# How do I hash a file of arbitrary size?
**URL:** https://forum.aztec.network/t/how-do-i-hash-a-file-of-arbitrary-size/6989
**Category:** Noir
**Created:** [November 4, 2024, 11:52am UTC](https://forum.aztec.network/t/how-do-i-hash-a-file-of-arbitrary-size/6989 "2024-11-04T11:52:13Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@qbzzt](https://forum.aztec.network/u/qbzzt)
#### Post date: [November 4, 2024, 11:52am UTC](https://forum.aztec.network/t/how-do-i-hash-a-file-of-arbitrary-size/6989/1 "2024-11-04T11:52:13Z")
I need to prove that a person knew a file with a publicly known blake3 hash. The way I’d like to do it is to provide the file as a private parameter, the blake3 hash as a public parameter, and a signed message as another public parameter (so we know the person signed the message and provided it to the zero-knowledge system along with the file).
How do I deal with hashing a variable-length file? I tried to feel the blake3 function a slice, but it didn’t work.
```auto
fn main() {
let mut slice: [u8] = &[0; 10];
let hash = std::hash::blake3(slice);
}
```
The error is:
```auto
ori@CryptoDocLaptop:~/noir/hello_world$ nargo execute
warning: unused variable hash
┌─ src/main.nr:7:9
│
7 │ let hash = std::hash::blake3(slice);
│ ---- unused variable
│
error: Expected type [u8; _], found type [u8]
┌─ src/main.nr:7:34
│
7 │ let hash = std::hash::blake3(slice);
│ -----
│
Aborting due to 1 previous error
ori@CryptoDocLaptop:~/noir/hello_world$
```
All the tricks I tried to get it to work also failed.
Is there a way to do it, or does the application need to compile a new circuit for every file size?
---
### Author:  [@muizcal](https://forum.aztec.network/u/muizcal)
#### Post date: [November 7, 2024, 6:31pm UTC](https://forum.aztec.network/t/how-do-i-hash-a-file-of-arbitrary-size/6989/2 "2024-11-07T18:31:18Z")
To fix the error, the code should use a `Vec` for variable-length input data and pass that data into the hashing function. This way, the application can handle files of any size without compiling new circuits for each different file.
---
### Author:  [@joshc](https://forum.aztec.network/u/joshc)
#### Post date: [November 7, 2024, 6:34pm UTC](https://forum.aztec.network/t/how-do-i-hash-a-file-of-arbitrary-size/6989/3 "2024-11-07T18:34:19Z")
You actually can’t provide a variable length input to a circuit. It’s a limitation of working with circuits. You would need to generate a new circuit for different sizes. A workaround that people use to reduce the number of circuits needed is to have several circuits of various lengths, and with a given input, use the circuit with the smallest input possible. Then pad the input so that the input size will match what is expected.