tarfile.py

        """Return the object as a ustar header block.
        """
        info["magic"] = POSIX_MAGIC

        if len(info["linkname"]) > LENGTH_LINK:
            raise ValueError("linkname is too long")

        if len(info["name"]) > LENGTH_NAME:
            info["prefix"], info["name"] = self._posix_split_name(info["name"])

        return self._create_header(info, USTAR_FORMAT)

    def create_gnu_header(self, info):
        """Return the object as a GNU header block sequence.
        """
        info["magic"] = GNU_MAGIC

        buf = ""
        if len(info["linkname"]) > LENGTH_LINK:
            buf += self._create_gnu_long_header(info["linkname"], GNUTYPE_LONGLINK)

        if len(info["name"]) > LENGTH_NAME:
            buf += self._create_gnu_long_header(info["name"], GNUTYPE_LONGNAME)

        return buf + self._create_header(info, GNU_FORMAT)

    def create_pax_header(self, info, encoding, errors):
        """Return the object as a ustar header block. If it cannot be
           represented this way, prepend a pax extended header sequence
           with supplement information.
        """
        info["magic"] = POSIX_MAGIC
        pax_headers = self.pax_headers.copy()

        # Test string fields for values that exceed the field length or cannot
        # be represented in ASCII encoding.
        for name, hname, length in (
                ("name", "path", LENGTH_NAME), ("linkname", "linkpath", LENGTH_LINK),
                ("uname", "uname", 32), ("gname", "gname", 32)):

            if hname in pax_headers:
                # The pax header has priority.
                continue

            val = info[name].decode(encoding, errors)

            # Try to encode the string as ASCII.
            try:
                val.encode("ascii")
            except UnicodeEncodeError:
                pax_headers[hname] = val
                continue

            if len(info[name]) > length:
                pax_headers[hname] = val

        # Test number fields for values that exceed the field limit or values
        # that like to be stored as float.
        for name, digits in (("uid", 8), ("gid", 8), ("size", 12), ("mtime", 12)):
            if name in pax_headers:
                # The pax header has priority. Avoid overflow.
                info[name] = 0
                continue

            val = info[name]
            if not 0 <= val < 8 ** (digits - 1) or isinstance(val, float):
                pax_headers[name] = unicode(val)
                info[name] = 0

        # Create a pax extended header if necessary.
        if pax_headers:
            buf = self._create_pax_generic_header(pax_headers)
        else:
            buf = ""

        return buf + self._create_header(info, USTAR_FORMAT)

    @classmethod
    def create_pax_global_header(cls, pax_headers):
        """Return the object as a pax global header block sequence.
        """
        return cls._create_pax_generic_header(pax_headers, type=XGLTYPE)

    def _posix_split_name(self, name):
        """Split a name longer than 100 chars into a prefix
           and a name part.
        """
        prefix = name[:LENGTH_PREFIX + 1]
        while prefix and prefix[-1] != "/":
            prefix = prefix[:-1]

        name = name[len(prefix):]
        prefix = prefix[:-1]

        if not prefix or len(name) > LENGTH_NAME:
            raise ValueError("name is too long")
        return prefix, name

    @staticmethod
    def _create_header(info, format):
        """Return a header block. info is a dictionary with file
           information, format must be one of the *_FORMAT constants.
        """
        parts = [
            stn(info.get("name", ""), 100),
            itn(info.get("mode", 0) & 07777, 8, format),
            itn(info.get("uid", 0), 8, format),
            itn(info.get("gid", 0), 8, format),
            itn(info.get("size", 0), 12, format),
            itn(info.get("mtime", 0), 12, format),
            "        ", # checksum field
            info.get("type", REGTYPE),
            stn(info.get("linkname", ""), 100),
            stn(info.get("magic", POSIX_MAGIC), 8),
            stn(info.get("uname", ""), 32),
            stn(info.get("gname", ""), 32),
            itn(info.get("devmajor", 0), 8, format),
            itn(info.get("devminor", 0), 8, format),
            stn(info.get("prefix", ""), 155)
        ]

        buf = struct.pack("%ds" % BLOCKSIZE, "".join(parts))
        chksum = calc_chksums(buf[-BLOCKSIZE:])[0]
        buf = buf[:-364] + "%06o\0" % chksum + buf[-357:]
        return buf

