Nate Vack njvack

Simple canvas-based color converter

To answer this StackOverflow question I wrote this — a small solution based on a never-rendered <canvas> element. It fills a 1-pixel canvas with the provided fill-style, and then reads the RGBA values of that pixel. It will work with any CSS color -- name, rgba(), hex, or even something more exotic like a gradient or pattern. Invalid colors are always returned as transparent black. Transparent colors are treated as painted on a newly-cleared canvas.

It's been tested in modern-ish versions of IE, Chrome, Safari, and Firefox. The API is:

color_convert.to_hex(color)   # Converts color to a hex-based RGB triple; to_hex('red') returns '#ff0000'
color_convert.to_rgba(color)  # Converts color to an rgba() string; to_rgba('red') returns 'rgba(255,0,0,1)'

	#!/usr/bin/env python

	"""
	Compute stimulus orderings for the mnemonic similarity task, described in:
	Stark SM, Stevenson R, Wu C, Rutledge S, & Stark CEL (2015). Stability of
	age-related deficits in the mnemonic similarity task across task variations.
	Behavioral Neuroscience 129(3), 257-268.

	Written by Nate Vack <[email protected]> at the Center for Healthy Minds,
	University of Wisconsin-Madison, for the mp2 study conducted by Dan Grupe

	#!/bin/bash

	# Written by Jonah Chaiken and Nate Vack
	# Support: [email protected]

	help=" Usage: convert_dcm2niix.sh <in_file> <out_file>

	This program will individually convert a dicom archive file to a NIfTI file.

	The input can be either a .tar file containing bzip2'd dicom slices, or a

	#!/usr/bin/env python
	"""
	Usage: dos_line_endings

	Takes stdin, changes all line endings from \r or \n to \r\n, strips UTF-8
	byte-order mark (if any), writes to stdout.
	"""

	import sys
	import re

	def convert_three_byte(frame):
	data_out = np.zeros(128, dtype=np.int32) # You're looking to make a 128-element array of int32.
	data_out.dtype = np.byte
	data_out = data_out.reshape(-1, 4) # Shape is now (128,4) and dtype is byte
	# Now, copy the data from chans08to00 into data_out. We need to specify which bytes to put it in.
	# Also, we need to reverse the data in frame.
	# You'd change 0:3 to 1:4 if the stuff in frame is big-endian, I think? You'll need to experiment.
	data_out[0:8, 0:3] = frame['chans08to00'][::-1]
	# ... and so on with channels
	data_out.dtype = np.int32

	<!DOCTYPE html>
	<html>
	<head>
	<meta charset="UTF-8" />
	<script type="text/javascript">

	window.console = window.console \|\| {}
	window.console.log = window.console.log \|\| function() {};

	console.log("Hello");

	<!DOCTYPE html>
	<meta charset="utf-8">
	<style>

	.node {
	font: 300 11px "Helvetica Neue", Helvetica, Arial, sans-serif;
	fill: #bbb;
	}

	.node:hover {

	#!/usr/bin/env ruby

	require 'mechanize'
	require 'pp'

	def coins_login(agent, username, password)
	page = agent.get "https://chronus.mrn.org/cas/login.php?rp=https%3A%2F%2Fchronus.mrn.org%2Fmicis%2Findex.php%3Fsubsite%3Dasmt"
	uname_match = page.body.match /name="([^"]+)" placeholder="Username/
	uname_field = uname_match[1]
	pw_match = page.body.match /name="([^"]+)" placeholder="Pa\$\$w0rd/

	#!/bin/bash

	GATING_DIR=${GATING_DIR:-/usr/g/service/log/gating}
	DEST_BASE=${DEST_BASE:-/mri-upload/ge_physio}
	MAX_AGE_MINUTES=${3:-120}

	study=$1
	ppt=$2

	if [[ "${study}" == "" ]] \|\| [[ "${ppt}" == "" ]] \|\| [[ "${study}" == "-h" ]]; then