Strip resolution information (pHYs chunk) from PNG file

png_strip_pHYs.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

"""
Strips the pHYs chunk (resolution information) of a .png file.
Handy if pdflatex fails with "dimension too large" due to the
PNG resolution being set to 1.

For usage information, call without any parameters.

Author: Jan Schlüter
"""

import sys
import os
import struct

KEEP_TIMESTAMP = True  # whether to keep the timestamp on in-place operation

def print_usage():
    print 'png_strip_pHYs.py: Removes the pHYs chunk of a .png file if present.'
    print 'Usage: %s INFILE [OUTFILE]' % sys.argv[0]
    print '  INFILE: .png file to read'
    print '  OUTFILE: .png file to write. Defaults to INFILE if omitted.'

def read_png_chunk(f):
    """
    Yields all chunks from a PNG file. Make sure to have read the magic number
    (first 8 bytes of the file) first.
    See http://www.w3.org/TR/PNG/#5PNG-file-signature for documentation.
    @return: A tuple of byte strings of a chunk: (length, type, data, crc).
    """
    while True:
        chunklen = f.read(4)
        if len(chunklen) != 4:
            break
        chunktype = f.read(4)
        chunkdata = f.read(struct.unpack('>L', chunklen)[0])
        chunkcrc = f.read(4)
        yield chunklen, chunktype, chunkdata, chunkcrc

def main():
    if len(sys.argv) not in (2,3):
        print_usage()
        return

    # 'parse' command line
    infile = sys.argv[1]
    if len(sys.argv) > 2:
        outfile = sys.argv[2]
    else:
        outfile = infile
        if KEEP_TIMESTAMP:
            fileinfo = os.stat(infile)

    # read input file
    buff = []
    with open(infile, 'rb') as f:
        # read magic number
        magic = f.read(8)
        if magic != '\x89PNG\r\n\x1a\n':
            print "This is not a valid .png file. Aborting."
            return
        buff.append(magic)
        # read input chunks
        for clen, ctype, cdata, ccrc in read_png_chunk(f):
            if ctype != 'pHYs':
                buff.extend([clen, ctype, cdata, ccrc])
    # write output file
    with open(outfile, 'wb') as f:
        f.writelines(buff)

    # keep timestamp if needed
    if KEEP_TIMESTAMP and infile == outfile:
        os.utime(outfile, (fileinfo.st_atime, fileinfo.st_mtime))

if __name__=="__main__":
    main()