From mboxrd@z Thu Jan  1 00:00:00 1970
References: <4c6dec5271e325c32ee144badfeae4c8@quintile.net>
	<25ea711db6f83d26ca16faf38466bdcc@ladd.quanstro.net>
In-Reply-To: <25ea711db6f83d26ca16faf38466bdcc@ladd.quanstro.net>
Mime-Version: 1.0 (Apple Message framework v1085)
Content-Type: text/plain; charset=us-ascii
Message-Id: <9BDABDE6-950E-4532-B895-ED2115DD2F4E@ar.aichi-u.ac.jp>
Content-Transfer-Encoding: 7bit
From: =?iso-2022-jp?B?GyRCTS1fNxsoQiAbJEI3cjwjGyhC?= <arisawa@ar.aichi-u.ac.jp>
Date: Thu, 10 Oct 2013 12:46:03 +0900
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Subject: Re: [9fans] Brdline and the last line
Topicbox-Message-UUID: 84eb5950-ead8-11e9-9d60-3106f5b1d025

Hello,

If speed is matter, try rdline() in Kirara.
http://plan9.aichi-u.ac.jp/netlib/kirara/kirara-1.2a.tgz

More improved version will be included in kirara-2.0 (to be released).

On 2013/10/10, at 1:04, erik quanstrom wrote:

> On Wed Oct  9 11:54:41 EDT 2013, steve@quintile.net wrote:
>> Hi all
>>
>> It has long been an irritation that Brdline returns failure (to match the
>> end of line token) at end of file if the file does not end with a newline.
>>
>> This is correct but annoying.
>>
>> does anyone had a neat snippet of code which ensures we parse the
>> last line correctly?
>>
>> For files on the filesystem I just fix them, but I am wrapping Brdline
>> around a tcp connection so I am not in control of how people terminate
>> their files...
>
> how about
>
> int
> brdline(Biobuf *b, char *buf, int nbuf)
> {
> 	int i, rv;
> 	Rune r;
>
> 	for(i = 0; i < nbuf - UTFmax - 1; ){
> 		rv = Bgetrune(b);
> 		if(rv == -1 || rv == '\n')
> 			break;
> 		r = rv;
> 		i += chartorune(&r, buf+i);
> 	}
> 	buf[i] = 0;
> 	return i;
> }
>
> - erik
>