Python 3 script to extract images from HTTP Archive (HAR) files (Tested & working on Python 3.11)

extract_har.py

#!/usr/bin/env python3

# Python 3 script to extract images from HTTP Archive (HAR) files (Tested & working on Python 3.11)
#
# Original code: kafran
#   https://gist.github.com/kafran/0257c13b3d0a79620695b73062334930
# Updated code: Lewiscowles1986
#   https://gist.github.com/Lewiscowles1986/645e79295efa84698f4e45cd06d610ea
# This code from: MrCheatEugene
#   https://gist.github.com/MrCheatEugene/46ad8173e83efb70cf6543cb36629403
# Updated & tweaked by: FurloSK [originally developed 2023-11-03]
#   https://gist.github.com/FurloSK/0477e01024f701db42341fc3223a5d8c
#
# Changes / Release notes:
#   2023-11-20
#     Prevents overwriting files by appending number to them + displays warning in such case.
#   2023-11-03
#     First working version.

import json
import base64
import os
import sys

# BEGIN config part

# Allowed mimetypes/extensions to be parsed:
# Determines which mimetypes will be saved when parsing the *.har file
mimetypes = {
    "image/webp": ".webp",
    "image/jpeg": ".jpeg", # *.jpg files have two possible extensions
    "image/jpeg": ".jpg",  #   (but .jpeg is official and thus preferred)
    "image/png": ".png",
    "image/svg+xml": ".svg"
}

# Output file path creation rules:
# Determines whether to create subfolders for domain/path when exporting files.
# When <outputPath> is True, <outputPathDepth> determines how many url path
#   parts (delimited by /) will be used, and from which end of url to start.
outputDomain = True
outputPath = True
outputPathDepth = 0 # 0=all, 2=first two path parts , -2=last two path parts

# END config part

#=============================================================================
# Start program
#=============================================================================

# check cmd arguments
if len(sys.argv) < 2 or len(sys.argv) > 3:
    print('extract_har: A simple script to extract all picture files from *.har file.')
    print('Usage: extract_har.py <input.har> [<output_dir>]')
    sys.exit(0)

# get *.har file to work with
if not os.path.exists(sys.argv[1]):
    print('Specified *.har file (' + sys.argv[1] + ') does not exist, exiting...')
    sys.exit(1)
with open(sys.argv[1], "r" ,encoding="utf8") as f:
    print('Loading *.har file: ' + sys.argv[1])
    har = json.loads(f.read())

# get base directory for extraction
if len(sys.argv) == 3:
    folder = sys.argv[2]
else:
    folder = os.path.join(os.getcwd(), sys.argv[1] + '_extract')
baseFolder = os.path.basename(os.path.normpath(folder))
print('Destination folder: ' + baseFolder)
if not os.path.isdir(folder):
    print('  Creating folder structure for extraction...')
    os.makedirs(folder)

# save extensions list from allowed mimetypes
extensions = tuple(mimetypes.values())

# print some info
print('\nExtraction output settings:')
if not outputDomain and not outputPath:
    print('  do not create any directory structure - extract images directly to base folder')
else:
    print('  create subfolders for domain: ' + str(outputDomain))
    if outputPath and outputPathDepth != 0:
        print('  create subfolders for URL path: ' + str(outputPath)
            + ' (only for ' + ('first ' if outputPathDepth > 0 else 'last ')
            + str(abs(outputPathDepth)) + ' parts)')
    else:
        print('  create subfolders for URL path: ' + str(outputPath))

# start parsing the individual entries
print('\nStarting extraction...')
entries = har["log"]["entries"]
count_total = 0
count_extracted = 0
for entry in entries:
    count_total += 1
    
    # only parse the entry if it is one of the wanted extensions
    mimetype = entry["response"]["content"]["mimeType"]
    if mimetype in mimetypes:
        count_extracted += 1
        
        # parse entry url
        url = entry["request"]["url"]
        urlProtocol, tmp = url.split("//", 1) # protocol
        tmp = tmp.split("/")
        urlDomain = tmp[0]                    # domain
        urlFullPath = tmp[1:]
        
        # extract file name and path
        urlPath = urlFullPath[:-1]            # path
        urlFilename = urlFullPath[-1]         # filename
        
        # if filename does not contain expected extension, append it
        if not urlFilename.endswith(extensions):
            urlFilename = urlFilename + mimetypes.get(mimetype)
        
        #print(str(urlProtocol), str(urlDomain), str(urlPath), str(urlFilename))
        
        # determine subfolders structure according to config settings
        pathStr = ''
        if outputDomain and outputPath:
            pathStr = os.path.join(urlDomain,
                *((urlPath[:outputPathDepth]) if outputPathDepth > 0 else (urlPath[outputPathDepth:]))
            )
        elif outputDomain:
            pathStr = urlDomain
        elif outputPath:
            pathStr = os.path.join(
                *((urlPath[:outputPathDepth]) if outputPathDepth > 0 else (urlPath[outputPathDepth:]))
            )
        
        # construct final file path to save extracted file
        if outputDomain or outputPath:
            subFolder = os.path.join(folder, pathStr)
        else:
            subFolder = folder
        outFile = os.path.join(subFolder, urlFilename)
        
        # prepare subfolder
        if not os.path.isdir(subFolder):
            os.makedirs(subFolder)
        
        # parse raw data to image data
        image64 = entry["response"]["content"]["text"]
        image = base64.b64decode(image64)
        
        # check if file exists
        print('  ' + urlFilename + ' [' + str(len(image)) + ' bytes]:')
        print('    extracting to: ' + (pathStr if len(pathStr) else baseFolder))
        if os.path.exists(outFile):
            print('    ⚠️ WARNING: file ' + urlFilename + ' already exists!')
            fixFilename, fixExtension = os.path.splitext(urlFilename)
            
            counter = 1
            while os.path.exists(outFile):
                fixFile = fixFilename + " (" + str(counter) + ")" + fixExtension
                outFile = os.path.join(subFolder, fixFile)
                counter += 1
            print('    ℹ️ Creating file ' + fixFile + '...')
        
        # save data to extracted file
        with open(outFile, "wb") as f:
            f.write(image)

print('\nFinished extracting ' + str(count_extracted) + ' (out of total ' + str(count_total) + ') files.')

hey, thanks for an update!

Lol, I love you and this. It's basically a different program to the original fork now.

I just updated my fork to contain more mime-types to image formats, using MDN and searhing for image/

https://gist.github.com/Lewiscowles1986/645e79295efa84698f4e45cd06d610ea#file-extract_har-py-L9-L21