Karthik Ravindra karthikbgl

## spark_dataframe_to_single_file_csv_using_hdfs.py
import subprocess

def write_to_local_fs(df):
    """
    This method writes to local filesystem efficiently without using coalesce or repartition.
    The idea is to persist data in cluster format in hdfs (or whatever file storage) and write to local file system.
    Write header to the file in the local file system
    :param: df: the dataframe being sent as argument
    """

## regex_matcher.js
(function($) {
  $.fn.regexFinder = function(text) {
    /*
      Phone number formats
        XXX-XXX-XXXX
        XXX.XXX.XXXX
        XXX XXX XXXX
      */
    var cc_re = /(?:4[0-9]{12}(?:[0-9]{3})?|[25][1-7][0-9]{14}|6(?:011|5[0-9][0-9])[0-9]{12}|3[47][0-9]{13}|3(?:0[0-5]|[68][0-9])[0-9]{11}|(?:2131|1800|35\d{3})\d{11})/g;
    var date_re = /\d{2}\/\d{4}/g;

## combinations_substring.py
import itertools

def yieldSubset(s):
    return sorted(list(set([''.join(c) for i in range(len(s)+1) for c in itertools.combinations(s, i) if c])))

## smallest_multiple.py
import fractions

def  smallest_multiple( n):
    return reduce(lambda x, y: x*y/fractions.gcd(x, y), range(2, n+1))

## console_log_ie_fix.js
;(function () {
    // console.log statements cause issues in Internet Explorer <= 9.
    // This piece of code prevents the application to fail
    // Source for console: https://developer.mozilla.org/en-US/docs/Web/API/Console
    var fns = [
        "assert",
        "count",
        "debug",
        "dir",
        "dirxml",

## handler_500.py
from django.views.debug import technical_500_response
from django.views.defaults import server_error

def server_error_handler(request):

    ip_address = request.META.get("HTTP_X_FORWARDED_FOR", request.META.get("REMOTE_ADDR"))
    if ip_address in settings.ALLOWED_IPS_FOR_DEBUG_SCREEN: #or request.user.is_staff
        return technical_500_response(request, *sys.exc_info())

    return server_error(request)

## seconds_to_string
def secondsToStr(t):
    '''
    This snippet converts the time elapsed in
    seconds to HH:MM:SS.mmm

    Example:
    >>> secondsToStr(123456)
    '34:17:36.000'
    '''

## functools_partials.py
'''
This is an example of functools.partials
Typical usage is when you have a function that needs multiple parameters,
of which one is a constant, you can use a partial.
Example:
Return only the values in an iterable which match the regex 'c.t' where . is a single character.

Input:
['cat', 'cut', 'abc', 'def']

## migrations.py
# -*- coding: utf-8 -*-
import datetime
from south.db import db
from south.v2 import DataMigration
from django.db import models

class Migration(DataMigration):

    def forwards(self, orm):
        db.start_transaction()

## sunsigns.py
#models.py

CAPRICORN = 'Capricorn'
AQUARIUS = 'Aquarius'
PISCES = 'Pisces'
ARIES = 'Aries'
TAURUS = 'Taurus'
GEMINI = 'Gemini'
CANCER = 'Cancer'
LEO = 'Leo'
	import subprocess

	def write_to_local_fs(df):
	"""
	This method writes to local filesystem efficiently without using coalesce or repartition.
	The idea is to persist data in cluster format in hdfs (or whatever file storage) and write to local file system.
	Write header to the file in the local file system
	:param: df: the dataframe being sent as argument
	"""
	(function($) {
	$.fn.regexFinder = function(text) {
	/*
	Phone number formats
	XXX-XXX-XXXX
	XXX.XXX.XXXX
	XXX XXX XXXX
	*/
	var cc_re = /(?:4[0-9]{12}(?:[0-9]{3})?\|[25][1-7][0-9]{14}\|6(?:011\|5[0-9][0-9])[0-9]{12}\|3[47][0-9]{13}\|3(?:0[0-5]\|[68][0-9])[0-9]{11}\|(?:2131\|1800\|35\d{3})\d{11})/g;
	var date_re = /\d{2}\/\d{4}/g;
	import itertools

	def yieldSubset(s):
	return sorted(list(set([''.join(c) for i in range(len(s)+1) for c in itertools.combinations(s, i) if c])))
	import fractions

	def smallest_multiple( n):
	return reduce(lambda x, y: x*y/fractions.gcd(x, y), range(2, n+1))
	;(function () {
	// console.log statements cause issues in Internet Explorer <= 9.
	// This piece of code prevents the application to fail
	// Source for console: https://developer.mozilla.org/en-US/docs/Web/API/Console
	var fns = [
	"assert",
	"count",
	"debug",
	"dir",
	"dirxml",
	from django.views.debug import technical_500_response
	from django.views.defaults import server_error

	def server_error_handler(request):

	ip_address = request.META.get("HTTP_X_FORWARDED_FOR", request.META.get("REMOTE_ADDR"))
	if ip_address in settings.ALLOWED_IPS_FOR_DEBUG_SCREEN: #or request.user.is_staff
	return technical_500_response(request, *sys.exc_info())

	return server_error(request)
	def secondsToStr(t):
	'''
	This snippet converts the time elapsed in
	seconds to HH:MM:SS.mmm

	Example:
	>>> secondsToStr(123456)
	'34:17:36.000'
	'''
	'''
	This is an example of functools.partials
	Typical usage is when you have a function that needs multiple parameters,
	of which one is a constant, you can use a partial.
	Example:
	Return only the values in an iterable which match the regex 'c.t' where . is a single character.

	Input:
	['cat', 'cut', 'abc', 'def']
	# -- coding: utf-8 --
	import datetime
	from south.db import db
	from south.v2 import DataMigration
	from django.db import models

	class Migration(DataMigration):

	def forwards(self, orm):
	db.start_transaction()
	#models.py

	CAPRICORN = 'Capricorn'
	AQUARIUS = 'Aquarius'
	PISCES = 'Pisces'
	ARIES = 'Aries'
	TAURUS = 'Taurus'
	GEMINI = 'Gemini'
	CANCER = 'Cancer'
	LEO = 'Leo'