[rust-dev] A question about implementation of &str

[rust-dev] A question about implementation of &str

[C K Kashyap]

[-- Attachment #1: Type: text/plain, Size: 670 bytes --]

Hi,
I am stuck in my kernel development where I find that I am not able to
iterate over a &str. The code is here -
https://github.com/ckkashyap/unix/blob/master/kernel/uart.rs in the
function uart_putc I find that the for-loop loops the right number of times
but it does not print the right character. To me it appears to be a linking
problem with my kernel. However, to debug this issue I wanted to get a
better understanding of what happens when we iterate over &str. I was
surprised to see that the length of the string literal that is determined
at compile time is being sent as an argument.

I'd appreciate any insights into how I can debug this.

Regards,
Kashyap

[-- Attachment #2: Type: text/html, Size: 1032 bytes --]

Re: [rust-dev] A question about implementation of &str

[Matthieu Monrocq]

[-- Attachment #1: Type: text/plain, Size: 1337 bytes --]

&str is simply a pair (length, pointer).

The reason that even for a literal the length is packed as an argument is
that &str does not ONLY work for literals (complete type &'static str) but
for any slice of characters, such as those produced by String::as_slice()
in which case the lifetime is different (only live as long as the
particular String instance) and the length is not necessarily known at
compile-time.

On Wed, Dec 3, 2014 at 6:34 PM, C K Kashyap <ckkashyap@gmail.com> wrote:

> Hi,
> I am stuck in my kernel development where I find that I am not able to
> iterate over a &str. The code is here -
> https://github.com/ckkashyap/unix/blob/master/kernel/uart.rs in the
> function uart_putc I find that the for-loop loops the right number of
> times but it does not print the right character. To me it appears to be a
> linking problem with my kernel. However, to debug this issue I wanted to
> get a better understanding of what happens when we iterate over &str. I was
> surprised to see that the length of the string literal that is determined
> at compile time is being sent as an argument.
>
> I'd appreciate any insights into how I can debug this.
>
> Regards,
> Kashyap
>
> _______________________________________________
> Rust-dev mailing list
> Rust-dev@mozilla.org
> https://mail.mozilla.org/listinfo/rust-dev
>
>

[-- Attachment #2: Type: text/html, Size: 2141 bytes --]

Re: [rust-dev] A question about implementation of &str

[C K Kashyap]

[-- Attachment #1: Type: text/plain, Size: 1503 bytes --]

Thanks Matthieu.
Regards,
Kashyap

On Wed, Dec 3, 2014 at 11:27 PM, Matthieu Monrocq <
matthieu.monrocq@gmail.com> wrote:

> &str is simply a pair (length, pointer).
>
> The reason that even for a literal the length is packed as an argument is
> that &str does not ONLY work for literals (complete type &'static str) but
> for any slice of characters, such as those produced by String::as_slice()
> in which case the lifetime is different (only live as long as the
> particular String instance) and the length is not necessarily known at
> compile-time.
>
> On Wed, Dec 3, 2014 at 6:34 PM, C K Kashyap <ckkashyap@gmail.com> wrote:
>
>> Hi,
>> I am stuck in my kernel development where I find that I am not able to
>> iterate over a &str. The code is here -
>> https://github.com/ckkashyap/unix/blob/master/kernel/uart.rs in the
>> function uart_putc I find that the for-loop loops the right number of
>> times but it does not print the right character. To me it appears to be a
>> linking problem with my kernel. However, to debug this issue I wanted to
>> get a better understanding of what happens when we iterate over &str. I was
>> surprised to see that the length of the string literal that is determined
>> at compile time is being sent as an argument.
>>
>> I'd appreciate any insights into how I can debug this.
>>
>> Regards,
>> Kashyap
>>
>> _______________________________________________
>> Rust-dev mailing list
>> Rust-dev@mozilla.org
>> https://mail.mozilla.org/listinfo/rust-dev
>>
>>
>

[-- Attachment #2: Type: text/html, Size: 2666 bytes --]