Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-25 Thread Quentin Rameau
Hi Roberto,

> If we go to increase that size, I would go to use dynamic memory. Having
> an array of 1MB statically allocated is a crazy idea

Yes it is!

> (and it is not C99 compliant, where the maximun allocated size is 128K).

That's actually 64k (“at least […] — 65535 bytes in an object (in a
hosted environment only)”) ;)



Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-25 Thread Silvan Jegen
On Tue, Sep 25, 2018 at 10:05 AM Roberto E. Vargas Caballero
 wrote:
> On Mon, Sep 24, 2018 at 05:45:29PM -0700, Eric Pruitt wrote:
> > I agree that the current buffer is too small. I'm pretty sure I've run
> > into this problem myself with Vim and Bash, but I hadn't gotten around
> > to digging into the problem.
>
> If we go to increase that size, I would go to use dynamic memory. Having
> an array of 1MB statically allocated is a crazy idea (and it is not
> C99 compliant, where the maximun allocated size is 128K).

On my machine, st uses 11MB up to 15MB of memory currently. If we just add 1MB
of statically-allocated space it would blow up that size by 9% in some cases so
I am also in favor of using dynamic memory there.



Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-25 Thread Roberto E. Vargas Caballero
On Mon, Sep 24, 2018 at 05:45:29PM -0700, Eric Pruitt wrote:
> I agree that the current buffer is too small. I'm pretty sure I've run
> into this problem myself with Vim and Bash, but I hadn't gotten around
> to digging into the problem.

If we go to increase that size, I would go to use dynamic memory. Having
an array of 1MB statically allocated is a crazy idea (and it is not
C99 compliant, where the maximun allocated size is 128K).



Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-24 Thread Eric Pruitt
On Mon, Sep 24, 2018 at 10:16:31PM +0200, Ingo Heimbach wrote:
> I have recognized that a buffer size of 128*4 bytes is quite small to
> store base64 encoded clipboard content (see OSC52), especially when
> copying multiple code lines in a terminal editor like vim.

I agree that the current buffer is too small. I'm pretty sure I've run
into this problem myself with Vim and Bash, but I hadn't gotten around
to digging into the problem.

Eric



Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-24 Thread Ingo Heimbach
I have recognized that a buffer size of 128*4 bytes is quite small to store 
base64 encoded clipboard content (see OSC52), especially when copying multiple 
code lines in a terminal editor like vim. 1048576 is 1 mega of 4 byte UTF-8 
characters. It is quite big but I simply chose a value that should even be big 
enough for very large text copy operations. Other values are also OK. The point 
is that 128*4 bytes are too small for the escape sequence buffer (in my 
opinion).

Best regards,
Ingo 

Am 24. September 2018 17:05:41 MESZ schrieb Laslo Hunhold :
>On Mon, 24 Sep 2018 15:52:41 +0100
>"Roberto E. Vargas Caballero"  wrote:
>
>Dear Roberto,
>
>> I would say, why 1048576 and not 1000? or 1?.
>> Is there a specific reason?
>
>1048576 is 2^20, so it's not completely arbitrary, though the point
>really stands as to why there's such a huge jump in size. Can you give
>a little information about that, Ingo?
>
>With best regards
>
>Laslo
>
>-- 
>Laslo Hunhold 


Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-24 Thread Laslo Hunhold
On Mon, 24 Sep 2018 15:52:41 +0100
"Roberto E. Vargas Caballero"  wrote:

Dear Roberto,

> I would say, why 1048576 and not 1000? or 1?.
> Is there a specific reason?

1048576 is 2^20, so it's not completely arbitrary, though the point
really stands as to why there's such a huge jump in size. Can you give
a little information about that, Ingo?

With best regards

Laslo

-- 
Laslo Hunhold 



Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-24 Thread Roberto E. Vargas Caballero
On Sun, Sep 23, 2018 at 03:56:43PM +0200, Ingo Heimbach wrote:
> What is incorrect? 

I would say, why 1048576 and not 1000? or 1?.
Is there a specific reason?


Roberto



Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-23 Thread Ingo Heimbach
What is incorrect? 

Am 23. September 2018 14:12:23 MESZ schrieb Hiltjo Posthuma 
:
>On Sun, Sep 23, 2018 at 12:12:27PM +0200, Ingo Heimbach wrote:
>> Before this commit, long escape sequences (e.g. OSC 52) could be
>trimmed
>> due to a quite small escape sequence buffer. This commit increases
>the
>> escape sequence buffer from 128 to 1048576 UTF-8 characters.
>> ---
>>  st.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>> 
>> diff --git a/st.c b/st.c
>> index 46c954b..e9d4536 100644
>> --- a/st.c
>> +++ b/st.c
>> @@ -31,7 +31,7 @@
>>  /* Arbitrary sizes */
>>  #define UTF_INVALID   0xFFFD
>>  #define UTF_SIZ   4
>> -#define ESC_BUF_SIZ   (128*UTF_SIZ)
>> +#define ESC_BUF_SIZ   (1048576*UTF_SIZ)
>>  #define ESC_ARG_SIZ   16
>>  #define STR_BUF_SIZ   ESC_BUF_SIZ
>>  #define STR_ARG_SIZ   ESC_ARG_SIZ
>> -- 
>> 2.19.0
>> 
>> 
>
>This is incorrect.
>
>-- 
>Kind regards,
>Hiltjo


Re: [hackers] [st][patch] Increase the buffer size for escape sequences

2018-09-23 Thread Hiltjo Posthuma
On Sun, Sep 23, 2018 at 12:12:27PM +0200, Ingo Heimbach wrote:
> Before this commit, long escape sequences (e.g. OSC 52) could be trimmed
> due to a quite small escape sequence buffer. This commit increases the
> escape sequence buffer from 128 to 1048576 UTF-8 characters.
> ---
>  st.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/st.c b/st.c
> index 46c954b..e9d4536 100644
> --- a/st.c
> +++ b/st.c
> @@ -31,7 +31,7 @@
>  /* Arbitrary sizes */
>  #define UTF_INVALID   0xFFFD
>  #define UTF_SIZ   4
> -#define ESC_BUF_SIZ   (128*UTF_SIZ)
> +#define ESC_BUF_SIZ   (1048576*UTF_SIZ)
>  #define ESC_ARG_SIZ   16
>  #define STR_BUF_SIZ   ESC_BUF_SIZ
>  #define STR_ARG_SIZ   ESC_ARG_SIZ
> -- 
> 2.19.0
> 
> 

This is incorrect.

-- 
Kind regards,
Hiltjo