Two small fixes, unrelated except that I noticed them both while using oiiotool.
1. By taking an early out of Oiiotool::read if the image is already read, avoid
accidentally setting output_tilewidth and output_dataformat except for images
as they are first read. Doing this wrong was, by a convo
Hello,
I've been messing around with the python binding and struggling to
understand how to consistently deal with different image formats. The
python examples in the repo and on the webpage appear to always create an
array with the "B" flag filled with empty bit. This appears to only
provide me
Is the idea that the ROI modulates all subsequent commands? Or did you mean
that -roi should set the crop?
On Apr 20, 2012, at 1:40 PM, Stephen Parker wrote:
and perhaps with no arguments, --roi could simply compute the 2d bounds for us
From: Stephen Parker
m
and perhaps with no arguments, --roi could simply compute the 2d bounds for us
From: Stephen Parker
To: OpenImageIO developers
Sent: Friday, April 20, 2012 1:37 PM
Subject: Re: [Oiio-dev] oiiotool crop help
For formats that support it, I think we should ha
For formats that support it, I think we should have a new command for setting
the region of interest (ie. --roi xmin,ymin,xmax,ymax).
From: Larry Gritz
To: OpenImageIO developers
Sent: Friday, April 20, 2012 11:54 AM
Subject: Re: [Oiio-dev] oiiotool crop he
oiiotool dpxtocrop.dpx --crop 1024x32+626+103 --origin +0+0 --fullpixels -o
out.dpx
The --crop reduces the data window as requested, the --origin removes the
offset, and the --fullpixels makes the display window conform to the data
window.
I assumed that --crop should mean "reduce the data w
