[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index]
[Thread Index]
- Subject: Re: io:lines() and \0
- From: Francisco Olarte <folarte@...>
- Date: Thu, 20 Feb 2014 14:00:46 +0100
Hi René
On Thu, Feb 20, 2014 at 1:33 PM, René Rebe <rene@exactcode.de> wrote:
> Yes, that patch was not perfect, see my new post from today.
I did see it, have not tried to analyse it, but it seem to me that it
syntetishes a '\n' for the last line if it is not there, which I
strongly dislike, and makes a lot of memsets. I find this as bad as
the truncation which the current version does ( just my opinion ).
I still like my solution more, but do not think it's adequate as it
will regress ( I would consider it a bug fix with a speed regression,
but from what I've read, if Roberto et al consider it a non-bug fix
with a speed regression, so that's what it is ). Given I'm not going
to use Lua for heavy duty text processing ( as I routinely use perl
and C/C++, it's difficult to beat these for that ), I'm just going to
flag it as a Lua quirk and follow on.
If I need more speed and need to work in lua, I will investigate a
simple replacement of the IO library written against POSIX or
something like that, it should be easy to do and with some adequate
loading I can load it at start and replace the standard.
Regards.
- References:
- io:lines() and \0, René Rebe
- Re: io:lines() and \0, steve donovan
- Re: io:lines() and \0, René Rebe
- Re: io:lines() and \0, Enrico Colombini
- Re: io:lines() and \0, steve donovan
- Re: io:lines() and \0, René Rebe
- Re: io:lines() and \0, Luiz Henrique de Figueiredo
- Re: io:lines() and \0, René Rebe
- Re: io:lines() and \0, Francisco Olarte
- Re: io:lines() and \0, René Rebe