcntl.flock(f, fcntl.LOCK_UN)
    f.close()
    if loud:
        sys.stdout.write("\r%6d" % counter)
        sys.stdout.write("\r  Trained %d out of %d messages\n" %
                         (trained, counter))

def mhdir_train(h, path, is_spam, force):
    """Train bayes with an mh directory"""

    if loud:
        print "  Reading as MH mailbox"

    import glob

    counter = 0
    trained = 0

    for fn in glob.glob(os.path.join(path, "[0-9]*")):
        counter += 1

        cfn = fn
        tfn = os.path.join(path, "spambayes.tmp")
        if loud and counter % 10 == 0:
            sys.stdout.write("\r%6d" % counter)
            sys.stdout.flush()
        f = file(fn, "rb")
        msg = get_message(f)
        f.close()
        if not msg:
            print "Malformed message: %s.  Skipping..." % cfn
            continue
        msg_train(h, msg, is_spam, force)
        trained += 1
        if not options["Headers", "include_trained"]:
            continue
        f = file(tfn, "wb")
        f.write(mboxutils.as_string(msg))
        f.close()
        shutil.copystat(cfn, tfn)

        # XXX: This will raise an exception on Windows.  Do any Windows
        # people actually use MH directories?
        os.rename(tfn, cfn)

    if loud:
        sys.stdout.write("\r%6d" % counter)
        sys.stdout.write("\r  Trained %d out of %d messages\n" %
                         (trained, counter))

def train(h, path, is_spam, force, trainnew, removetrained):
    if not os.path.exists(path):
        raise ValueError("Nonexistent path: %s" % path)
    elif os.path.isfile(path):
        mbox_train(h, path, is_spam, force)
    elif os.path.isdir(os.path.join(path, "cur")):
        maildir_train(h, os.path.join(path, "cur"), is_spam, force,
                      removetrained)
        if trainnew:
            maildir_train(h, os.path.join(path, "new"), is_spam, force,
                          removetrained)
    elif os.path.isdir(path):
        mhdir_train(h, path, is_spam, force)
    else:
        raise ValueError("Unable to determine mailbox type: " + path)


def usage(code, msg=''):
    """Print usage message and sys.exit(code)."""
    if msg:
        print >> sys.stderr, msg
        print >> sys.stderr
    print >> sys.stderr, __doc__ % globals()
    sys.exit(code)

def main():
    """Main program; parse options and go."""

    global loud

    try:
        opts, args = getopt.getopt(sys.argv[1:], 'hfqnrd:p:g:s:o:')
    except getopt.error, msg:
        usage(2, msg)

    if not opts:
        usage(2, "No options given")

    force = False
    trainnew = False
    removetrained = False
    good = []
    spam = []
    for opt, arg in opts:
        if opt == '-h':
            usage(0)
        elif opt == "-f":
            force = True
        elif opt == "-n":
            trainnew = True
        elif opt == "-q":
            loud = False
        elif opt == '-g':
            good.append(arg)
        elif opt == '-s':
            spam.append(arg)
        elif opt == "-r":
            removetrained = True
        elif opt == '-o':
            options.set_from_cmdline(arg, sys.stderr)
    pck, usedb = storage.database_type(opts)
    if args:
        usage(2, "Positional arguments not allowed")

    if usedb == None:
        # Use settings in configuration file.
        usedb = options["Storage", "persistent_use_database"]
        pck = get_pathname_option("Storage",
                                          "persistent_storage_file")

    h = hammie.open(pck, usedb, "c")

    for g in good:
        if loud:
            print "Training ham (%s):" % g
        train(h, g, False, force, trainnew, removetrained)
        sys.stdout.flush()
        save = True

    for s in spam:
        if loud:
            print "Training spam (%s):" % s
        train(h, s, True, force, trainnew, removetrained)
        sys.stdout.flush()
        save = True

    if save:
        h.store()


if __name__ == "__main__":
    main()