From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mimir.eigenstate.org ([206.124.132.107]) by ewsd; Tue Feb 18 09:43:34 EST 2020
Received: from abbatoir.fios-router.home (pool-162-83-132-245.nycmny.fios.verizon.net [162.83.132.245])
	by mimir.eigenstate.org (OpenSMTPD) with ESMTPSA id 7a75051e (TLSv1.2:ECDHE-RSA-AES256-SHA:256:NO);
	Tue, 18 Feb 2020 06:43:19 -0800 (PST)
Message-ID: <26CBEB2B018684C137675D9A0C6AD08A@eigenstate.org>
To: rgl@antares-labs.eu, 9front@9front.org
Subject: Re: [9front] rotate: perform arbitrary rotations
Date: Tue, 18 Feb 2020 06:43:18 -0800
From: ori@eigenstate.org
In-Reply-To: <A78EA9B15B57033D2D46121A23042EC4@antares-labs.eu>
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: full-stack component descriptor scripting-oriented database 

> hi all,
> 
> would it be OK to commit this?
> 
> 
> -rodri
> 
> +Memimage*
> +rotate(Memimage *m, double θ)
> +{
> +	Memimage *w;
> +	Rectangle r;
> +	Point p, c;
> +
> +	if((w = allocmemimage(m->r, m->chan)) == nil)
> +		sysfatal("allocmemimage: %r");
> +	memfillcolor(w, DTransparent);
> +	c = divpt(addpt(m->r.min, m->r.max), 2);
> +	for(r = m->r; r.min.y < m->r.max.y; r.min.y++){
> +		r.min.x = m->r.min.x;
> +		r.max.y = r.min.y+1;
> +		for(; r.min.x < m->r.max.x; r.min.x++){
> +			r.max.x = r.min.x+1;
> +			p = rotatept(r.min, θ, c);
> +			if(ptinrect(p, m->r))
> +				memimagedraw(w, r, m, p, nil, ZP, S);
> +		}
> +	}
> +	freememimage(m);
> +	return w;
> +}

This is probably not going to give very good results,
since you're not doing any filtering.

There are good algorithms for doing affine transformations.
It's probably worth doing that.