    @staticmethod
    def _create_payload(payload):
        """Return the string payload filled with zero bytes
           up to the next 512 byte border.
        """
        blocks, remainder = divmod(len(payload), BLOCKSIZE)
        if remainder > 0:
            payload += (BLOCKSIZE - remainder) * NUL
        return payload

    @classmethod
    def _create_gnu_long_header(cls, name, type):
        """Return a GNUTYPE_LONGNAME or GNUTYPE_LONGLINK sequence
           for name.
        """
        name += NUL

        info = {}
        info["name"] = "././@LongLink"
        info["type"] = type
        info["size"] = len(name)
        info["magic"] = GNU_MAGIC

        # create extended header + name blocks.
        return cls._create_header(info, USTAR_FORMAT) + \
                cls._create_payload(name)

    @classmethod
    def _create_pax_generic_header(cls, pax_headers, type=XHDTYPE):
        """Return a POSIX.1-2001 extended or global header sequence
           that contains a list of keyword, value pairs. The values
           must be unicode objects.
        """
        records = []
        for keyword, value in pax_headers.iteritems():
            keyword = keyword.encode("utf8")
            value = value.encode("utf8")
            l = len(keyword) + len(value) + 3   # ' ' + '=' + '\n'
            n = p = 0
            while True:
                n = l + len(str(p))
                if n == p:
                    break
                p = n
            records.append("%d %s=%s\n" % (p, keyword, value))
        records = "".join(records)

        # We use a hardcoded "././@PaxHeader" name like star does
        # instead of the one that POSIX recommends.
        info = {}
        info["name"] = "././@PaxHeader"
        info["type"] = type
        info["size"] = len(records)
        info["magic"] = POSIX_MAGIC

        # Create pax header + record blocks.
        return cls._create_header(info, USTAR_FORMAT) + \
                cls._create_payload(records)

    @classmethod
    def frombuf(cls, buf):
        """Construct a TarInfo object from a 512 byte string buffer.
        """
        if len(buf) == 0:
            raise EmptyHeaderError("empty header")
        if len(buf) != BLOCKSIZE:
            raise TruncatedHeaderError("truncated header")
        if buf.count(NUL) == BLOCKSIZE:
            raise EOFHeaderError("end of file header")

        chksum = nti(buf[148:156])
        if chksum not in calc_chksums(buf):
            raise InvalidHeaderError("bad checksum")

        obj = cls()
        obj.buf = buf
        obj.name = nts(buf[0:100])
        obj.mode = nti(buf[100:108])
        obj.uid = nti(buf[108:116])
        obj.gid = nti(buf[116:124])
        obj.size = nti(buf[124:136])
        obj.mtime = nti(buf[136:148])
        obj.chksum = chksum
        obj.type = buf[156:157]
        obj.linkname = nts(buf[157:257])
        obj.uname = nts(buf[265:297])
        obj.gname = nts(buf[297:329])
        obj.devmajor = nti(buf[329:337])
        obj.devminor = nti(buf[337:345])
        prefix = nts(buf[345:500])

        # Old V7 tar format represents a directory as a regular
        # file with a trailing slash.
        if obj.type == AREGTYPE and obj.name.endswith("/"):
            obj.type = DIRTYPE

        # Remove redundant slashes from directories.
        if obj.isdir():
            obj.name = obj.name.rstrip("/")

        # Reconstruct a ustar longname.
        if prefix and obj.type not in GNU_TYPES:
            obj.name = prefix + "/" + obj.name
        return obj

    @classmethod
    def fromtarfile(cls, tarfile):
        """Return the next TarInfo object from TarFile object
           tarfile.
        """
        buf = tarfile.fileobj.read(BLOCKSIZE)
        obj = cls.frombuf(buf)
        obj.offset = tarfile.fileobj.tell() - BLOCKSIZE
        return obj._proc_member(tarfile)

