def correct_file_extension(file_bytes: bytes, target_path):
"""
Identify the correct file type and extension based on the byte content of the file and
change the target file extensions to the correct suffix.
There is a limitation to what files can be recognised. See https://pypi.org/project/filetype/#supported-types
Parameters
----------
file_bytes : bytes
bytes read form the source file, this can be the entire file or a minimum of the first 261 bytes.
target_path : Path
Path object to the audio file that need to have the correct extension applied
Returns
-------
Path
Pathlib Path to the audio file with the correct extension applied if file type was of a type that could be
identified and corrected.
"""
kind_of_file = filetype.guess(file_bytes)
if kind_of_file:
correct_extension = f".{kind_of_file.extension}"
target_path = target_path.with_suffix(correct_extension)
return target_path
def test_conversion(self):
"""Test if the library can just convert video files and produce MP3 files."""
potato_mp3_path = convert_video(os.path.join(video_dir, "potato.mp4"),
os.path.join(temp_dir, "potato.mp3"))
# File should exist.
assert (os.path.exists(potato_mp3_path))
# We should be able to open it.
open(potato_mp3_path, 'r').close()
# It should look like an MP3.
assert filetype.guess(potato_mp3_path).extension == "mp3"
# It should REALLY look like an MP3.
assert (filetype.guess(potato_mp3_path).mime == "audio/mpeg")
def _validate_filetype(obj):
"""
Checks for valid file type and returns extension and mime type.
Args:
obj (str, bytes, bytearray or readable): Path to the file, a readable object or the file's content
as str, bytes or bytearray. See filetype documentation:
https://h2non.github.io/filetype.py/v1.0.0/filetype.m.html
Returns:
tuple: The file's extension and mime type.
Raises:
TypeError: Raised if type is invalid or could not be recognized.
"""
try:
ft = filetype.guess(obj)
except IOError:
ft = None
# Check for SVG content if filetype.guess fails
if ft is None:
# Test for readable object first
try:
content = obj.read()
except AttributeError:
# Test for file path
try:
with open(obj) as f:
content = f.read()
except Exception:
# Use input as content as last option
content = obj
# Convert to string
if isinstance(content, bytes) or isinstance(content, bytearray):
content = content.decode('utf-8')
# Check for SVG content
if ImageRecord.SVG_PATTERN.search(content):
result = ('svg', 'image/svg+xml')
else:
raise TypeError('Unrecognized file type')
else:
result = (ft.extension, ft.mime)
# Check for extension in valid file types
if result[0] not in ImageRecord.VALID_FILE_TYPES:
raise TypeError('Invalid file type: {invalid}. Valid types are: {valid}.'.format(
invalid=ft.extension,
valid=ImageRecord.VALID_FILE_TYPES
))
return result
def move_file_to_typed_direct(source_file, target_file, file_type):
if os.path.isfile(target_file) or os.path.isfile(source_file):
return
filepath_set = set(os.listdir(source_file))
dir = os.path.dirname(source_file)
for v in filepath_set:
full_path = os.path.join(dir, v)
if os.path.isfile(full_path):
kind = filetype.guess(full_path)
if kind and kind.EXTENSION == file_type:
shutil.move(full_path, target_file)
def file_extension_from_bytes(file_bytes: bytes) -> Optional[str]:
"""
From file byte content identify the type of file, if recognised return the suffix extension, else returns None
Parameters
==========
file_bytes : bytes
Returns
=======
str or None
Str - suffix including dot is file is recognised, e.g. '.jpg'
None - File type was not recognised
"""
kind_of_file = filetype.guess(file_bytes)
if kind_of_file:
return f".{kind_of_file.extension}"
def download_file(url: str, ideal_filename: str = None, out_dir: str = None, headers: List[Tuple[str, str]] = None, with_progress_bar: bool = True,
cache: bool = True, duplicate_handler: DuplicateHandler = DuplicateHandler.FIND_VALID_FILE, ignored_content_types: List[str] = None,
max_filename_length=DEFAULT_MAX_FILENAME_LENGTH, group_by: GroupByMapping = None) -> DownloadedFile:
download_cache = shelve.open(CACHE_WEBSITE_LINKS_FILE, writeback=True)
if cache and url in download_cache:
return download_cache[url]
configure_urllib_opener(headers)
filename = tempfile.TemporaryFile(delete=False).name
file_download = download_file_impl(url, filename, download_cache, with_progress_bar=with_progress_bar)
if not file_download:
downloaded_file = add_to_download_cache(download_cache, url, result=DownloadedFileResult.FAIL)
return downloaded_file
if type(file_download) == DownloadedFile:
return file_download
if not file_download[0]:
downloaded_file = add_to_download_cache(download_cache, url, result=DownloadedFileResult.FAIL)
return downloaded_file
url, old_url, res_headers = file_download
