rjollet

<div style="white-space:nowrap;font-family:Helvetica, Arial;font-size: 14px;vertical-align:middle;white-space:nowrap">
    <h1 style="white-space:nowrap;font-size:18px; color:#36465d;line-height: 1;margin: 0 0 8px 0">Renaud Jollet De Lorenzo</h1>
    <h4 style="white-space:nowrap;font-size:14px; color:#333;line-height: 1;margin: 0 0 8px 0;font-weight: 400;">Master Student | <a href="http://blog.renaudjollet.info" style="white-space:nowrap;text-decoration: none; color: #333">blog.renaudjollet.info</a></h4>
    <h6 style="white-space:nowrap;font-size:14px; color:#999;line-height: 1;margin: 0;font-weight: 400;"><a href="https://keybase.io/rjollet" style="white-space:nowrap;text-decoration: none; color: #999">BE2A D008 9411 A141</a> | <span style="white-space:nowrap;text-decoration:none!important;color: #999!important">+886 9 780 184 30</span></h6>
    <div style="white-space:nowrap;margin-top: 8px">
        <a style="white-space:nowrap;display:inline-block;" href="http://blog.renaudjollet.info">
            <img

rjollet

// mongo <host>/<dbname> -u <username> -p <password> export.js > out.csv
// mongo local --quiet export_text.js > tweets_text.csv

print("id\tlang\ttext");
db.sample_stream.find({}, { 'tweet.id_str': 1, 'tweet.lang': 1,'tweet.text': 1, _id: 0}).forEach(function(doc){
  print(doc.tweet.id_str + "\t" + doc.tweet.lang + "\t" + doc.tweet.text.replace(/\t|\n|\r/g, ' '));
});

rjollet

conda create --name spark python=3
source activate spark
conda install jupyter
pip install --pre toree
jupyter toree install --user

rjollet

// mongo <host>/<dbname> -u <username> -p <password> export.js > out.csv
// mongo local --quiet export.js > out.csv

print("id, in_reply_to_status_id, user_id");
db.sample_stream.find({}, { 'tweet.id_str': 1, 'tweet.in_reply_to_status_id_str': 1 , 'tweet.user.id_str': 1, _id: 0}).forEach(function(doc){
  print(doc.tweet.id_str + "," + doc.tweet.in_reply_to_status_id_str + "," + doc.tweet.user.id_str);
});

rjollet

require 'yaml'
require 'csv'

def hashs_to_array(hashs)
  array = []
  hashs.each do |hash|
    array << hash.keys if array == []
    array << hash.values
  end
  array

rjollet

require 'yaml'
require 'csv'

def array_to_hashs(array)
  keys = array[0]
  data = array[1..-1]
  data.map { |values| Hash[keys.zip(values)] }
end

array = CSV.read(ARGV[0], col_sep: "\t")

rjollet

def fizzbuzz(size, &output)
  (1..size).map do |number|
    res = ''
    res << 'Fizz' if (number % 3).zero?
    res << 'Buzz' if (number % 5).zero?
    res = res == '' ? number : res
    yield res if output
    res
  end
end

rjollet

Keybase proof

I hereby claim:

• I am rjollet on github.
• I am rjollet (https://keybase.io/rjollet) on keybase.
• I have a public key whose fingerprint is 602D 9B5A D59A BD02 419D DCF6 BE2A D008 9411 A141

To claim this, I am signing this object:

rjollet

import json
import argparse

parser = argparse.ArgumentParser(description='Compare two corpus of words each corpus is a text file containing one word per line.')
parser.add_argument("--corpus1", help="file contening the corpus1")
parser.add_argument("--corpus2", help="file file contening the corpus2")
args = parser.parse_args()

if(args.corpus1 and args.corpus2):
    corpus1 = set(open(args.corpus1,'r'))

rjollet

class ShortStringPacker
  ## Packs a short string into a Fixnum
  # Arguments:
  #   str - String object
  # Returns: a Fixnum object
  def self.pack(str, nb_bits=5)
    res = 0
    str.each_char.with_index do |char, index|
      res = res | ((char.ord - 'a'.ord + 1)  << nb_bits*(str.length-(index+1)))
    end