    #--------------------------------------------------------------------------
    # The following are methods that are called depending on the type of a
    # member. The entry point is _proc_member() which can be overridden in a
    # subclass to add custom _proc_*() methods. A _proc_*() method MUST
    # implement the following
    # operations:
    # 1. Set self.offset_data to the position where the data blocks begin,
    #    if there is data that follows.
    # 2. Set tarfile.offset to the position where the next member's header will
    #    begin.
    # 3. Return self or another valid TarInfo object.
    def _proc_member(self, tarfile):
        """Choose the right processing method depending on
           the type and call it.
        """
        if self.type in (GNUTYPE_LONGNAME, GNUTYPE_LONGLINK):
            return self._proc_gnulong(tarfile)
        elif self.type == GNUTYPE_SPARSE:
            return self._proc_sparse(tarfile)
        elif self.type in (XHDTYPE, XGLTYPE, SOLARIS_XHDTYPE):
            return self._proc_pax(tarfile)
        else:
            return self._proc_builtin(tarfile)

    def _proc_builtin(self, tarfile):
        """Process a builtin type or an unknown type which
           will be treated as a regular file.
        """
        self.offset_data = tarfile.fileobj.tell()
        offset = self.offset_data
        if self.isreg() or self.type not in SUPPORTED_TYPES:
            # Skip the following data blocks.
            offset += self._block(self.size)
        tarfile.offset = offset

        # Patch the TarInfo object with saved global
        # header information.
        self._apply_pax_info(tarfile.pax_headers, tarfile.encoding, tarfile.errors)

        return self

    def _proc_gnulong(self, tarfile):
        """Process the blocks that hold a GNU longname
           or longlink member.
        """
        buf = tarfile.fileobj.read(self._block(self.size))

        # Fetch the next header and process it.
        try:
            next = self.fromtarfile(tarfile)
        except HeaderError:
            raise SubsequentHeaderError("missing or bad subsequent header")

        # Patch the TarInfo object from the next header with
        # the longname information.
        next.offset = self.offset
        if self.type == GNUTYPE_LONGNAME:
            next.name = nts(buf)
        elif self.type == GNUTYPE_LONGLINK:
            next.linkname = nts(buf)

        return next

    def _proc_sparse(self, tarfile):
        """Process a GNU sparse header plus extra headers.
        """
        buf = self.buf
        sp = _ringbuffer()
        pos = 386
        lastpos = 0L
        realpos = 0L
        # There are 4 possible sparse structs in the
        # first header.
        for i in xrange(4):
            try:
                offset = nti(buf[pos:pos + 12])
                numbytes = nti(buf[pos + 12:pos + 24])
            except ValueError:
                break
            if offset > lastpos:
                sp.append(_hole(lastpos, offset - lastpos))
            sp.append(_data(offset, numbytes, realpos))
            realpos += numbytes
            lastpos = offset + numbytes
            pos += 24

        isextended = ord(buf[482])
        origsize = nti(buf[483:495])

        # If the isextended flag is given,
        # there are extra headers to process.
        while isextended == 1:
            buf = tarfile.fileobj.read(BLOCKSIZE)
            pos = 0
            for i in xrange(21):
                try:
                    offset = nti(buf[pos:pos + 12])
                    numbytes = nti(buf[pos + 12:pos + 24])
                except ValueError:
                    break
                if offset > lastpos:
                    sp.append(_hole(lastpos, offset - lastpos))
                sp.append(_data(offset, numbytes, realpos))
                realpos += numbytes
                lastpos = offset + numbytes
                pos += 24
            isextended = ord(buf[504])

        if lastpos < origsize:
            sp.append(_hole(lastpos, origsize - lastpos))

        self.sparse = sp

        self.offset_data = tarfile.fileobj.tell()
        tarfile.offset = self.offset_data + self._block(self.size)
        self.size = origsize

        return self

    def _proc_pax(self, tarfile):
        """Process an extended or global header as described in
           POSIX.1-2001.
        """
        # Read the header information.
        buf = tarfile.fileobj.read(self._block(self.size))

