josePhoenix

• -sameq has been removed, -qscale 0 is ambiguous, -qscale:v 0 might do the right thing?
• -ss <start_time_secs>, -t <duration_secs> to extract a segment of the video
• burning in subtitles: ffmpeg -i video.mkv -vf subtitles=video.mkv out.avi (or videosubs.srt after subtitles= if they're a separate file
• switching containers: ffmpeg -i input.mkv -vcodec copy -acodec copy output.mp4

josePhoenix

{
 "metadata": {
  "name": "",
  "signature": "sha256:f5ab02d3ac08e1202c6393db0ad4b6eff20422abc57c1b7e44602f5c8db2621b"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [

josePhoenix

{
 "metadata": {
  "name": "",
  "signature": "sha256:3e543cb29723aa5100b58a0246a54f1d496ff8f96a62e0e6b78736a5e352287a"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [

josePhoenix

{
 "metadata": {
  "name": "",
  "signature": "sha256:01532456f242d56d6873439d39dda566e17787c7f3a4150601ecce6539b8ff63"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [

josePhoenix

In [1]: data_specs = [
   ...:     (1.234, 'Foo', 3),
   ...:     (55.4, 'Bar', 1),
   ...:     (6.7, 'Foo', 4),
   ...:     (1.9, 'Foo', 4),
   ...: ]

In [2]: # Find the matching tuples

In [3]: data_matching = [item for item in data_specs if item[1] == 'Foo' and item[2] == 4]

josePhoenix

josePhoenix

I loathed this book when I was an undergraduate student, more so than any textbook I've ever encountered before or since. And what do you know? Reading it again, sixteen years later, I loathe it just as much. This was the book that instilled in me a hatred of everything and anything to do with optics, a hatred that persists to this day. It is the very fact that I hate optics so much that had me reading this book again. I figured that if I were to try to learn optics again, this time more successfully, perhaps the best way to start was to return to where it all first went wrong, to go over Hecht & Zajac again to see what they say, evaluate it in the light of the the current older, more knowledgeable me, and remember the points they cover for comparison when I read a subsequent text. Just what is it that makes this book so very bad? + Any textbook can have convoluted physical explanations that make no sense. + Likewise any textbook can include plodding mathematical explanations that get so worn down explaining 

josePhoenix

josePhoenix

$ brew install -vv gal-sim==> Installing gal-sim from homebrew/homebrew-science
/usr/bin/env python -c import pyfits
/usr/bin/env python -c import numpy
Warning: Formula file is modified!
Building from source because homebrew/Library/Taps/homebrew/homebrew-science/gal-sim.rb has local changes
To install from a bottle instead, run with --force-bottle
/usr/bin/env python -c import pyfits
/usr/bin/env python -c import numpy
==> Downloading https://github.com/GalSim-developers/GalSim/archive/v1.2.0.tar.gz
Already downloaded: /Library/Caches/Homebrew/gal-sim-1.2.0.tar.gz

josePhoenix

file /private/tmp/gal-sim20150730-38945-y869q2/GalSim-1.2.0/SConstruct,line 1655:
	Configure(confdir = .sconf_temp)
scons: Configure: Checking for C++ header file fftw3.h... 
.sconf_temp/conftest_0.cpp <-
  |
  |#include <fftw3.h>
  |
  |
g++ -o .sconf_temp/conftest_0.o -c -O2 -fno-strict-aliasing -Iinclude .sconf_temp/conftest_0.cpp
scons: Configure: yes