From a793ee30b81fc348b653e0e4c9f00d33969003e7 Mon Sep 17 00:00:00 2001
From: Philipp Gesang <philipp.gesang@intra2net.com>
Date: Tue, 29 Aug 2017 14:45:31 +0200
Subject: [PATCH] lift block alignment requirement for tar archive rescue

It is unlikely that damaged archives have correctly aligned tar
headers. Thus we need to check each header-like section whether
it contains the right magic and the checksum matches. Objects
without a correct checksum (which spans the better part of the
header) are discarded similar to what file(1) does.
---
 deltatar/tarfile.py     |   26 ++++++++++++++++++++------
 testing/test_recover.py |    2 +-
 2 files changed, 21 insertions(+), 7 deletions(-)

diff --git a/deltatar/tarfile.py b/deltatar/tarfile.py
index b2de2d5..4987d7d 100644
--- a/deltatar/tarfile.py
+++ b/deltatar/tarfile.py
@@ -3384,6 +3384,7 @@ TAR_FMT_HDR = (# See tar(5):
 #      valid, though the header is not truly POSIX conforming.  */
 #
 #
+TAR_HDR_OFF_MAGIC    = 257
 TAR_FMT_OLDGNU_MAGIC = b"ustar "
 
 def read_gnu_tar_hdr (data):
@@ -3445,6 +3446,14 @@ def read_gnu_tar_hdr (data):
         }
 
 
+def tar_hdr_check_chksum (data):
+    hdr = read_gnu_tar_hdr (data)
+    if hdr is None:
+        return False
+    s = calc_chksums (data)
+    return nti (hdr ["checksum"]) in s
+
+
 def readable_tar_objects_offsets (ifd):
     """
     Traverse blocks in file, trying to extract tar headers.
@@ -3452,14 +3461,19 @@ def readable_tar_objects_offsets (ifd):
     pos     = 0
     offsets = []
 
+    mm = mmap.mmap(ifd, 0, mmap.MAP_SHARED, mmap.PROT_READ)
+    pos = TAR_HDR_OFF_MAGIC
+
     while True:
-        blk = os.read (ifd, BLOCKSIZE)
-        if len (blk) != BLOCKSIZE:
+        pos = mm.find (TAR_FMT_OLDGNU_MAGIC, pos)
+        if pos == -1:
             break
-        hdr = read_gnu_tar_hdr (blk)
-        if hdr is not None:
-            offsets.append (pos)
-        pos += BLOCKSIZE
+        off = pos - TAR_HDR_OFF_MAGIC
+        mm.seek (off)
+        blk = mm.read (BLOCKSIZE)
+        if tar_hdr_check_chksum (blk) is True:
+            offsets.append (off)
+        pos += 1
 
     return offsets
 
diff --git a/testing/test_recover.py b/testing/test_recover.py
index f2147dd..6e8ab99 100644
--- a/testing/test_recover.py
+++ b/testing/test_recover.py
@@ -156,7 +156,7 @@ def corrupt_leading_garbage (_, fname, compress, encrypt):
     assert os.lseek (infd, 0, os.SEEK_SET) == 0
     outfd = os.open (os.path.dirname (aname), os.O_WRONLY | os.O_TMPFILE,
                      stat.S_IRUSR | stat.S_IWUSR)
-    junk  = os.urandom (512) # tar block sized
+    junk  = os.urandom (42)
 
     # write new file with garbage prepended
     done = 0
-- 
1.7.1