Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
executable file 120 lines (101 sloc) 3.73 KB
#! /usr/bin/python
# Keep Photo Dump is meant to get your photos out of Google Keep pages
# as exported from Google Takeout.
# The photos are base64 encoded.
# This script goes through a given html file and finds all the base64
# encoded image tags and writes them to disk as JPG. It should work just
# as well for extracting photos out of any web page with base64 encoded images.
# May use lots of RAM!
# Usage:
# ./ my_keep_file.html
import argparse
import base64
import os
from HTMLParser import HTMLParser
# Writes a base64 encoded string to a file, as binary.
# Args:
# base64_img: An image encoded as base 64.
# filename: The name of the file to write to.
def decode_and_write_file(base64_img, filename):
decoded_img = base64.b64decode(base64_img)
with open(filename, 'wb') as file:
# The string that corresponds to a base64 encoded jpg.
base64_jpg_prefix = "data:image/jpeg;base64"
# Is the given stirng a base64 encoded image from the <img src=""> tag
# and attribute?
def is_base64_encoded_src(src):
# I'm assuming there can be other image types that I might want to
# decode here. But for now, we only recognize this one string.
return src.startswith(base64_jpg_prefix)
# For an encoded image src attribute, returns the encoding type and
# the data as a part.
def encoding_data_pair(src):
partition = src.partition(',')
return (partition[0], partition[2])
# Returns the value of the src attribute or None.
# Args:
# attrs: A list of (attribute name, value) pairs.
# Returns:
# The value corresponding to 'src' or None if src is not in the list.
def find_source_attribute(attrs):
for (attr_name, attr_value) in attrs:
if attr_name == "src":
return attr_value
return None
# This class is an HTML parser that looks for <img> tags containing
# base64 images (i.e., those whose source starts with
# "data:image/jpeg;base64,"). When the feed() method completes on an
# HTML string, the encoding_pairs() method will return those images.
class Base64ImgGrabber(HTMLParser):
def __init__(self):
self.__encoding_pairs = []
def handle_starttag(self, tag, attrs):
if tag == "img":
src_value = find_source_attribute(attrs)
if src_value and is_base64_encoded_src(src_value):
encoding_pair = encoding_data_pair(src_value)
# Returns the images and their encodings seen during parsing.
# Returns:
# A list of pairs. The key in each pair is the encoding type. The value
# is the data.
def encoding_pairs(self):
return self.__encoding_pairs
# From a path, returns the file name without extension or path.
# "/a/b/c.txt" --> "c"
def file_base(path):
base = os.path.basename(path)
return os.path.splitext(base)[0]
# Parses the html file, dumps all its images to disk.
# Args:
# html_file: The html file to parse.
# file_suffix: When writing image files to disk, use this suffix.
def parse_html_dump_images(html_file, file_suffix):
html_parser = Base64ImgGrabber()
file_index = 1
for (encoding, data) in html_parser.encoding_pairs():
filename = file_base( + str(file_index) + file_suffix
decode_and_write_file(data, filename)
file_index = file_index + 1
# The "main" function.
# Set up flags.
parser = argparse.ArgumentParser(description="""
In a Google Keep HTML file, finds all the images encoded as base64 and writes
them to disk as JPG images.
parser.add_argument('keep_html_file', metavar='keep_html_file',
help='the input html file path')
args = parser.parse_args()
# Parse html and dump.
parse_html_dump_images(args.keep_html_file, ".jpg")