benchmarking the Tableau Extract API

testing_extract_api

from random import randrange, random, choice
import dataextract as tde
import os
import string
import time
import csv

def write_to_extract(rows_to_write, name):

    start = time.time()

    current_file = name + '.tde'
    with tde.Extract(current_file) as extract:
        tableDef = tde.TableDefinition()
        tableDef.addColumn('rowID', tde.Type.INTEGER)
        tableDef.addColumn('tag', tde.Type.CHAR_STRING)
        tableDef.addColumn('value', tde.Type.DOUBLE)
        table = None

        # 'Extract' is a magic term, you just need to use it.
        if not extract.hasTable('Extract'):
            # Table does not exist, so create it.
            print "Writing", rows_to_write, "rows to a tde"
            table = extract.addTable('Extract', tableDef)
        else:
            # Table exists, so append the new data.
            print "Appending", rows_to_write, "rows to a tde"
            table = extract.openTable('Extract')

        new_row = tde.Row(tableDef)

        print "Progress:",
        end = rows_to_write
        progress = 0
        for i in xrange(end):
            new_row.setInteger(0, i)
            new_row.setCharString(1, choice(string.ascii_lowercase))
            new_row.setDouble(2, random())
            table.insert(new_row)
            # this is just a complicated way to show a progress bar
            if int((float(i) / end) * 100) % 5 == 0 and progress != int((float(i) / end) * 100):
                progress = int((float(i) / end) * 100)
                if progress % 10 == 0:
                    print str(progress)+'%,',
        print "done!"
        
    elapsed_time = round(time.time()-start, 1)
    return elapsed_time

def main():
    rows_to_write = int(raw_input("Create a .tde with how many rows? "))

    name = "test"
    tde_time = write_to_extract(rows_to_write, name)
    tde_size = os.path.getsize(name + ".tde")
    print "*" * 60

    print "\nSUMMARY:"
    print "Time taken to create tde:", tde_time, "seconds"
    print "Size of tde file (bytes):", tde_size
    print "\r"

main()

This script will append data to an already existing .tde or create a new file (lines 20-28). There is a great tutorial on the Extract API here: http://www.tableau.com/learn/tutorials/on-demand/extract-api-introduction and you can download the dataextract Python module here: http://www.tableau.com/data-extract-api