        # A pax header stores supplemental information for either
        # the following file (extended) or all following files
        # (global).
        if self.type == XGLTYPE:
            pax_headers = tarfile.pax_headers
        else:
            pax_headers = tarfile.pax_headers.copy()

        # Parse pax header information. A record looks like that:
        # "%d %s=%s\n" % (length, keyword, value). length is the size
        # of the complete record including the length field itself and
        # the newline. keyword and value are both UTF-8 encoded strings.
        regex = re.compile(r"(\d+) ([^=]+)=", re.U)
        pos = 0
        while True:
            match = regex.match(buf, pos)
            if not match:
                break

            length, keyword = match.groups()
            length = int(length)
            value = buf[match.end(2) + 1:match.start(1) + length - 1]

            keyword = keyword.decode("utf8")
            value = value.decode("utf8")

            pax_headers[keyword] = value
            pos += length

        # Fetch the next header.
        try:
            next = self.fromtarfile(tarfile)
        except HeaderError:
            raise SubsequentHeaderError("missing or bad subsequent header")

        if self.type in (XHDTYPE, SOLARIS_XHDTYPE):
            # Patch the TarInfo object with the extended header info.
            next._apply_pax_info(pax_headers, tarfile.encoding, tarfile.errors)
            next.offset = self.offset

            if "size" in pax_headers:
                # If the extended header replaces the size field,
                # we need to recalculate the offset where the next
                # header starts.
                offset = next.offset_data
                if next.isreg() or next.type not in SUPPORTED_TYPES:
                    offset += next._block(next.size)
                tarfile.offset = offset

        return next

    def _apply_pax_info(self, pax_headers, encoding, errors):
        """Replace fields with supplemental information from a previous
           pax extended or global header.
        """
        for keyword, value in pax_headers.iteritems():
            if keyword not in PAX_FIELDS:
                continue

            if keyword == "path":
                value = value.rstrip("/")

            if keyword in PAX_NUMBER_FIELDS:
                try:
                    value = PAX_NUMBER_FIELDS[keyword](value)
                except ValueError:
                    value = 0
            else:
                value = uts(value, encoding, errors)

            setattr(self, keyword, value)

        self.pax_headers = pax_headers.copy()

    def _block(self, count):
        """Round up a byte count by BLOCKSIZE and return it,
           e.g. _block(834) => 1024.
        """
        blocks, remainder = divmod(count, BLOCKSIZE)
        if remainder:
            blocks += 1
        return blocks * BLOCKSIZE

    def isreg(self):
        return self.type in REGULAR_TYPES
    def isfile(self):
        return self.isreg()
    def isdir(self):
        return self.type == DIRTYPE
    def issym(self):
        return self.type == SYMTYPE
    def islnk(self):
        return self.type == LNKTYPE
    def ischr(self):
        return self.type == CHRTYPE
    def isblk(self):
        return self.type == BLKTYPE
    def isfifo(self):
        return self.type == FIFOTYPE
    def issparse(self):
        return self.type == GNUTYPE_SPARSE
    def isdev(self):
        return self.type in (CHRTYPE, BLKTYPE, FIFOTYPE)
# class TarInfo

class TarFile(object):
    """The TarFile Class provides an interface to tar archives.
    """

    debug = 0                   # May be set from 0 (no msgs) to 3 (all msgs)

    dereference = False         # If true, add content of linked file to the
                                # tar file, else the link.

    ignore_zeros = False        # If true, skips empty or invalid blocks and
                                # continues processing.

    errorlevel = 1              # If 0, fatal errors only appear in debug
                                # messages (if debug >= 0). If > 0, errors
                                # are passed to the caller as exceptions.

    format = DEFAULT_FORMAT     # The format to use when creating an archive.

    encoding = ENCODING         # Encoding for 8-bit character strings.

    errors = None               # Error handler for unicode conversion.

    tarinfo = TarInfo           # The default TarInfo class to use.

    fileobject = ExFileObject   # The default ExFileObject class to use.

