mhlib.py

来自「mallet是自然语言处理、机器学习领域的一个开源项目。」· Python 代码 · 共 1,004 行 · 第 1/3 页

        mimetools.Message.__init__(self, fp)

    def __repr__(self):
        """String representation."""
        return 'Message(%s, %s)' % (repr(self.folder), self.number)

    def getheadertext(self, pred = None):
        """Return the message's header text as a string.  If an
        argument is specified, it is used as a filter predicate to
        decide which headers to return (its argument is the header
        name converted to lower case)."""
        if not pred:
            return ''.join(self.headers)
        headers = []
        hit = 0
        for line in self.headers:
            if not line[0].isspace():
                i = line.find(':')
                if i > 0:
                    hit = pred(line[:i].lower())
            if hit: headers.append(line)
        return ''.join(headers)

    def getbodytext(self, decode = 1):
        """Return the message's body text as string.  This undoes a
        Content-Transfer-Encoding, but does not interpret other MIME
        features (e.g. multipart messages).  To suppress decoding,
        pass 0 as an argument."""
        self.fp.seek(self.startofbody)
        encoding = self.getencoding()
        if not decode or encoding in ('', '7bit', '8bit', 'binary'):
            return self.fp.read()
        from StringIO import StringIO
        output = StringIO()
        mimetools.decode(self.fp, output, encoding)
        return output.getvalue()

    def getbodyparts(self):
        """Only for multipart messages: return the message's body as a
        list of SubMessage objects.  Each submessage object behaves
        (almost) as a Message object."""
        if self.getmaintype() != 'multipart':
            raise Error, 'Content-Type is not multipart/*'
        bdry = self.getparam('boundary')
        if not bdry:
            raise Error, 'multipart/* without boundary param'
        self.fp.seek(self.startofbody)
        mf = multifile.MultiFile(self.fp)
        mf.push(bdry)
        parts = []
        while mf.next():
            n = str(self.number) + '.' + `1 + len(parts)`
            part = SubMessage(self.folder, n, mf)
            parts.append(part)
        mf.pop()
        return parts

    def getbody(self):
        """Return body, either a string or a list of messages."""
        if self.getmaintype() == 'multipart':
            return self.getbodyparts()
        else:
            return self.getbodytext()


class SubMessage(Message):

    def __init__(self, f, n, fp):
        """Constructor."""
        Message.__init__(self, f, n, fp)
        if self.getmaintype() == 'multipart':
            self.body = Message.getbodyparts(self)
        else:
            self.body = Message.getbodytext(self)
        self.bodyencoded = Message.getbodytext(self, decode=0)
            # XXX If this is big, should remember file pointers

    def __repr__(self):
        """String representation."""
        f, n, fp = self.folder, self.number, self.fp
        return 'SubMessage(%s, %s, %s)' % (f, n, fp)

    def getbodytext(self, decode = 1):
        if not decode:
            return self.bodyencoded
        if type(self.body) == type(''):
            return self.body

    def getbodyparts(self):
        if type(self.body) == type([]):
            return self.body

    def getbody(self):
        return self.body


class IntSet:
    """Class implementing sets of integers.

    This is an efficient representation for sets consisting of several
    continuous ranges, e.g. 1-100,200-400,402-1000 is represented
    internally as a list of three pairs: [(1,100), (200,400),
    (402,1000)].  The internal representation is always kept normalized.

    The constructor has up to three arguments:
    - the string used to initialize the set (default ''),
    - the separator between ranges (default ',')
    - the separator between begin and end of a range (default '-')
    The separators must be strings (not regexprs) and should be different.

    The tostring() function yields a string that can be passed to another
    IntSet constructor; __repr__() is a valid IntSet constructor itself.
    """

    # XXX The default begin/end separator means that negative numbers are
    #     not supported very well.
    #
    # XXX There are currently no operations to remove set elements.

    def __init__(self, data = None, sep = ',', rng = '-'):
        self.pairs = []
        self.sep = sep
        self.rng = rng
        if data: self.fromstring(data)

    def reset(self):
        self.pairs = []

    def __cmp__(self, other):
        return cmp(self.pairs, other.pairs)

    def __hash__(self):
        return hash(self.pairs)

    def __repr__(self):
        return 'IntSet(%s, %s, %s)' % (`self.tostring()`,
                  `self.sep`, `self.rng`)

    def normalize(self):
        self.pairs.sort()
        i = 1
        while i < len(self.pairs):
            alo, ahi = self.pairs[i-1]
            blo, bhi = self.pairs[i]
            if ahi >= blo-1:
                self.pairs[i-1:i+1] = [(alo, max(ahi, bhi))]
            else:
                i = i+1

    def tostring(self):
        s = ''
        for lo, hi in self.pairs:
            if lo == hi: t = `lo`
            else: t = `lo` + self.rng + `hi`
            if s: s = s + (self.sep + t)
            else: s = t
        return s