content_type = get_content_type_from_headers(res_headers)
if ignorable_content_type(ignored_content_types, content_type):
downloaded_file = add_to_download_cache(download_cache, url, result=DownloadedFileResult.SKIPPED)
return downloaded_file
actual_name = None
content_disposition = res_headers.get('Content-Disposition', failobj=None)
if content_disposition:
result = re.findall('filename="(.+)"', content_disposition)
result = first_or_none(result)
if get_file_extension(result):
actual_name = result
potential_filename = os.path.basename(urlparse(url).path)
if not actual_name and potential_filename:
if content_type:
potential_mimetype = content_type
else:
potential_mimetype = mimetypes_extended.guess_type(potential_filename)[0]
if potential_mimetype:
ext = mimetypes_extended.guess_extension(potential_mimetype, include_period=True)
if ext and potential_filename.endswith(ext):
actual_name = potential_filename
if not actual_name or not get_file_extension(actual_name):
ext = None
if content_type:
ext = mimetypes_extended.guess_extension(content_type, include_period=True)
if ext:
if actual_name:
actual_name += ext
else:
actual_name = get_filename(filename) + ext
else:
kind = filetype.guess(filename)
if kind:
actual_name = f'{get_filename(filename)}.{kind.extension}'
if actual_name:
if not out_dir:
out_dir = str(Path(filename).parent)
filename_split = split_filename(actual_name, include_ext_period=True)
if ideal_filename and filename_split[1]:
ideal_filename = os.path.splitext(ideal_filename)[0]
actual_name = ideal_filename + filename_split[1]
else:
actual_name = get_filename_from_url(url)
if not actual_name:
actual_name = os.path.basename(filename)
out_path = join_path(out_dir, filename=actual_name)
if out_dir and group_by:
directory, filename_only, ext = split_path_components(out_path, fatal=False, include_ext_period=True)
sub_dir = f'/{group_by.fail_dir}'
if ext in group_by:
sub_dir = f'/{group_by[ext]}'
filename_with_ext = join_filename_with_ext(filename_only, ext)
out_path = join_path(directory, sub_dir, filename=filename_with_ext)
out_path = shorten_file_name(out_path, max_length=max_filename_length)
filename = move_file(filename, out_path, make_dirs=True, duplicate_handler=duplicate_handler)
downloaded_file = add_to_download_cache(download_cache, url, old_url, headers=res_headers, filename=filename)
download_cache.close() # synchronizes automatically
return downloaded_file
def document_handler(update: Update, context: CallbackContext):
"""
This function has multiple scope of responsibility:
1. Downloading a document without extension;
2. Guessing a file type. If type isn't supported file removes from server;
3. Checking is this file type supported by application. If type isn't supported file removes from server;
4. Removing metadata from document via calling 'metadata_worker.py';
2. Applying face hiding tool;
3. Sending cloaked and metadata-free file;
4. Removing original, metadata-free, and cloaked files from server.
Yep, this function definitely should be refactored...
"""
logger = getLogger()
is_faces_found = False
logger.info("document_handler started")
file = context.bot.getFile(update.message.document.file_id)
file.download("documents/image")
logger.info("Guessing file type")
kind = filetype.guess("documents/image")
if kind is None:
logger.error("Cannot guess file type!")
update.message.reply_text(
"Cannot guess file type. This file type not supported")
try:
logger.info("Preparing for file deletion from server (kind guess)")
remove("documents/image")
update.message.reply_text("File successfully removed from server")
except Exception as e:
logger.error("Can't remove file (kind guess)")
logger.error(e.args)
update.message.reply_text("Error at removing file from server")
return
logger.info("File MIME type: %s", kind.mime)
if kind.mime not in SUPPORTED_MIME_LIST:
update.message.reply_text("{} not supported!".format(kind.mime))
logger.info("Removing file...")
try:
remove("documents/image")
update.message.reply_text("File successfully removed from server")
except Exception as e:
logger.error("Can't remove file")
logger.error(e.args)
update.message.reply_text("Error at removing file from server")
return
else:
logger.info("Metadata removing started")
if kind.mime == "image/png":
try:
delete_metadata_from_png("documents/image")
logger.info("Metadata was successfully deleted")
except Exception as e:
logger.error("Metadata wasn't deleted")
logger.error(e.args)
remove_original_doc_from_server(logger, update)
update.message.reply_text(
"Error at removing metadata from PNG file\nFile removed from server"
)
remove_original_doc_from_server(logger, update)
try:
logger.info("Goes into fawkes section")
update.message.reply_text("Applying face hider tools, wait...")
