added support for compressed records. kodump now supports a --width option.

2 files changed, 39 insertions, 8 deletions

diff --git a/crodump.py b/crodump.py
index 8d69df9..80daab9 100644
--- a/crodump.py
+++ b/crodump.py
@@ -6,6 +6,7 @@ from binascii import b2a_hex
 from hexdump import hexdump, asasc, tohex, unhex, strescape
 from koddecoder import kodecode
 from readers import ByteReader
+import zlib
 
 """
 python3 crodump.py crodump chechnya_proverki_ul_2012
@@ -126,6 +127,8 @@ class Datafile:
                 decdat = encdat
                 decflag = ' '
 
+            if args.decompress and self.iscompressed(decdat):
+                decdat = self.decompress(decdat)
             print("%5d: %08x-%08x: (%02x:%08x) %s %s%s %s" % (i+1, ofs, ofs+ln, flags, chk, infostr, decflag, toout(args, decdat), tohex(tail)))
 
         if args.verbose:
@@ -134,6 +137,22 @@ class Datafile:
                 dat = self.readdata(o, l)
                 print("%08x-%08x: %s" % (o, o+l, toout(args, dat)))
 
+    def iscompressed(self, data):
+        if len(data)<11:
+            return
+        size, flag = struct.unpack_from(">HH", data, 0)
+        if size+5 != len(data):
+            return
+        if flag!=0x800:
+            return
+        if data[-3:] != b"\x00\x00\x02":
+            return
+        return True
+
+    def decompress(self, data):
+        C = zlib.decompressobj(-15)
+        return C.decompress(data[8:-3])
+
 def destruct_bank_definition(args, data):
     """
     decode the 'bank' / database definition
@@ -284,16 +303,13 @@ def decode_kod(args, data):
     """
     if args.nokod:
         # plain hexdump, no KOD decode
-        if args.ascdump:
-            print(asasc(data))
-        else:
-            hexdump(args.offset, data)
+        hexdump(args.offset, data, args)
 
     elif args.shift:
         # explicitly specified shift.
         args.shift = int(args.shift, 0)
         enc = kodecode(args.shift, data)
-        hexdump(args.offset, enc)
+        hexdump(args.offset, enc, args)
     elif args.increment:
         # explicitly specified shift.
         for s in range(256):
@@ -317,6 +333,11 @@ def kod_hexdump(args):
         args.endofs = int(args.endofs, 0)
         args.length = args.endofs - args.offset
 
+    if args.width:
+        args.width = int(args.width, 0)
+    else:
+        args.width = 64 if args.ascdump else 16
+
     if args.filename:
         with open(args.filename, "rb") as fh:
             if args.length is None:
@@ -365,6 +386,7 @@ def main():
     ko = subparsers.add_parser('kodump', help='KOD/hex dumper')
     ko.add_argument('--offset', '-o', type=str, default="0")
     ko.add_argument('--length', '-l', type=str)
+    ko.add_argument('--width', '-w', type=str)
     ko.add_argument('--endofs', '-e', type=str)
     ko.add_argument('--unhex', '-x', action='store_true', help="assume the input contains hex data")
     ko.add_argument('--shift', '-s', type=str, help="KOD decode with the specified shift")
@@ -380,6 +402,7 @@ def main():
     cro.add_argument('--ascdump', '-a', action='store_true')
     cro.add_argument('--nokod', '-n', action='store_true')
     cro.add_argument('--struonly', action='store_true')
+    cro.add_argument('--nodecompress', action='store_false', dest='decompress', default='true')
     cro.add_argument('dbdir', type=str)
     cro.set_defaults(handler=cro_dump)
 
diff --git a/hexdump.py b/hexdump.py
index cbba3c9..984a19f 100644
--- a/hexdump.py
+++ b/hexdump.py
@@ -26,9 +26,17 @@ def aschr(b):
     return "."
 def asasc(line):
     return "".join(aschr(_) for _ in line)
-def hexdump(ofs, data):
-    for o in range(0, len(data), 16):
-        print("%08x: %-47s  %s" % (o+ofs, ashex(data[o:o+16]), asasc(data[o:o+16])))
+def hexdump(ofs, data, args):
+    w = args.width
+    if args.ascdump:
+        fmt = "%08x: %s"
+    else:
+        fmt = "%%08x: %%-%ds  %%s" % (3*w-1)
+    for o in range(0, len(data), w):
+        if args.ascdump:
+            print(fmt % (o+ofs, asasc(data[o:o+w])))
+        else:
+            print(fmt % (o+ofs, ashex(data[o:o+w]), asasc(data[o:o+w])))
 
 def tohex(data):
     return b2a_hex(data).decode('ascii')