def send_file_helper(context, chat_id, filename, file_content, caption):
"""
Helper for handle generic files.
If the file is a pdf, it will be converted and send as image/s.
If the file is already an image it will not be processed.
If the file is any other type, it will be send as document.
"""
mime = filetype.guess_mime(file_content)
if mime == 'application/pdf':
pages = convert_from_bytes(file_content, 500)
images = []
for page in pages:
file = io.BytesIO()
page.save(file, 'png')
file.name = filename
file.seek(0)
images.append(file)
send_images_helper(context, chat_id, images, caption, '')
elif filetype.image(file_content) is not None:
send_images_helper(context, chat_id, [io.BytesIO(file_content)],
caption, '')
else:
content = io.BytesIO(file_content)
context.bot.send_document(chat_id=chat_id,
document=content,
filename=filename,
parse_mode=ParseMode.MARKDOWN,
caption=caption)
def is_valid_image_file(parser, filename):
"""Simple check the filename is an image file"""
if not os.path.isfile(filename):
parser.error(f'The file {filename} does not exist!')
if not filetype.image(filename):
parser.error(f'The file {filename} not an image file!')
return filename
def assertFileType(content_type=None, name=None):
if content_type == "application/zip" or content_type == "application/x-zip-compressed":
return "zip", ".zip"
elif content_type and content_type.find("image") == 0:
return "img", get_type_obj(content_type, getExtName(name)).extension
else:
try:
if filetype.guess_mime(name) == "application/zip":
return "zip", ".zip"
elif filetype.image(name):
return "img", get_type_obj(content_type,
getExtName(name)).extension
except FileNotFoundError:
pass
raise RequestHandleFailException(415, "输入的文件不是支持的图片或zip类型！")
def unzipAllChooseImages(zipPath):
exts = os.path.splitext(zipPath)
foldername = exts[0]
if len(exts) < 2 or exts[1] != ".zip":
os.rename(zipPath, foldername + ".zip")
zipPath = foldername + ".zip"
with zipfile.ZipFile(zipPath, 'r') as zipf:
zipf.extractall(path=foldername)
res = []
for root, dirs, files in os.walk(foldername):
for file in files:
realPath = os.path.join(root, file)
kind = filetype.image(realPath)
if kind:
res.append(realPath)
if len(res) <= 0:
os.remove(zipPath)
os.remove(foldername)
raise RequestHandleFailException(415, "上传的压缩包中不含有任何的图片类型文件，因此无法处理")
return res, foldername
def get_file_type(self):
"""Determine type of file.
Check extension before bytes, to correctly identify documents
and PDF files. All other types are detected by bytes.
"""
extension = splitext(self.component_file.name)[1][1:]
for type_code, type_data in self.COMPONENT_TYPE_DATA.items():
if extension in type_data.get('extensions', set()):
return type_code
file_obj = self.component_file.open()
if filetype.image(file_obj):
return self.TYPE_IMAGE
elif filetype.video(file_obj):
return self.TYPE_VIDEO
elif filetype.audio(file_obj):
return self.TYPE_AUDIO
elif filetype.archive(file_obj):
return self.TYPE_ARCHIVE
else:
return self.TYPE_OTHER
import filetype
filename = "/path/to/file.jpg"
if filetype.image(filename):
print(f"{filename} is a valid image...")
elif filetype.video(filename):
print(f"{filename} is a valid video...")
else:
print("not image file")
def wx_file_type(self, path):
if filetype.image(str(path)) is not None:
return 'img'
if filetype.video(str(path)) is not None:
return 'vid'
return 'fil'
def get_image_info(fp):
"""Reads some image info from a file descriptor."""
head = fp.read(32)
fp.seek(0)
if len(head) < 24:
return 'unknown', None, None
magic_bytes = b'<?xml', b'<svg'
if any(map(head.strip().startswith, magic_bytes)):
return get_svg_info(fp)
_type = filetype.image(bytearray(head))
fmt = _type.mime.split("/")[1] if _type else None
width = None
height = None
if fmt == 'png':
check = struct.unpack('>i', head[4:8])[0]
if check == 0x0d0a1a0a:
width, height = struct.unpack('>ii', head[16:24])
elif fmt == 'gif':
width, height = struct.unpack('<HH', head[6:10])
elif fmt == 'jpeg':
# specification available under
# http://www.w3.org/Graphics/JPEG/itu-t81.pdf
# Annex B (page 31/35)
# we are looking for a SOF marker ("start of frame").
# skip over the "start of image" marker
# (filetype detection took care of that).
fp.seek(2)
while True:
byte = fp.read(1)
# "All markers are assigned two-byte codes: an X’FF’ byte
# followed by a byte which is not equal to 0 or X’FF’."
if not byte or ord(byte) != 0xff:
raise Exception("Malformed JPEG image.")
# "Any marker may optionally be preceded by any number
# of fill bytes, which are bytes assigned code X’FF’."
while ord(byte) == 0xff:
byte = fp.read(1)
if ord(byte) not in _JPEG_SOF_MARKERS:
# header length parameter takes 2 bytes for all markers
length = struct.unpack('>H', fp.read(2))[0]
fp.seek(length - 2, 1)
continue
# else...
# see Figure B.3 – Frame header syntax (page 35/39) and
# Table B.2 – Frame header parameter sizes and values
# (page 36/40)
fp.seek(3, 1) # skip header length and precision parameters
height, width = struct.unpack('>HH', fp.read(4))
if height == 0:
# "Value 0 indicates that the number of lines shall be
# defined by the DNL marker [...]"
#
# DNL is not supported by most applications,
# so we won't support it either.
raise Exception("JPEG with DNL not supported.")
break
# if the file is rotated, we want, for all intents and purposes,
# to return the dimensions swapped. (all client apps will display
# the image rotated, and any template computations are likely to want
# to make decisions based on the "visual", not the "real" dimensions.
# thumbnail code also depends on this behaviour.)
fp.seek(0)
if is_rotated(fp):
width, height = height, width
else:
fmt = None
return fmt, width, height
