Skip to main content

Python library to scrape data from Tableau viz

Project description

Tableau Scraper

PyPI CI codecov License

Python library to scrape data from Tableau viz

R library is under development but a script is available to get the worksheets, see this

Python

Install

pip install TableauScraper

Usage

Get worksheets data

from tableauscraper import TableauScraper as TS

url = "https://public.tableau.com/views/PlayerStats-Top5Leagues20192020/OnePlayerSummary"

ts = TS()
ts.loads(url)
workbook = ts.getWorkbook()

for t in workbook.worksheets:
    print(f"worksheet name : {t.name}") #show worksheet name
    print(t.data) #show dataframe for this worksheet

Try this on repl.it

Get a specific worksheet

from tableauscraper import TableauScraper as TS

url = "https://public.tableau.com/views/PlayerStats-Top5Leagues20192020/OnePlayerSummary"

ts = TS()
ts.loads(url)

ws = ts.getWorksheet("ATT MID CREATIVE COMP")
print(ws.data)

select a selectable item

from tableauscraper import TableauScraper as TS

url = "https://public.tableau.com/views/PlayerStats-Top5Leagues20192020/OnePlayerSummary"

ts = TS()
ts.loads(url)

ws = ts.getWorksheet("ATT MID CREATIVE COMP")

# show selectable values
selections = ws.getSelectableItems()
print(selections)

# select that value
dashboard = ws.select("ATTR(Player)", "Vinicius Júnior")

# display worksheets
for t in dashboard.worksheets:
    print(t.data)

Try this on repl.it

set parameter

Get list of parameters with workbook.getParameters() and set parameter value using workbook.setParameter("column_name", "value") :

from tableauscraper import TableauScraper as TS

url = "https://public.tableau.com/views/PlayerStats-Top5Leagues20192020/OnePlayerSummary"

ts = TS()
ts.loads(url)
workbook = ts.getWorkbook()

# show parameters values / column
parameters = workbook.getParameters()
print(parameters)

# set parameters column / value
workbook = workbook.setParameter("P.League 2", "Ligue 1")

# display worksheets
for t in workbook.worksheets:
    print(t.data)

Try this on repl.it

set filter

Get list of filters with worksheet.getFilters and set filter value using worksheet.setFilter("column_name", "value"):

from tableauscraper import TableauScraper as TS

url = 'https://public.tableau.com/views/WomenInOlympics/Dashboard1'
ts = TS()
ts.loads(url)

# show original data for worksheet
ws = ts.getWorksheet("Bar Chart")
print(ws.data)

# get filters columns and values
filters = ws.getFilters()
print(filters)

# set filter value
wb = ws.setFilter('Olympics', 'Winter')

# show the new data for worksheet
countyWs = wb.getWorksheet("Bar Chart")
print(countyWs.data)

Try this on repl.it

More advanced filtering options

  • You can specify dashboardFilter=True in order to use dashboard-categorical-filter API instead of categorical-filter-by-index API (related)

  • You can discard membershipTarget property from being sent in setFilter using setFilter('COLUMN','VALUE', membershipTarget=False) (related)

  • You can specify multiple filters for filters that enable that feature using setFilter('COLUMN', ['VALUE1','VALUE2'])

  • You can specify a "filter-delta" filter type adding the parameter filterDelta=True like the following setFilter('COLUMN','VALUE', filterDelta=True). This will discard all filters and add the one corresponding to ['VALUE'] in this case. This is helpful when all or some filters are selected by default, and you want to unselect them. The default behaviour (filterDelta=False) is filter-replace which sometimes doesn't work when filter multi-selection is possible in the dashboard. example

  • In last recourse, you can use indexValues property to directly specify the indices (if there is a bug in the library or anything comes up): setFilter('COLUMN', [], indexValues=[0,1,2])

Story points

Some Tableau dashboard have storypoints where you can navigate. To list the storypoints and go to a specific storypoints:

from tableauscraper import TableauScraper as TS

url = 'https://public.tableau.com/views/EarthquakeTrendStory2/Finished-Earthquakestory'
ts = TS()
ts.loads(url)
wb = ts.getWorkbook()

print(wb.getStoryPoints())
print("go to specific storypoint")
sp = wb.goToStoryPoint(storyPointId=10)

print(sp.getWorksheetNames())
print(sp.getWorksheet("Timeline").data)

Try this on repl.it

Level drill Up/Down

On some graph/table, there is a drill up/down feature used to zoom in or out data like this drill up/down

from tableauscraper import TableauScraper as TS

url = 'https://tableau.azdhs.gov/views/ELRv2testlevelandpeopletested/PeopleTested'
ts = TS()
ts.loads(url)
wb = ts.getWorkbook()

sheetName = "P1 - Tests by Day W/ % Positivity (Both) (2)"

drillDown1 = wb.getWorksheet(sheetName).levelDrill(drillDown=True, position=1)
drillDown2 = drillDown1.getWorksheet(sheetName).levelDrill(drillDown=True, position=1)
drillDown3 = drillDown2.getWorksheet(sheetName).levelDrill(drillDown=True, position=1)

print(drillDown1.getWorksheet(sheetName).data)
print(drillDown2.getWorksheet(sheetName).data)
print(drillDown3.getWorksheet(sheetName).data)

Try this on repl.it

The position parameter is default to 0. It doesn't seem to be present in the json configuration. If the default is not working try incrementing it or checkout the network tabs using Chrome devtools.

