From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on inbox.vuxu.org
X-Spam-Level: 
X-Spam-Status: No, score=0.2 required=5.0 tests=DKIM_INVALID,DKIM_SIGNED
	autolearn=no autolearn_force=no version=3.4.4
Received: (qmail 22774 invoked from network); 25 Nov 2021 11:01:37 -0000
Received: from 4ess.inri.net (216.126.196.42)
  by inbox.vuxu.org with ESMTPUTF8; 25 Nov 2021 11:01:37 -0000
Received: from mail.9lab.org ([168.119.8.41]) by 4ess; Thu Nov 25 05:54:22 -0500 2021
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=9lab.org; s=20210803;
	t=1637836858;
	h=from:from:reply-to:subject:subject:date:date:message-id:message-id:
	 to:to:cc:mime-version:mime-version:content-type:content-type:
	 content-transfer-encoding:content-transfer-encoding:  in-reply-to:in-reply-to;
	bh=PK3EHzs+t2T8HyQTVIDaSgKHVy/IKdrS9dzQAnB4qOw=;
	b=aFMmlt/J5oIJBsyrRxXvwTXZbXC1KOFSB4/EEfc73j4v+ZgC/inhaqb/7v070YaNHjAatJ
	NBj/4Ok2WZIdrGgh7KXePpxRcgbBQLlSl8P7R3Chhz/0dJqdkQ+dWdwSuXNEGeTVYb1xUr
	fcJINEQkExMYkrR1PHsrtToSFK4cP9Y=
Received: from pjw (host-185-64-155-70.ecsnet.at [185.64.155.70])
	by mail.9lab.org (OpenSMTPD) with ESMTPSA id c1732f9c (TLSv1.2:ECDHE-RSA-CHACHA20-POLY1305:256:NO)
	for <9front@9front.org>;
	Thu, 25 Nov 2021 11:40:57 +0100 (CET)
Message-ID: <7CEF83159FC92104223B8FEC4E0A594B@9lab.org>
To: 9front@9front.org
Date: Thu, 25 Nov 2021 11:40:56 +0100
From: igor@9lab.org
In-Reply-To: <B884909AAC83761D18BC037BEB08577E@felloff.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit
List-ID: <9front.9front.org>
List-Help: <http://lists.9front.org>
X-Glyph: ➈
X-Bullshit: private hypervisor rich-client replication DOM manager 
Subject: Re: [9front] [PATCH] rio: allow spaces in working directory path (-cd) when creating a new window
Reply-To: 9front@9front.org
Precedence: bulk

Thanks for looking over it!

Quoth cinap_lenrek@felloff.net:
[…]
> probably is better to parse everything using gettokens().
> as your patch doesnt handle quoting the quotes.
[…]

Thanks for the tip. That indeed sounds like the cleaner solution.

> the whole window command also will probably have issues with
> this kind of stuff... the quoting is definitely messed up in
> there.

Yeah, that is where my journey started, thinking that window cmd
was the culprit, until I realised that the limitation goes all
the way to `/sys/src/cmd/rio/wctl.c:/^parsewctl`.

> 
> tho it is possible to quote from rc, like:
> 
> fn q {a=$1 whatis a | sed 's!^a=!!;q'}
> 
> or even get rid of wctl all together... it is a bit silly given
> you can just fork and mount $wsys yourself... then there are not
> two different code paths for creating a window.

I think I understand what you mean. The below is just checking I am
on the same page.

You are referring to the two highlighted ('^^^') code paths in
/rc/bin/window:

…
  if(~ $#mflag 1) {
  ^^^
  	…
  		if(mount $wsys /mnt/wsys 'new '$"spec){
  			bind -b /mnt/wsys /dev
  			exec $argv0 -x $cmd </dev/cons >/dev/cons >[2]/dev/cons
  		}
  	}&
  	exit ''
  }
  if not {
  ^^^  	if(~ $wctl ''){
  		if(test -f /dev/wctl) wctl=/dev/wctl
  		if not if(test -f /mnt/term/dev/wctl) wctl=/mnt/term/dev/wctl
  		if not if(test -r /mnt/term/env/wctl) wctl=/mnt/term^`{cat /mnt/term/env/wctl}
  		if not {
  			echo $argv0: '$wctl' not defined >[1=2]
  			exit bad
  		}
  	}
  
  	if(! ~ $#wdir 0)
  		spec=($spec -cd $wdir)
  	echo new $spec $argv0 -x $cmd >>$wctl
  }
…

…if I understand you correctly the suggestion is to drop the `if not  { … }`
part that writes to $wctl and modify the `if(~ $#mflag 1) { … }` in
such a way that it can handle both the `-m` and non `-m` variant.

Cheers,
Igor