Script to change the color of the highlight annotations in PDF document(s).

pdf-recolor.py

#!/usr/bin/env python
"""
Change the color of the highlight annotations in PDF document(s).
The destination color is hardcoded in this script.
"""
# Adapted from:
# https://unix.stackexchange.com/a/118492/112647
# PDF format reference:
# http://wwwimages.adobe.com/www.adobe.com/content/dam/Adobe/en/devnet/pdf/pdfs/PDF32000_2008.pdf
# pages 12 to 13 define the character sets
import re
import os

# Color specification
PREFIX = b'/C ['
SUFFIX = b']'
COLOR = b'1.0 0.969 0.416'  # your favorite color here!

# Regular expressions
_to_char_group = lambda chars: '[' + re.escape(chars) + ']'
FLOAT = r'(?=\+|\-)?\d+(\.\d*)?'
WHITESPACE = '\x00\x09\x0A\x0C\x0D\x20'
DELIMITER = '()<>[]{}/%'
WS = _to_char_group(WHITESPACE)
DELIM = _to_char_group(DELIMITER)
SPECIAL = _to_char_group(WHITESPACE + DELIMITER)
REGEX_OBJ = re.compile(
    fr'(?<={SPECIAL})obj{WS}*<<.*?>>{WS}*endobj(?={SPECIAL})'.encode(), re.DOTALL
)
REGEX_ANNOT = re.compile(
    f'/Type{WS}*/Annot(?={SPECIAL})'.encode()
)
REGEX_HIGH = re.compile(
    f'/Subtype{WS}*/Highlight(?={SPECIAL})'.encode()
)
REGEX_COLOR = re.compile(
    fr'/C{WS}*\[{WS}*({FLOAT}){WS}+({FLOAT}){WS}+({FLOAT}){WS}*\]'.encode()
)


def process_color(match):
    diff = len(match.group(0)) - len(COLOR) - len(PREFIX) - len(SUFFIX)
    if diff < 0:
        raise NotImplementedError(
            'Replacement is too long and would require the size of the PDF '
            'file to change. This is not implemented.'
        )
    result = PREFIX + diff * b' ' + COLOR + SUFFIX
    return result


def process_obj(match):
    is_annot = REGEX_ANNOT.search(match.group(0))
    is_highlight = REGEX_HIGH.search(match.group(0))
    if is_annot and is_highlight:
        return REGEX_COLOR.sub(process_color, match.group(0))
    else:
        return match.group(0)


def process_file(filename, overwrite=False):
    with open(filename, 'rb') as f:
        data = f.read()
    data = REGEX_OBJ.sub(process_obj, data)
    if not overwrite:
        filename, _ = os.path.splitext(filename)
        filename += '_recolored.pdf'
    with open(filename, 'wb') as f:
        f.write(data)


if __name__ == '__main__':
    import argparse
    parser = argparse.ArgumentParser(description=__doc__, add_help=False)
    parser.add_argument(
        '-h',
        '--help',
        action='help',
        help='Show this help message and exit.'
    )
    parser.add_argument(
        '-o',
        '--overwrite',
        action='store_true',
        help=(
            'Whether to overwrite the files instead of creating new ones '
            "with the suffix '_recolored.pdf'. Default behavior is the latter."
        )
    )
    parser.add_argument(
        'filename',
        nargs='*',
        type=str,
        help='PDF file(s) to process.'
    )
    args = parser.parse_args()

    for filename in args.filename:
        print(f'Processing file: {filename!r}')
        process_file(filename, overwrite=args.overwrite)