    def __init__(self, name=None, mode="r", fileobj=None, format=None,
            tarinfo=None, dereference=None, ignore_zeros=None, encoding=None,
            errors=None, pax_headers=None, debug=None, errorlevel=None):
        """Open an (uncompressed) tar archive `name'. `mode' is either 'r' to
           read from an existing archive, 'a' to append data to an existing
           file or 'w' to create a new file overwriting an existing one. `mode'
           defaults to 'r'.
           If `fileobj' is given, it is used for reading or writing data. If it
           can be determined, `mode' is overridden by `fileobj's mode.
           `fileobj' is not closed, when TarFile is closed.
        """
        if len(mode) > 1 or mode not in "raw":
            raise ValueError("mode must be 'r', 'a' or 'w'")
        self.mode = mode
        self._mode = {"r": "rb", "a": "r+b", "w": "wb"}[mode]

        if not fileobj:
            if self.mode == "a" and not os.path.exists(name):
                # Create nonexistent files in append mode.
                self.mode = "w"
                self._mode = "wb"
            fileobj = bltn_open(name, self._mode)
            self._extfileobj = False
        else:
            if name is None and hasattr(fileobj, "name"):
                name = fileobj.name
            if hasattr(fileobj, "mode"):
                self._mode = fileobj.mode
            self._extfileobj = True
        self.name = os.path.abspath(name) if name else None
        self.fileobj = fileobj

        # Init attributes.
        if format is not None:
            self.format = format
        if tarinfo is not None:
            self.tarinfo = tarinfo
        if dereference is not None:
            self.dereference = dereference
        if ignore_zeros is not None:
            self.ignore_zeros = ignore_zeros
        if encoding is not None:
            self.encoding = encoding

        if errors is not None:
            self.errors = errors
        elif mode == "r":
            self.errors = "utf-8"
        else:
            self.errors = "strict"

        if pax_headers is not None and self.format == PAX_FORMAT:
            self.pax_headers = pax_headers
        else:
            self.pax_headers = {}

        if debug is not None:
            self.debug = debug
        if errorlevel is not None:
            self.errorlevel = errorlevel

        # Init datastructures.
        self.closed = False
        self.members = []       # list of members as TarInfo objects
        self._loaded = False    # flag if all members have been read
        self.offset = self.fileobj.tell()
                                # current position in the archive file
        self.inodes = {}        # dictionary caching the inodes of
                                # archive members already added

        try:
            if self.mode == "r":
                self.firstmember = None
                self.firstmember = self.next()

            if self.mode == "a":
                # Move to the end of the archive,
                # before the first empty block.
                while True:
                    self.fileobj.seek(self.offset)
                    try:
                        tarinfo = self.tarinfo.fromtarfile(self)
                        self.members.append(tarinfo)
                    except EOFHeaderError:
                        self.fileobj.seek(self.offset)
                        break
                    except HeaderError, e:
                        raise ReadError(str(e))

            if self.mode in "aw":
                self._loaded = True

                if self.pax_headers:
                    buf = self.tarinfo.create_pax_global_header(self.pax_headers.copy())
                    self.fileobj.write(buf)
                    self.offset += len(buf)
        except:
            if not self._extfileobj:
                self.fileobj.close()
            self.closed = True
            raise

    def _getposix(self):
        return self.format == USTAR_FORMAT
    def _setposix(self, value):
        import warnings
        warnings.warn("use the format attribute instead", DeprecationWarning,
                      2)
        if value:
            self.format = USTAR_FORMAT
        else:
            self.format = GNU_FORMAT
    posix = property(_getposix, _setposix)

    #--------------------------------------------------------------------------
    # Below are the classmethods which act as alternate constructors to the
    # TarFile class. The open() method is the only one that is needed for
    # public use; it is the "super"-constructor and is able to select an
    # adequate "sub"-constructor for a particular compression using the mapping
    # from OPEN_METH.
    #
    # This concept allows one to subclass TarFile without losing the comfort of
    # the super-constructor. A sub-constructor is registered and made available
    # by adding it to the mapping in OPEN_METH.