_ = call(["fawkes", "-d", "documents", "--mode", FAWKES_MODE])
logger.info(result_of("ls documents"))
if path.exists("documents/clean_image_cloaked.png"):
is_faces_found = True
logger.info("Does faces found?: %s", is_faces_found)
logger.info("fawkes try-catch finished")
except Exception as e:
logger.error("EXCEPTION at fawkes section")
logger.error(e.args)
update.message.reply_text("Error at hiding faces")
if is_faces_found:
logger.info("Preparing for sending cloaked file")
send_file(logger, update, context, "cloaked")
logger.info("Preparing for clean file deletion on server")
try:
remove("documents/clean_image.png")
update.message.reply_text(
"Clean version of file successfully removed from server"
)
logger.info("Clean version of file successfully removed")
except Exception as e:
logger.error("Can't remove clean version of file")
logger.error(e.args)
update.message.reply_text(
"Error at removing clean version of file from server")
logger.info("Preparing for cloaked photo deletion on server")
try:
remove("documents/clean_image_cloaked.png")
update.message.reply_text(
"Cloaked file successfully removed from server")
logger.info("Cloaked file successfully removed")
except Exception as e:
logger.error("Can't remove cloaked file")
logger.error(e.args)
update.message.reply_text(
"Error at removing cloaked file from server")
else:
logger.info("No faces found")
update.message.reply_text("Can't find any faces")
logger.info("Preparing for sending photo without metadata")
send_file(logger, update, context, "clean")
try:
remove("documents/clean_image.png")
update.message.reply_text(
"File without metadata successfully removed from server"
)
logger.info("File without metadata successfully removed")
except Exception as e:
logger.error("Can't remove file without metadata")
logger.error(e.args)
update.message.reply_text(
"Error at removing file without metadata from server")
else:
delete_metadata("documents/image")
logger.info("Metadata was successfully deleted")
update.message.reply_text("Metadata was successfully deleted")
remove_original_doc_from_server(logger, update)
try:
logger.info("Goes into fawkes section")
update.message.reply_text("Applying face hider tools, wait...")
_ = call(["fawkes", "-d", "documents", "--mode", FAWKES_MODE])
logger.info(result_of("ls documents"))
if path.exists("documents/clean_image_cloaked.png"):
is_faces_found = True
logger.info("Does faces found?: %s", is_faces_found)
logger.info("fawkes try-catch finished")
except Exception as e:
logger.error("EXCEPTION at fawkes section")
logger.error(e.args)
update.message.reply_text("Error at hiding faces")
if is_faces_found:
logger.info("Preparing for sending cloaked file")
send_file(logger, update, context, "cloaked")
logger.info("Preparing for clean file deletion on server")
try:
remove("documents/clean_image.jpg")
update.message.reply_text(
"Clean version of file successfully removed from server"
)
logger.info("Clean version of file successfully removed")
except Exception as e:
logger.error("Can't remove clean version of file")
logger.error(e.args)
update.message.reply_text(
"Error at removing clean version of file from server")
logger.info("Preparing for cloaked photo deletion on server")
try:
remove("documents/clean_image_cloaked.png")
update.message.reply_text(
"Cloaked file successfully removed from server")
logger.info("Cloaked file successfully removed")
except Exception as e:
logger.error("Can't remove cloaked file")
logger.error(e.args)
update.message.reply_text(
"Error at removing cloaked file from server")
else:
logger.info("No faces found")
update.message.reply_text("Can't find any faces")
logger.info("Preparing for sending photo without metadata")
send_file(logger, update, context, "clean")
try:
remove("documents/clean_image.jpg")
update.message.reply_text(
"File without metadata successfully removed from server"
)
logger.info("File without metadata successfully removed")
except Exception as e:
logger.error("Can't remove file without metadata")
logger.error(e.args)
update.message.reply_text(
"Error at removing file without metadata from server")
def does_file_mime_has(file, mime_keyword):
guess = filetype.guess(file)
return guess and mime_keyword in guess.mime
def get_filetype(cls, url):
with urllib.request.urlopen(url) as f:
return filetype.guess(f.read(261))
