A simple class that enables you to download (workbooks) or (csv from views) from a Tableau Server.

tableau_server_export.py

import tableauserverclient as TSC
import pandas as pd
from io import StringIO

class Tableau_Server(object):
    
    """docstring for ClassName"""
    def __init__(self,username, password,site_id,url, https = False):
        super().__init__() # http://stackoverflow.com/questions/576169/understanding-python-super-with-init-methods

        # authorize 
        tableau_auth = TSC.TableauAuth(username, password, site_id)
        server = TSC.Server(url)
        server.add_http_options({'verify': https}) # if not https server skip warnings
        server.auth.sign_in(tableau_auth)
        server.use_highest_version() # make sure to use the same api version as server
        server.server_info.get()
        
        self.server = server
        
        # get all workbooks in site
        self.all_books, pagination_item = self.server.workbooks.get()
        self.all_books_names = [wb.name for wb in self.all_books]
        print('\nThere are {} workbooks on "{}" Tableau Server site.'.format(pagination_item.total_available,site_id))

    def check_wb_name(self,wb_name):
        # check if workbook exists
        if wb_name not in self.all_books_names:
            print('Workbook with name {} not found.'.format(wb_name))
            return


    def download_wb(self,wb_name):
        # downlaod workbook given workbook name
        self.check_wb_name(wb_name)
        wb = [w for w in self.all_books if w.name == wb_name][0]
        self.server.workbooks.download(wb.id)
    
    
    def download_view_csv(self,wb_name, view_name = None):
        # downlaod view csv given workbook name and view name (optional)
        self.check_wb_name(wb_name)
        wb = [w for w in self.all_books if w.name == wb_name][0]
        # request views
        self.server.workbooks.populate_views(wb) 
        views = [view for view in wb.views]
        view_names = [view.name for view in wb.views]
        
        if view_name is not None and view_name not in view_names:
            print('View with name "{}" not found in workbook "{}". Below are available views.'.format(view_name,wb_name))
            print(view_names)
            return
        
        # get either first view or user defined view
        view_item = [view for view in views][0]
        if view_name is not None:
            view_item = [view for view in views if view.name == view_name][0]
        self.server.views.populate_csv(view_item) # request view csv
        
        # Perform byte join on the CSV data
        string = StringIO(b''.join(view_item.csv).decode("utf-8"))
        df = pd.read_csv(string, sep=",")
        # pivot view csv so its in wide format if there is "Measure Values" column
        # if not just save to csv
        if 'Measure Values' not in df.columns.values:
            df.to_csv('{}.csv'.format(view_item.name), index = False)
            return
        df['Measure Values'] = pd.to_numeric(df['Measure Values'].str.replace('\\,|\\$|\\%', ''))
        cols = [c for c in df.columns.values if c not in ('Measure Values','Measure Names')]
        df = pd.pivot_table(df,values = 'Measure Values', columns = 'Measure Names', index = cols).reset_index()
        df.to_csv('{}.csv'.format(view_item.name), index = False)

def main():
    username = ''
    password = ''
    site_id = ''
    url = ''

    ts = Tableau_Server(username,password,site_id,url)
    
    # download a workbook
    ts.download_wb('First Touch Report')
    
    # download a view csv
    ts.download_view_csv('First Touch Report','First Touch Report')
    
if __name__ == '__main__':
    main()
            

Thank you very much, lots of pages have the populate_csv example but none of them thought about pivoting the table, the "Measure Values" in a single column is awful.

Hello again.

Allow me to suggest something. In order to avoid the alphabetic order for the columns you can store the original order in a list and then pass that list to the final datafarme. Also order the rows by the "cols" index would be helpfull.

Something like this:

cols_order = table_dataframe_raw['Measure_Names'].unique().tolist()
cols = [c for c in table_dataframe_raw.columns.values if c not in (
    'Measure_Values', 'Measure_Names')]
table_dataframe = table_dataframe_raw.pivot(
    index=cols, columns=['Measure_Names'], values='Measure_Values')[cols_order]
table_dataframe.sort_values(cols)

Let me know what you think!

hey @rodrigopizzano,
i need to retain columns order of the view (avoid the alphabetic order), so can you help with the case when Measure Names and Measure Values is not present. It would be great if you can share your reorganise function here in the gist comments. Thanks.