djay
commented
1 year ago @federicoscabbia I have the same problem. I wrote this code to get around it. It handles both the case where there is no worksheet (dashboardFilter) and when the filter doesn't list the values but still accepts them.
def force_setFilter(wb, ws_name, columnName, values):
"setFilter but ignore the listed filter options. also gets around wrong ordinal value which makes index value incorrect"
# TODO: remove if they fix https://github.com/bertrandmartel/tableau-scraping/issues/50
scraper = wb._scraper
tableauscraper.api.delayExecution(scraper)
ws = next((ws for ws in wb.worksheets if ws.name == ws_name), None)
filter = next((
{
"globalFieldName": t["globalFieldName"],
}
for t in (ws.getFilters() if ws is not None else [])
if t["column"] == columnName
), None
)
try:
if filter is None:
resp = tableauscraper.api.dashboardFilter(scraper, columnName, [values] if not isinstance(values, list) else values)
else:
payload = (
("dashboard", scraper.dashboard),
("globalFieldName", (None, filter["globalFieldName"])),
("qualifiedFieldCaption", (None, columnName)),
("membershipTarget", (None, "filter")),
("exclude", (None, "false")),
("filterValues", (None, json.dumps(values))),
("filterUpdateType", (None, "filter-replace"))
)
r = scraper.session.post(
f'{scraper.host}{scraper.tableauData["vizql_root"]}/sessions/{scraper.tableauData["sessionid"]}/commands/tabdoc/dashboard-categorical-filter',
files=payload,
verify=scraper.verify
)
scraper.lastActionTime = time.time()
if r.status_code >= 400:
raise requests.exceptions.RequestException(r.content)
resp = r.json()
errors = [
res['commandReturn']['commandValidationPresModel']['errorMessage']
for res in resp['vqlCmdResponse']['cmdResultList']
if not res['commandReturn'].get('commandValidationPresModel', {}).get('valid', True)
]
if errors:
wb._scraper.logger.error(str(", ".join(errors)))
raise tableauscraper.api.APIResponseException(", ".join(errors))
wb.updateFullData(resp)
return tableauscraper.dashboard.getWorksheetsCmdResponse(scraper, resp)
except ValueError as e:
scraper.logger.error(str(e))
return tableauscraper.TableauWorkbook(
scraper=scraper, originalData={}, originalInfo={}, data=[]
)
except tableauscraper.api.APIResponseException as e:
wb._scraper.logger.error(str(e))
return tableauscraper.TableauWorkbook(
scraper=scraper, originalData={}, originalInfo={}, data=[]
)
david-hammond
miriamventurini
Hi everyone, I think i have found a tableau that has selectables not in the specific worksheet but in the workbook instead and i do not know how to scrape all the years. I tried this code to select the different years but when i look for the filters it gives back an empty list. Probably the the filtes are set at the workbook level and not at the worksheet level, however 'TableauWorkbook' object has no attribute 'getFilters'.
Could you please help me with that?
This is the table i'm referring to: https://public.tableau.com/views/SocialistPartyScandinavianFederation/Story1
And this is the code I used to scrape the data (as you can see i can only scrape the data from 1916 as it is set as the default year in the table).