    @classmethod
    def open(cls, name=None, mode="r", fileobj=None, bufsize=RECORDSIZE, **kwargs):
        """Open a tar archive for reading, writing or appending. Return
           an appropriate TarFile class.

           mode:
           'r' or 'r:*' open for reading with transparent compression
           'r:'         open for reading exclusively uncompressed
           'r:gz'       open for reading with gzip compression
           'r:bz2'      open for reading with bzip2 compression
           'a' or 'a:'  open for appending, creating the file if necessary
           'w' or 'w:'  open for writing without compression
           'w:gz'       open for writing with gzip compression
           'w:bz2'      open for writing with bzip2 compression

           'r|*'        open a stream of tar blocks with transparent compression
           'r|'         open an uncompressed stream of tar blocks for reading
           'r|gz'       open a gzip compressed stream of tar blocks
           'r|bz2'      open a bzip2 compressed stream of tar blocks
           'w|'         open an uncompressed stream for writing
           'w|gz'       open a gzip compressed stream for writing
           'w|bz2'      open a bzip2 compressed stream for writing
        """

        if not name and not fileobj:
            raise ValueError("nothing to open")

        if mode in ("r", "r:*"):
            # Find out which *open() is appropriate for opening the file.
            for comptype in cls.OPEN_METH:
                func = getattr(cls, cls.OPEN_METH[comptype])
                if fileobj is not None:
                    saved_pos = fileobj.tell()
                try:
                    return func(name, "r", fileobj, **kwargs)
                except (ReadError, CompressionError), e:
                    if fileobj is not None:
                        fileobj.seek(saved_pos)
                    continue
            raise ReadError("file could not be opened successfully")

        elif ":" in mode:
            filemode, comptype = mode.split(":", 1)
            filemode = filemode or "r"
            comptype = comptype or "tar"

            # Select the *open() function according to
            # given compression.
            if comptype in cls.OPEN_METH:
                func = getattr(cls, cls.OPEN_METH[comptype])
            else:
                raise CompressionError("unknown compression type %r" % comptype)
            return func(name, filemode, fileobj, **kwargs)

        elif "|" in mode:
            filemode, comptype = mode.split("|", 1)
            filemode = filemode or "r"
            comptype = comptype or "tar"

            if filemode not in "rw":
                raise ValueError("mode must be 'r' or 'w'")

            t = cls(name, filemode,
                    _Stream(name, filemode, comptype, fileobj, bufsize),
                    **kwargs)
            t._extfileobj = False
            return t

        elif mode in "aw":
            return cls.taropen(name, mode, fileobj, **kwargs)

        raise ValueError("undiscernible mode")

    @classmethod
    def taropen(cls, name, mode="r", fileobj=None, **kwargs):
        """Open uncompressed tar archive name for reading or writing.
        """
        if len(mode) > 1 or mode not in "raw":
            raise ValueError("mode must be 'r', 'a' or 'w'")
        return cls(name, mode, fileobj, **kwargs)

    @classmethod
    def gzopen(cls, name, mode="r", fileobj=None, compresslevel=9, **kwargs):
        """Open gzip compressed tar archive name for reading or writing.
           Appending is not allowed.
        """
        if len(mode) > 1 or mode not in "rw":
            raise ValueError("mode must be 'r' or 'w'")

        try:
            import gzip
            gzip.GzipFile
        except (ImportError, AttributeError):
            raise CompressionError("gzip module is not available")

        if fileobj is None:
            fileobj = bltn_open(name, mode + "b")

        try:
            t = cls.taropen(name, mode,
                gzip.GzipFile(name, mode, compresslevel, fileobj),
                **kwargs)
        except IOError:
            raise ReadError("not a gzip file")
        t._extfileobj = False
        return t

    @classmethod
    def bz2open(cls, name, mode="r", fileobj=None, compresslevel=9, **kwargs):
        """Open bzip2 compressed tar archive name for reading or writing.
           Appending is not allowed.
        """
        if len(mode) > 1 or mode not in "rw":
            raise ValueError("mode must be 'r' or 'w'.")

        try:
            import bz2
        except ImportError:
            raise CompressionError("bz2 module is not available")

        if fileobj is not None:
            fileobj = _BZ2Proxy(fileobj, mode)
        else:
            fileobj = bz2.BZ2File(name, mode, compresslevel=compresslevel)

        try:
            t = cls.taropen(name, mode, fileobj, **kwargs)
        except (IOError, EOFError):
            raise ReadError("not a bzip2 file")
        t._extfileobj = False
        return t