Download CSV data

For Tableau URL that have the download feature enabled, you can download full data using:

from tableauscraper import TableauScraper as TS

url = 'https://public.tableau.com/views/WYCOVID-19Dashboard/WyomingCOVID-19CaseDashboard'
ts = TS()
ts.loads(url)
wb = ts.getWorkbook()
data = wb.getCsvData(sheetName='case map')

print(data)

Note that in some Tableau server, the prefix used in the API url is different. As it's set in the javascript, it must be set manually if it's not the same as public.tableau.com like:

wb.getCsvData(sheetName='worksheet1', prefix="vud")

The prefix values, I've encountered are: vud and vudcsv. The default is vudcsv.

Try this on repl.it

Download Cross Tab data

For Tableau URL that have the crosstab feature enabled, you can download the crosstab using:

from tableauscraper import TableauScraper as TS

url = "https://tableau.soa.org/t/soa-public/views/USPostLevelTermMortalityExperienceInteractiveTool/DataTable2"

ts = TS()
ts.loads(url)
wb = ts.getWorkbook()

wb.setParameter(inputName="Count or Amount", value="Amount")

data = wb.getCrossTabData(
    sheetName="Data Table 2 - Premium Jump & PLT Duration")

print(data)

Go to sheet

Get list of all sheets with subsheets visible or invisible, ability to send a go-to-sheet command (dashboar button) :

from tableauscraper import TableauScraper as TS

url = "https://public.tableau.com/views/COVID-19VaccineTrackerDashboard_16153822244270/Dosesadministered"
ts = TS()
ts.loads(url)
workbook = ts.getWorkbook()

sheets = workbook.getSheets()
print(sheets)

nycAdults = workbook.goToSheet("NYC Adults")
for t in nycAdults.worksheets:
    print(f"worksheet name : {t.name}")  # show worksheet name
    print(t.data)  # show dataframe for this worksheet

Render tooltip

Get the tooltip html output when render-tooltip-server API is called. This is particularly useful when dealing with server side rendering dashboard:

from tableauscraper import TableauScraper as TS

url = "https://public.tableau.com/views/CMI-2_0/CMI"

ts = TS()
ts.loads(url)
workbook = ts.getWorkbook()
ws = workbook.getWorksheet("US Map - State - CMI")

tooltipHtml = ws.renderTooltip(x=387, y=196)
print(tooltipHtml)

Sample usecases

Server side rendering

If the tableau url you're working on is using server side rendering, data can't be extracted as is.

You can checkout if your tableau url is using server side rendering by opening chrome development console / network tab. You would notice the API calls have renderMode properties set to render-mode-server:

server side render mode

Server side rendering means that no data is sent to the browser. Instead, the server is rendering the tableau chart using images only and detects selection using mouse coordinates.

To extract the data, one thing that has worked with some tableau url was to trigger a specific filter that is not server-side-rendered. You can checkout the network tab on Chrome development console to check if the filter call is using or not server-side rendering or client-side-rendering with renderMode:

client side rendering

If the filter is only using client side rendering, you can list all filters and perform the filter for each value. This technique only works if the tableau data has "cleared" the filter by default otherwise the data is already cached when the tableau data is loaded, and since it's using server side rendering you can't access this data

Checkout the following repl.it for examples with tableau url using server side rendering:

Testing Python script

To discover all worksheets, selectable columns and dropdowns, run prompt.py script under scripts directory :

git clone git@github.com:bertrandmartel/tableau-scraping.git
cd tableau-scraping/scripts

#get worksheets data
python3 prompt.py -get workbook -url "https://public.tableau.com/views/PlayerStats-Top5Leagues20192020/OnePlayerSummary"

#select a selectable item
python3 prompt.py -get select -url "https://public.tableau.com/views/MKTScoredeisolamentosocial/VisoGeral"

#set a parameter
python3 prompt.py -get parameter -url "https://public.tableau.com/views/COVID-19DailyDashboard_15960160643010/Casesbyneighbourhood"

Settings

TableauScraper class has the following optional parameters :

Parameters default value description
logLevel logging.INFO log level
delayMs 500 minimum delay in millis between api calls

R

under R directory :

Rscript tableau.R

R library is under development

Dependencies

requirements.txt

  • pandas
  • requests
  • beautifulsoup4

Stackoverflow Questions

See those stackoverflow posts about this topic

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

TableauScraper-0.1.29.tar.gz (18.3 kB view details)

Uploaded Source

Built Distribution

TableauScraper-0.1.29-py3-none-any.whl (22.5 kB view details)

Uploaded Python 3

File details

Details for the file TableauScraper-0.1.29.tar.gz.

File metadata

  • Download URL: TableauScraper-0.1.29.tar.gz
  • Upload date:
  • Size: 18.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.1 importlib_metadata/4.6.1 pkginfo/1.6.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.43.0 CPython/3.7.6

File hashes

Hashes for TableauScraper-0.1.29.tar.gz
Algorithm Hash digest
SHA256 7fa031b9b4aacecb8b525ad1d5908623456c3975fc46bf88fb2b6fd13156288e
MD5 13ba9b3b1e8a620289262ba159dc2626
BLAKE2b-256 5ca19a8872dd93d598a710b8a61ffcd75897cf45eb1c8d0ac3507ab4496d683f

See more details on using hashes here.

File details

Details for the file TableauScraper-0.1.29-py3-none-any.whl.

File metadata

  • Download URL: TableauScraper-0.1.29-py3-none-any.whl
  • Upload date:
  • Size: 22.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.1 importlib_metadata/4.6.1 pkginfo/1.6.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.43.0 CPython/3.7.6

File hashes

Hashes for TableauScraper-0.1.29-py3-none-any.whl
Algorithm Hash digest
SHA256 bd0011d6c8da8b39f07147a6dc32ffb9008dfe0994070e4caf40e3573ffdb846
MD5 a2a4dc28342edda078c348a4fc1d60e1
BLAKE2b-256 f427383c7996dbaedbd743af92e77bea2478ff5a2a2f89401bec8005681ff1f5

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page