Python: Breaking "For" Loop Doesn't Give Desired Results

[dodle]

I've created a python script that takes corrupted jpeg images and removes garbage from the beginning of the file. I use a for loop to iterate through the file and find the JFIF header. Sometimes the images contain more than one JFIF header so I have to break out of the loop after the first header is found. But for some reason the break causes the program to report that the image is not corrupted. I haven't been able to figure out what my problem is yet.

Here is the portion of the code giving me the problem:

...
...
# JFIF header
jfif = (('ff', 'd8', 'ff', 'e0', '00', '10'), ('4a', '46', '49', '46'))
...
...
jpeg = open(image, 'rb')

jpeg_data = jpeg.read()
jpeg_data = array.array('B', jpeg_data)

jpeg.close()

def ToHex(b):
return '%02x' % b

def HexToInt(h):
return int(h, 16)

try:
# If JFIF header is not found, image is reported as not fixable
corrupt = 2
for byte in xrange(len(jpeg_data)):
if ToHex(jpeg_data[byte]) == jfif[1][0]:
if ToHex(jpeg_data[byte+1]) == jfif[1][1]:
if ToHex(jpeg_data[byte+2]) == jfif[1][2]:
if ToHex(jpeg_data[byte+3]) == jfif[1][3]:
# JFIF header found
corrupt = 0
garbage = list(jpeg_data[:byte])
for g in xrange(len(garbage)):
garbage[g] = ToHex(garbage[g])
if garbage != list(jfif[0]):
# If JFIF header is found with garbage before, it is reported as fixable
corrupt = 1
new_jpeg_data = array.array('B', jpeg_data[byte:])
prefix_bytes = len(jfif[0])
for p in xrange(prefix_bytes):
prefix_bytes -= 1
new_jpeg_data.insert(0, HexToInt(jfif[0][prefix_bytes]))
# Ensure that iteration ends after first JFIF header
break
if corrupt:
if corrupt == 2:
print('\nERROR: Could not repair %s!\n' % image)
logger.error('Could not repair %s!' % image)
else:
name = '.'.join(os.path.basename(image).split('.')[:-1])
name = '%s_fix.jpg' % name
print('Fixing %s and saving to %s' % (image, name))
logger.info('Fixing %s and saving to %s' % (image, name))
new_jpeg = open(name, 'wb')
new_jpeg.write(new_jpeg_data)
new_jpeg.close()
else:
print('\nWARNING: %s does not appear to be corrupt, skipping!\n' % image)
logger.warning('%s does not appear to be corrupt, skipping!' % image)
#time.sleep(4)
except IndexError:
print('\ERROR: Could not repair %s!\n' % image)
logger.error('Could not repair %s!' % image)

I created a test script to see if I could find the error, but the test results the way I want it.

#! /usr/bin/env python

jfif = (('ff', 'd8', 'ff', 'e0', '00', '10'), ('4a', '46', '49', '46'))

# Pseudo jpeg data with garbage before JFIF header
jpeg_data = ('aa', 'a0', '4a', '46', '49', '46', 'b9', '7d', '4a', '46', '49', '46')

def HexToInt(h):
return int(h, 16)

# If JFIF header is not found, image is reported as not fixable
corrupt = 2
for byte in xrange(len(jpeg_data)):
if jpeg_data[byte] == jfif[1][0]:
print('4a')
if jpeg_data[byte+1] == jfif[1][1]:
print('46')
if jpeg_data[byte+2] == jfif[1][2]:
print('49')
if jpeg_data[byte+3] == jfif[1][3]:
print('46')
corrupt = 0
garbage = list(jpeg_data[:byte])
for g in xrange(len(garbage)):
garbage[g] = garbage[g]
if garbage != list(jfif[0]):
# If JFIF header is found with garbage before, it is reported as fixable
corrupt = 1
new_jpeg_data = list(jpeg_data[byte:])
prefix_bytes = len(jfif[0])
for p in xrange(prefix_bytes):
prefix_bytes -= 1
new_jpeg_data.insert(0, jfif[0][prefix_bytes])
# Ensure that iteration ends after first JFIF header
break

if corrupt:
if corrupt == 2:
print('\nERROR: Could not repair image!\n')
else:
print('Fixing image and saving')
print(new_jpeg_data)
else:
print('\nWARNING: image does not appear to be corrupt, skipping!\n')


----- EDIT -----

I think the problem is in this area of code:

if garbage != list(jfif[0]):
# If JFIF header is found with garbage before, it is reported as fixable
corrupt = 1


----- SOLVED -----

Okay, my mistake. It seems to be working fine. The jpeg image I was testing with wasn't what I thought.