    # All *open() methods are registered here.
    OPEN_METH = {
        "tar": "taropen",   # uncompressed tar
        "gz":  "gzopen",    # gzip compressed tar
        "bz2": "bz2open"    # bzip2 compressed tar
    }

    #--------------------------------------------------------------------------
    # The public methods which TarFile provides:

    def close(self):
        """Close the TarFile. In write-mode, two finishing zero blocks are
           appended to the archive.
        """
        if self.closed:
            return

        if self.mode in "aw":
            self.fileobj.write(NUL * (BLOCKSIZE * 2))
            self.offset += (BLOCKSIZE * 2)
            # fill up the end with zero-blocks
            # (like option -b20 for tar does)
            blocks, remainder = divmod(self.offset, RECORDSIZE)
            if remainder > 0:
                self.fileobj.write(NUL * (RECORDSIZE - remainder))

        if not self._extfileobj:
            self.fileobj.close()
        self.closed = True

    def getmember(self, name):
        """Return a TarInfo object for member `name'. If `name' can not be
           found in the archive, KeyError is raised. If a member occurs more
           than once in the archive, its last occurrence is assumed to be the
           most up-to-date version.
        """
        tarinfo = self._getmember(name)
        if tarinfo is None:
            raise KeyError("filename %r not found" % name)
        return tarinfo

    def getmembers(self):
        """Return the members of the archive as a list of TarInfo objects. The
           list has the same order as the members in the archive.
        """
        self._check()
        if not self._loaded:    # if we want to obtain a list of
            self._load()        # all members, we first have to
                                # scan the whole archive.
        return self.members

    def getnames(self):
        """Return the members of the archive as a list of their names. It has
           the same order as the list returned by getmembers().
        """
        return [tarinfo.name for tarinfo in self.getmembers()]

    def gettarinfo(self, name=None, arcname=None, fileobj=None):
        """Create a TarInfo object for either the file `name' or the file
           object `fileobj' (using os.fstat on its file descriptor). You can
           modify some of the TarInfo's attributes before you add it using
           addfile(). If given, `arcname' specifies an alternative name for the
           file in the archive.
        """
        self._check("aw")

        # When fileobj is given, replace name by
        # fileobj's real name.
        if fileobj is not None:
            name = fileobj.name

        # Building the name of the member in the archive.
        # Backward slashes are converted to forward slashes,
        # Absolute paths are turned to relative paths.
        if arcname is None:
            arcname = name
        drv, arcname = os.path.splitdrive(arcname)
        arcname = arcname.replace(os.sep, "/")
        arcname = arcname.lstrip("/")

        # Now, fill the TarInfo object with
        # information specific for the file.
        tarinfo = self.tarinfo()
        tarinfo.tarfile = self

        # Use os.stat or os.lstat, depending on platform
        # and if symlinks shall be resolved.
        if fileobj is None:
            if hasattr(os, "lstat") and not self.dereference:
                statres = os.lstat(name)
            else:
                statres = os.stat(name)
        else:
            statres = os.fstat(fileobj.fileno())
        linkname = ""

        stmd = statres.st_mode
        if stat.S_ISREG(stmd):
            inode = (statres.st_ino, statres.st_dev)
            if not self.dereference and statres.st_nlink > 1 and \
                    inode in self.inodes and arcname != self.inodes[inode]:
                # Is it a hardlink to an already
                # archived file?
                type = LNKTYPE
                linkname = self.inodes[inode]
            else:
                # The inode is added only if its valid.
                # For win32 it is always 0.
                type = REGTYPE
                if inode[0]:
                    self.inodes[inode] = arcname
        elif stat.S_ISDIR(stmd):
            type = DIRTYPE
        elif stat.S_ISFIFO(stmd):
            type = FIFOTYPE
        elif stat.S_ISLNK(stmd):
            type = SYMTYPE
            linkname = os.readlink(name)
        elif stat.S_ISCHR(stmd):
            type = CHRTYPE
        elif stat.S_ISBLK(stmd):
            type = BLKTYPE
        else:
            return None