    def tolist(self):
        l = []
        for lo, hi in self.pairs:
            m = range(lo, hi+1)
            l = l + m
        return l

    def fromlist(self, list):
        for i in list:
            self.append(i)

    def clone(self):
        new = IntSet()
        new.pairs = self.pairs[:]
        return new

    def min(self):
        return self.pairs[0][0]

    def max(self):
        return self.pairs[-1][-1]

    def contains(self, x):
        for lo, hi in self.pairs:
            if lo <= x <= hi: return 1
        return 0

    def append(self, x):
        for i in range(len(self.pairs)):
            lo, hi = self.pairs[i]
            if x < lo: # Need to insert before
                if x+1 == lo:
                    self.pairs[i] = (x, hi)
                else:
                    self.pairs.insert(i, (x, x))
                if i > 0 and x-1 == self.pairs[i-1][1]:
                    # Merge with previous
                    self.pairs[i-1:i+1] = [
                            (self.pairs[i-1][0],
                             self.pairs[i][1])
                          ]
                return
            if x <= hi: # Already in set
                return
        i = len(self.pairs) - 1
        if i >= 0:
            lo, hi = self.pairs[i]
            if x-1 == hi:
                self.pairs[i] = lo, x
                return
        self.pairs.append((x, x))

    def addpair(self, xlo, xhi):
        if xlo > xhi: return
        self.pairs.append((xlo, xhi))
        self.normalize()

    def fromstring(self, data):
        new = []
        for part in data.split(self.sep):
            list = []
            for subp in part.split(self.rng):
                s = subp.strip()
                list.append(int(s))
            if len(list) == 1:
                new.append((list[0], list[0]))
            elif len(list) == 2 and list[0] <= list[1]:
                new.append((list[0], list[1]))
            else:
                raise ValueError, 'bad data passed to IntSet'
        self.pairs = self.pairs + new
        self.normalize()


# Subroutines to read/write entries in .mh_profile and .mh_sequences

def pickline(file, key, casefold = 1):
    try:
        f = open(file, 'r')
    except IOError:
        return None
    pat = re.escape(key) + ':'
    prog = re.compile(pat, casefold and re.IGNORECASE)
    while 1:
        line = f.readline()
        if not line: break
        if prog.match(line):
            text = line[len(key)+1:]
            while 1:
                line = f.readline()
                if not line or not line[0].isspace():
                    break
                text = text + line
            return text.strip()
    return None

def updateline(file, key, value, casefold = 1):
    try:
        f = open(file, 'r')
        lines = f.readlines()
        f.close()
    except IOError:
        lines = []
    pat = re.escape(key) + ':(.*)\n'
    prog = re.compile(pat, casefold and re.IGNORECASE)
    if value is None:
        newline = None
    else:
        newline = '%s: %s\n' % (key, value)
    for i in range(len(lines)):
        line = lines[i]
        if prog.match(line):
            if newline is None:
                del lines[i]
            else:
                lines[i] = newline
            break
    else:
        if newline is not None:
            lines.append(newline)
    tempfile = file + "~"
    f = open(tempfile, 'w')
    for line in lines:
        f.write(line)
    f.close()
    os.rename(tempfile, file)


# Test program

def test():
    global mh, f
    os.system('rm -rf $HOME/Mail/@test')
    mh = MH()
    def do(s): print s; print eval(s)
    do('mh.listfolders()')
    do('mh.listallfolders()')
    testfolders = ['@test', '@test/test1', '@test/test2',
                   '@test/test1/test11', '@test/test1/test12',
                   '@test/test1/test11/test111']
    for t in testfolders: do('mh.makefolder(%s)' % `t`)
    do('mh.listsubfolders(\'@test\')')
    do('mh.listallsubfolders(\'@test\')')
    f = mh.openfolder('@test')
    do('f.listsubfolders()')
    do('f.listallsubfolders()')
    do('f.getsequences()')
    seqs = f.getsequences()
    seqs['foo'] = IntSet('1-10 12-20', ' ').tolist()
    print seqs
    f.putsequences(seqs)
    do('f.getsequences()')
    testfolders.reverse()
    for t in testfolders: do('mh.deletefolder(%s)' % `t`)
    do('mh.getcontext()')
    context = mh.getcontext()
    f = mh.openfolder(context)
    do('f.getcurrent()')
    for seq in ['first', 'last', 'cur', '.', 'prev', 'next',
                'first:3', 'last:3', 'cur:3', 'cur:-3',
                'prev:3', 'next:3',
                '1:3', '1:-3', '100:3', '100:-3', '10000:3', '10000:-3',
                'all']:
        try:
            do('f.parsesequence(%s)' % `seq`)
        except Error, msg:
            print "Error:", msg
        stuff = os.popen("pick %s 2>/dev/null" % `seq`).read()
        list = map(int, stuff.split())
        print list, "<-- pick"
    do('f.listmessages()')


if __name__ == '__main__':
    test()