paultopia’s gists

paultopia / addcss.py

Created October 8, 2015 22:00

Easy way to add CSS or whatever to header of a bunch of HTML files at once.

	# USAGE:
	#
	# To add code to the end of every <head> tag (like a css link, a font link, etc.) to quick-format an entire website:
	# 1. Start in the top-level-directory of the site. Put this file there.
	# 2. Add your formtting for the <head> tag to the formatme variable.
	# EXAMPLE: mine was '<link href="https://fonts.googleapis.com/css?family=Halant:300" rel="stylesheet" type="text/css"><link rel="stylesheet" href="http://paul-gowder.com/conlawII/prettify.css">'
	# be sure to either escape quotes or use single quotes to demarcate the string and double-quotes in the html/vice versa
	# 3. Run this script.
	# 4. Bam. Every html page in in the top-level directory and all its subdirectories now has the formatting you want.

paultopia / wordcount.py

Last active November 18, 2015 20:52

quick and dirty script to take a bunch of documents and output a printed list of word counts for each document as well as for whole

	# assumes documents are provided in the form of a list of (docid, doctext) tuples named thedocslist. docid = int/string/float; doctext = string

	import nltk
	import string
	from collections import Counter

	# get rid of punctuation, numbers; make all lowercase. no stemming.

	counterslist = []
	for onedocument in thedocslist:

paultopia / toolchain.md

Last active November 30, 2015 20:16

The toolchain of a techy-ish political theorist/lawyer

My toolchain (on OSX)

I do all four of the following often:

Write lengthy academic prose.
Write code.
Analyze data and do other math-y things.

paultopia / makeslide.py

Last active May 3, 2017 20:41 — forked from aaronwolen/slides.md

Pandoc template to generate reveal.js slideshows. (corrected from original to be compatible with 3.2.0 release of reveal.js); also added a quick python script to generate with less command-line ugliness.

	import argparse

	# this first bit is to enable multiline help text. apparently this is a known problem with argparse.
	# Solution jacked from http://stackoverflow.com/questions/3853722/python-argparse-how-to-insert-newline-in-the-help-text

	import textwrap as _textwrap
	class MultilineFormatter(argparse.HelpFormatter):
	def _fill_text(self, text, width, indent):
	text = self._whitespace_matcher.sub(' ', text).strip()
	paragraphs = text.split('\|n ')

paultopia / pgmd.py

Created December 1, 2015 05:42

Quick and easy command-line wrapper for pandoc conversions to html, pdf, and docx

	# The point of this script is that pandoc commandline syntax is painful and hard to remember.
	# I really only produce html, pdf, and docx. And I only ever use the defaults. Ergo, a script
	# (subsequently to be put in $PATH with path to python added to top to be runnable trivially) to
	# make it simple.
	#
	# usage: python pgmd.py INPUTFILE FORMAT[html/pdf/word]
	# that's it. easy.
	#
	# there are a handful of other options (output file, overwrite output file, append scripts and
	# css and such to html headers), details are in the commandline help via -h flag

paultopia / quickscrape.py

Last active December 27, 2015 18:24

OBSOLETE: use this instead: https://github.com/paultopia/spideyscrape

	# OBSOLETE.
	# GO HERE INSTEAD: https://github.com/paultopia/spideyscrape

	# very basic scraper-spider for those html books where there's a table of contents page that links to a
	# bunch of sub-pages with actual content. (Like the documentation for a bunch of libraries.)
	# WARNING: has no validation, assumes pages contain relative links and are all on the same site.
	# (this is an easy tweak but I don't have time today)
	# also assumes all content is vanilla html or at least can be accessed through vanilla html.
	#
	# pass ToC page through raw_input. This script scrapes every unique page linked from ToC and

paultopia / val_quickscrape.py

Last active December 27, 2015 18:25

OBSOLETE: use this instead: https://github.com/paultopia/spideyscrape

	# EDIT: this has now been upgraded to a full-fledged repo and is accepting PRs. This gist is no longer updating.
	# go here: https://github.com/paultopia/spideyscrape

	# This is a very basic scraper-spider for those html books where there's a table of contents page that links to a
	# bunch of sub-pages with actual content (like the documentation for a bunch of libraries).
	#
	# Dependencies: Beautiful soup 4 on Python 2.7.
	#
	# It assumes all content is vanilla html or at least can be accessed through vanilla html.
	#

paultopia / scrapewrap.py

Last active December 29, 2015 17:39

scrapewrap.py

	import sys
	import spideyscrape
	import console
	import os

	args = sys.argv[1:] # see if the user gave us a command line argument
	start = args[0] if args else raw_input('URL to crawl: ')
	html = spideyscrape.scrape(start)
	filename = spideyscrape.savePage(html)
	console.open_in(filename)

paultopia / worst_python_ever.py

Last active December 30, 2015 03:44

worst_python_ever.py

	# I think I've discovered a bit of Python code even more dangerous than https://github.com/ajalt/fuckitpy

	class string(str):
	def __call__(self):
	try:
	exec self
	except Exception:
	pass

	evil = string('print "EVIL"')

paultopia / never_do_this.py

Last active June 3, 2016 22:10

never_do_this.py

	# NEVER DO THIS EXCEPT AS A PRANK ON YOUR WORST ENEMY

	class foo(str):
	def __call__(self):
	try:
	exec self
	except Exception:
	pass

	str = foo

Paul Gowder paultopia

My toolchain (on OSX)