PlaidCTF 2019: A Whaley Good Joke

whale.py

import tarfile
import json
from datetime import datetime
import sys
import shutil
import tempfile
from pathlib import Path
from io import BytesIO
import hashlib

# https://stackoverflow.com/a/44873382
def sha256sum(filename):
    h  = hashlib.sha256()
    b  = bytearray(128*1024)
    mv = memoryview(b)
    with open(filename, 'rb', buffering=0) as f:
        for n in iter(lambda : f.readinto(mv), 0):
            h.update(mv[:n])
    return h.hexdigest()

image_filename = sys.argv[1]

def layer_timestamps(image_dir):
    for layer_tar in Path(image_dir).glob("*/layer.tar"):
        with tarfile.open(layer_tar) as layer:
            max_timestamp = max(info.mtime for info in layer)
            yield max_timestamp, layer_tar

with tempfile.TemporaryDirectory() as tmpdir:
    with tarfile.open(image_filename, 'r:gz') as image:
        image.extractall(path=tmpdir)
    
    with Path(tmpdir, "manifest.json").open('r') as f:
        manifest = json.load(f)
        manifest_layers = manifest[0]["Layers"] = []
    
    config_filename = manifest[0]["Config"]
    with Path(tmpdir, config_filename).open('r') as f:
        config = json.load(f)
        diff_ids = config["rootfs"]["diff_ids"] = []
    
    print("Sorting layers...")
    layers = sorted(layer_timestamps(tmpdir))

    for timestamp, layer_tar in layers:
        layer_hash = layer_tar.parts[-2]
        layer_tar = Path(tmpdir, layer_hash, "layer.tar")
        layer_diffid = sha256sum(layer_tar)
        manifest_layers.append(f"{layer_hash}/layer.tar")
        diff_ids.append(f"sha256:{layer_diffid}")

    print("Rebuilding image...")
    with tarfile.open(f"{image_filename}-fixed", 'w:gz') as image_fixed:
        image_fixed.add(Path(tmpdir, "repositories"), arcname="repositories")

        # Add layer files
        for image_file in Path(tmpdir).glob("*/*"):
            # TODO: skip json?
            image_fixed.add(image_file, arcname=image_file.relative_to(tmpdir))
            # print(image_file)

        manifest_file = BytesIO(json.dumps(manifest).encode("utf-8"))
        manifest_tarinfo = tarfile.TarInfo("manifest.json")
        manifest_tarinfo.size = manifest_file.getbuffer().nbytes
        image_fixed.addfile(manifest_tarinfo, manifest_file)

        config_file = BytesIO(json.dumps(config).encode("utf-8"))
        config_tarinfo = tarfile.TarInfo(config_filename)
        config_tarinfo.size = config_file.getbuffer().nbytes
        image_fixed.addfile(config_tarinfo, config_file)

# Load with:
#   docker load -i pctf-whales_169aeb74f82dcdceb76e36a6c4c22a89-fixed
# Run:
#   docker run --rm -it pctf-whales:latest ./flag.sh

# This soluition has files 12 and 15 missing some reason.
# I extracted them manually.

# pctf{1_b3t_u_couldnt_c0nt4in3r_ur_l4ught3r}