splitting Vec<u8> of non unicode characters

[bobi6666]

hello, I have a u8 vector, there can be non-unicode characters in that vector and I needed to divide it wherever there is a \0, but with the fact that if there is a \n or \r\n, it will not stop but will continue. Can you please show me example?

[BurntSushi]

I think you probably want to use bytes::Regex for this.

As for a specific example, I don't really understand what you're saying. Could you please provide some input and the desired output?

[bobi6666]

i would just want example on how to build Regex::new from regex::bytes crate to cover my question 2022-11-17 13:32 GMT+01:00, Andrew Gallant ***@***.***>:

…

I think you probably want to use [`bytes::Regex`](https://docs.rs/regex/latest/regex/bytes/struct.Regex.html) for this. As for a specific example, I don't really understand what you're saying. Could you please provide some input and the desired output? -- Reply to this email directly or view it on GitHub: #928 (comment) You are receiving this because you authored the thread. Message ID: ***@***.***>

[BurntSushi]

It's right there in the link I gave you. :-) For example: https://docs.rs/regex/latest/regex/bytes/struct.Regex.html#method.find

[bobi6666]

yesterday i wanted to build with unicode disabled and when i did let re = regex::bytes::Regex::new(r"0x0").unwrap(); it didn't work 2022-11-17 14:18 GMT+01:00, Andrew Gallant ***@***.***>:

…

It's right there in the link I gave you. :-) For example: https://docs.rs/regex/latest/regex/bytes/struct.Regex.html#method.find -- Reply to this email directly or view it on GitHub: #928 (comment) You are receiving this because you authored the thread. Message ID: ***@***.***>

[BurntSushi]

If you can't answer my original question

Could you please provide some input and the desired output?

then I can't help you. I won't be responding further if your next comment doesn't answer that question. I apologize if I come across as rude, but this has been a common pattern with you in past discussions. They just go back-and-forth endlessly without my understanding what you actually want to do.

All I'm asking for is a simple example that shows the input and what you want your output to be. If that isn't something you know how to provide, then I'm sorry, but I just don't have the time to give the help required here.

[BurntSushi]

This might help: https://jvns.ca/blog/good-questions/

(I linked to ESR's version of the same thing previously, but I forgot just how patronizing it was.)

[bobi6666]

https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=65826762b40d386ae495a1781305930f 2022-11-17 14:45 GMT+01:00, Andrew Gallant ***@***.***>:

…

This might help: https://jvns.ca/blog/good-questions/ (I linked to [ESR's version of the same thing](http://www.catb.org/%7Eesr/faqs/smart-questions.html) previously, but I forgot just how patronizing it was.) -- Reply to this email directly or view it on GitHub: #928 (reply in thread) You are receiving this because you authored the thread. Message ID: ***@***.***>

[BurntSushi]

Thanks, but you still didn't answer my question! You didn't tell me what output you wanted. And the program you've given me has zero output.

Anyway, I took a guess: https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=d315997c5951b2aca3fd1e974fccbc2a

[BurntSushi]

Notice that you don't even need a bytes::Regex because a NUL byte is valid UTF-8.

[bobi6666]

output will be later Vec<&[u8]> but about that i am fine and if your regex idea will not stop after non utf16 characters then you did what i wanted 2022-11-17 15:10 GMT+01:00, Andrew Gallant ***@***.***>:

…

Notice that you don't even need a `bytes::Regex` because a NUL byte is valid UTF-8. -- Reply to this email directly or view it on GitHub: #928 (reply in thread) You are receiving this because you authored the thread. Message ID: ***@***.***>

[bobi6666]

I'll try to translate it so you can better understand what I need: in the example I sent you, there may be bytes that are not utf8 but may be utf16. I need to split them, but in such a way that they can be collected at the end into Vec<&[u8]> or Vec<Vec<u8>>, but I believe that if they are split so their appearance will be preserved as they are and I will not be blamed if they do not carry utf 8 2022-11-17 15:14 GMT+01:00, Peter Kubek ***@***.***>:

…

output will be later Vec<&[u8]> but about that i am fine and if your regex idea will not stop after non utf16 characters then you did what i wanted 2022-11-17 15:10 GMT+01:00, Andrew Gallant ***@***.***>: > Notice that you don't even need a `bytes::Regex` because a NUL byte is > valid > UTF-8. > > -- > Reply to this email directly or view it on GitHub: > #928 (reply in thread) > You are receiving this because you authored the thread. > > Message ID: > ***@***.***>

[BurntSushi]

A &str can be converted to &[u8] via str::as_bytes.

[bobi6666]

do you think this split will be faster than normal split from std if my file that i work with will be big? 2022-11-17 15:39 GMT+01:00, Andrew Gallant ***@***.***>:

…

A `&str` can be converted to `&[u8]` via [`str::as_bytes`](https://doc.rust-lang.org/std/primitive.str.html#method.as_bytes). -- Reply to this email directly or view it on GitHub: #928 (reply in thread) You are receiving this because you authored the thread. Message ID: ***@***.***>

[BurntSushi]

Maybe. Benchmark it.

[bobi6666]

hello, I want to ask you if aho_corasick supports processing non utf8 bytes if i am using replace_all_with method? 2022-11-17 16:08 GMT+01:00, Andrew Gallant ***@***.***>:

…

Maybe. Benchmark it. -- Reply to this email directly or view it on GitHub: #928 (reply in thread) You are receiving this because you authored the thread. Message ID: ***@***.***>

[BurntSushi]

Yes... Of course...