I tried this: (python)
file = "some data of file passed by frontend/client"
filepath = InputFile.from_bytes(file)
print(filepath)
response = storage.create_file('attachments', id_class.unique(), filepath)
and it gives the appwrite.exception.AppwriteException: Empty file passed to the endpoint. error. I double checked the file var does contain the contents of the upload file from the frontend
the bucket has all perms for "Any" role
wait, if the file is not binary, does appwrite storage not see it as a file?
InputFile.from_bytes needs your input to be a byte array
In case your input is of a different type, you'll have to convert it first
ohh, any idea on how to do that?
also, does the bucket accept all file types by default?
This depends on the data type you want to convert
For strings, this might help: https://blog.enterprisedna.co/python-string-to-bytes-3-easy-methods/
Yes
ok, thanks, I'll try this and get back to u
so I encode the file data to UTF-8? will appwrite accept it?
file = str(file).encode("utf-8")
print(f"FILE: {bytearray(file)}")
filepath = InputFile.from_bytes(bytearray(file))
print(f"FILEPATH: {filepath}")
response = storage.create_file('attachments', id_class.unique(), filepath)
is this correct?
The initial encoding depends on what datatype file was present in, in the first place
For example, if it was a string
text_data = "Text data";
file = bytearray(text_data, "utf-8");
filepath = InputFile.from_bytes(file);
response = storage.create_file("attachments", id_class.unique(), filepath)
That being said, if instead of text_data, there's a variable representing a different type, you'll have to handle it accordingly
well, the data is going to be in string format. so I will try this method
I printed file and it shows this type of bytearray, still appwrite classifies it at "empty":
(message.txt)
I'll try testing on my end too once, if that's the case
Ok, Thanks
😆 Oh yeah this is very common. Python has to many different types of bin, byte, byte array, hex, all kinds of stuff
I don't remember the exact encoding you need for it to work properly when sent over HTTP
Lemme check past projects to see if we have anything that would help
Oof, I've dealt with byteencoding before and it was always a pain
@Ryan You know what it is?
InputFile.from_bytes(file) is missing filename
🤣
The error message is confusing but it's because filename is optional in our SDK, but it shouldn't be 😬
Recommended threads
- Appwrite Cloud Services. getting 500 Int...
I'm not able to access any appwrite services on appwrite cloud website, getting a 500 internal error on all services even though status.appwrite.online says the...
- [FEAT REQ] Bun runtime support
Instead of just Nodejs, it would be helpful to also add support for the Bun runtime which can speed up deployments by a ton.
- [ENHANCEMENT] Use the custom Next.js dep...
Deployment adapters on Next.js are now stable! This means that we don't have to stick to the limitations of sniffing out build outputs and manually moving or mo...