All scans now accept negative offsets for the --offset parameter

src/binwalk/__init__.py

@@ -454,6 +454,8 @@ class Binwalk(object):
 		if fd is None:
 			fd = BlockFile(target_file, length=self.scan_length, offset=offset)
 			i_opened_fd = True
+			# If offset is negative (bytes from EOF), BlockFile class will autmoatically calculate the right offset
+			offset = fd.offset
 
 		# Seek to the starting offset.
 		#fd.seek(offset)

src/binwalk/common.py

@@ -114,17 +114,22 @@ class BlockFile(file):
 		@fname  - Path to the file to be opened.
 		@mode   - Mode to open the file in.
 		@length - Maximum number of bytes to read from the file via self.block_read().
+		@offset - Offset at which to start reading from the file.
 
 		Returns None.
 		'''
 		self.total_read = 0
-		self.offset = offset
 		
 		try:
 			self.size = file_size(fname)
 		except:
 			self.size = 0
 
+		if offset < 0:
+			self.offset = self.size + offset
+		else:
+			self.offset = offset
+
 		if length:
 			self.length = length
 		else:

src/binwalk/compression.py

@@ -153,7 +153,7 @@ class CompressionEntropyAnalyzer(object):
 		if self.fp.READ_BLOCK_SIZE < self.BLOCK_SIZE:
 			self.fp.READ_BLOCK_SIZE = self.BLOCK_SIZE
 
-		self.start = start
+		self.start = self.fp.offset
 		self.length = length
 		self.binwalk = binwalk
 

src/binwalk/entropy.py

@@ -135,6 +135,7 @@ class FileEntropy(object):
 			self.block = self.DEFAULT_BLOCK_SIZE
 			
 		self.fd = common.BlockFile(file_name, 'rb', offset=self.start, length=self.length)
+		self.start = self.fd.offset
 		self.fd.MAX_TRAILING_SIZE = 0
 		if self.fd.READ_BLOCK_SIZE < self.block:
 			self.fd.READ_BLOCK_SIZE = self.block

src/binwalk/hexdiff.py

@@ -117,6 +117,9 @@ class HexDiff(object):
 			fp.MAX_TRAILING_SIZE = 0
 			fps.append(fp)
 
+		# BlockFile handles calculation of negative offsets, if one was specified
+		offset = fps[0].offset
+
 		while total < size:
 			i = 0
 			for fp in fps:

src/binwalk/smartstrings.py

@@ -79,6 +79,7 @@ class FileStrings(object):
 		# TODO: This is not optimal. We should read in larger chunks and process it into self.block chunks.
 		self.fd.READ_BLOCK_SIZE = self.block
 		self.fd.MAX_TRAILING_SIZE = 0
+		self.start = self.fd.offset
 
 		# Set the total_scanned and scan_length values for plugins and status display messages
 		self.binwalk.total_scanned = 0