[Python-checkins] python/dist/src/Lib tarfile.py,1.10,1.11

[akuchling at users.sourceforge.net]

Update of /cvsroot/python/python/dist/src/Lib
In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv7570

Modified Files:
	tarfile.py 
Log Message:
[Patch 988444]
Read multiple special headers
- fixed/improved handling of extended/special headers
in read-mode (adding new extended headers should be
less painful now).
- improved nts() function.
- removed TarFile.chunks datastructure which is not
(and was never) needed.
- fixed TarInfo.tobuf(), fields could overflow with too
large values, values are now clipped.


Index: tarfile.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/tarfile.py,v
retrieving revision 1.10
retrieving revision 1.11
diff -C2 -d -r1.10 -r1.11
*** tarfile.py	2 Jan 2004 15:44:29 -0000	1.10
--- tarfile.py	10 Jul 2004 22:02:11 -0000	1.11
***************
*** 136,140 ****
      """Convert a null-terminated string buffer to a python string.
      """
!     return s.split(NUL, 1)[0]
  
  def calc_chksum(buf):
--- 136,140 ----
      """Convert a null-terminated string buffer to a python string.
      """
!     return s.rstrip(NUL)
  
  def calc_chksum(buf):
***************
*** 714,718 ****
              ):
              l = len(value)
!             parts.append(value + (fieldsize - l) * NUL)
  
          buf = "".join(parts)
--- 714,718 ----
              ):
              l = len(value)
!             parts.append(value[:fieldsize] + (fieldsize - l) * NUL)
  
          buf = "".join(parts)
***************
*** 797,801 ****
          self.members     = []       # list of members as TarInfo objects
          self.membernames = []       # names of members
-         self.chunks      = [0]      # chunk cache
          self._loaded     = False    # flag if all members have been read
          self.offset      = 0L       # current position in the archive file
--- 797,800 ----
***************
*** 1282,1288 ****
              self.offset += blocks * BLOCKSIZE
  
!         self.members.append(tarinfo)
!         self.membernames.append(tarinfo.name)
!         self.chunks.append(self.offset)
  
      def extract(self, member, path=""):
--- 1281,1285 ----
              self.offset += blocks * BLOCKSIZE
  
!         self._record_member(tarinfo)
  
      def extract(self, member, path=""):
***************
*** 1552,1556 ****
  
          # Read the next block.
!         self.fileobj.seek(self.chunks[-1])
          while True:
              buf = self.fileobj.read(BLOCKSIZE)
--- 1549,1553 ----
  
          # Read the next block.
!         self.fileobj.seek(self.offset)
          while True:
              buf = self.fileobj.read(BLOCKSIZE)
***************
*** 1570,1574 ****
                  else:
                      # Block is empty or unreadable.
!                     if self.chunks[-1] == 0:
                          # If the first block is invalid. That does not
                          # look like a tar archive we can handle.
--- 1567,1571 ----
                  else:
                      # Block is empty or unreadable.
!                     if self.offset == 0:
                          # If the first block is invalid. That does not
                          # look like a tar archive we can handle.
***************
*** 1593,1602 ****
          # method is registered in the TYPE_METH. If so, then call it.
          if tarinfo.type in self.TYPE_METH:
!             tarinfo = self.TYPE_METH[tarinfo.type](self, tarinfo)
!         else:
!             tarinfo.offset_data = self.offset
!             if tarinfo.isreg() or tarinfo.type not in SUPPORTED_TYPES:
!                 # Skip the following data blocks.
!                 self.offset += self._block(tarinfo.size)
  
          if tarinfo.isreg() and tarinfo.name[:-1] == "/":
--- 1590,1599 ----
          # method is registered in the TYPE_METH. If so, then call it.
          if tarinfo.type in self.TYPE_METH:
!             return self.TYPE_METH[tarinfo.type](self, tarinfo)
! 
!         tarinfo.offset_data = self.offset
!         if tarinfo.isreg() or tarinfo.type not in SUPPORTED_TYPES:
!             # Skip the following data blocks.
!             self.offset += self._block(tarinfo.size)
  
          if tarinfo.isreg() and tarinfo.name[:-1] == "/":
***************
*** 1604,1610 ****
              tarinfo.type = DIRTYPE
  
!         self.members.append(tarinfo)
!         self.membernames.append(tarinfo.name)
!         self.chunks.append(self.offset)
          return tarinfo
  
--- 1601,1605 ----
              tarinfo.type = DIRTYPE
  
!         self._record_member(tarinfo)
          return tarinfo
  
***************
*** 1621,1625 ****
      # 2. set self.offset to the position where the next member's header will
      #    begin.
!     # 3. return a valid TarInfo object.
  
      def proc_gnulong(self, tarinfo):
--- 1616,1622 ----
      # 2. set self.offset to the position where the next member's header will
      #    begin.
!     # 3. call self._record_member() if the tarinfo object is supposed to
!     #    appear as a member of the TarFile object.
!     # 4. return tarinfo or another valid TarInfo object.
  
      def proc_gnulong(self, tarinfo):
***************
*** 1637,1658 ****
              count -= BLOCKSIZE
  
!         if tarinfo.type == GNUTYPE_LONGNAME:
!             name = nts(buf)
!         if tarinfo.type == GNUTYPE_LONGLINK:
!             linkname = nts(buf)
! 
!         buf = self.fileobj.read(BLOCKSIZE)
  
!         tarinfo = TarInfo.frombuf(buf)
!         tarinfo.offset = self.offset
!         self.offset += BLOCKSIZE
!         tarinfo.offset_data = self.offset
!         tarinfo.name = name or tarinfo.name
!         tarinfo.linkname = linkname or tarinfo.linkname
  
!         if tarinfo.isreg() or tarinfo.type not in SUPPORTED_TYPES:
!             # Skip the following data blocks.
!             self.offset += self._block(tarinfo.size)
!         return tarinfo
  
      def proc_sparse(self, tarinfo):
--- 1634,1647 ----
              count -= BLOCKSIZE
  
!         # Fetch the next header
!         next = self.next()
  
!         next.offset = tarinfo.offset
!         if tarinfo.type == GNUTYPE_LONGNAME:
!             next.name = nts(buf)
!         elif tarinfo.type == GNUTYPE_LONGLINK:
!             next.linkname = nts(buf)
  
!         return next
  
      def proc_sparse(self, tarinfo):
***************
*** 1710,1713 ****
--- 1699,1704 ----
          self.offset += self._block(tarinfo.size)
          tarinfo.size = origsize
+ 
+         self._record_member(tarinfo)
          return tarinfo
  
***************
*** 1746,1749 ****
--- 1737,1746 ----
                  return self.members[i]
  
+     def _record_member(self, tarinfo):
+         """Record a tarinfo object in the internal datastructures.
+         """
+         self.members.append(tarinfo)
+         self.membernames.append(tarinfo.name)
+ 
      def _load(self):
          """Read through the entire archive file and look for readable