        # Fill the TarInfo object with all
        # information we can get.
        tarinfo.name = arcname
        tarinfo.mode = stmd
        tarinfo.uid = statres.st_uid
        tarinfo.gid = statres.st_gid
        if type == REGTYPE:
            tarinfo.size = statres.st_size
        else:
            tarinfo.size = 0L
        tarinfo.mtime = statres.st_mtime
        tarinfo.type = type
        tarinfo.linkname = linkname
        if pwd:
            try:
                tarinfo.uname = pwd.getpwuid(tarinfo.uid)[0]
            except KeyError:
                pass
        if grp:
            try:
                tarinfo.gname = grp.getgrgid(tarinfo.gid)[0]
            except KeyError:
                pass

        if type in (CHRTYPE, BLKTYPE):
            if hasattr(os, "major") and hasattr(os, "minor"):
                tarinfo.devmajor = os.major(statres.st_rdev)
                tarinfo.devminor = os.minor(statres.st_rdev)
        return tarinfo

    def list(self, verbose=True):
        """Print a table of contents to sys.stdout. If `verbose' is False, only
           the names of the members are printed. If it is True, an `ls -l'-like
           output is produced.
        """
        self._check()

        for tarinfo in self:
            if verbose:
                print filemode(tarinfo.mode),
                print "%s/%s" % (tarinfo.uname or tarinfo.uid,
                                 tarinfo.gname or tarinfo.gid),
                if tarinfo.ischr() or tarinfo.isblk():
                    print "%10s" % ("%d,%d" \
                                    % (tarinfo.devmajor, tarinfo.devminor)),
                else:
                    print "%10d" % tarinfo.size,
                print "%d-%02d-%02d %02d:%02d:%02d" \
                      % time.localtime(tarinfo.mtime)[:6],

            print tarinfo.name + ("/" if tarinfo.isdir() else ""),

            if verbose:
                if tarinfo.issym():
                    print "->", tarinfo.linkname,
                if tarinfo.islnk():
                    print "link to", tarinfo.linkname,
            print

    def add(self, name, arcname=None, recursive=True, exclude=None, filter=None):
        """Add the file `name' to the archive. `name' may be any type of file
           (directory, fifo, symbolic link, etc.). If given, `arcname'
           specifies an alternative name for the file in the archive.
           Directories are added recursively by default. This can be avoided by
           setting `recursive' to False. `exclude' is a function that should
           return True for each filename to be excluded. `filter' is a function
           that expects a TarInfo object argument and returns the changed
           TarInfo object, if it returns None the TarInfo object will be
           excluded from the archive.
        """
        self._check("aw")

        if arcname is None:
            arcname = name

        # Exclude pathnames.
        if exclude is not None:
            import warnings
            warnings.warn("use the filter argument instead",
                    DeprecationWarning, 2)
            if exclude(name):
                self._dbg(2, "tarfile: Excluded %r" % name)
                return

        # Skip if somebody tries to archive the archive...
        if self.name is not None and os.path.abspath(name) == self.name:
            self._dbg(2, "tarfile: Skipped %r" % name)
            return

        self._dbg(1, name)

        # Create a TarInfo object from the file.
        tarinfo = self.gettarinfo(name, arcname)

        if tarinfo is None:
            self._dbg(1, "tarfile: Unsupported type %r" % name)
            return

        # Change or exclude the TarInfo object.
        if filter is not None:
            tarinfo = filter(tarinfo)
            if tarinfo is None:
                self._dbg(2, "tarfile: Excluded %r" % name)
                return

        # Append the tar header and data to the archive.
        if tarinfo.isreg():
            f = bltn_open(name, "rb")
            self.addfile(tarinfo, f)
            f.close()

        elif tarinfo.isdir():
            self.addfile(tarinfo)
            if recursive:
                for f in os.listdir(name):
                    self.add(os.path.join(name, f), os.path.join(arcname, f),
                            recursive, exclude, filter)

        else:
            self.addfile(tarinfo)