https://kedro.org/ logo
Join Slack
Powered by
I'm having issues, because when I import GBQQueeyD...
# plugins-integrations
o
I'm having issues, because when I import GBQQueeyDataSet with the querie and project, everything run fine, taking the credentials in the env generated with 
gcloud auth login
But when I run 
kedro run
and its slicing variations or 
kedro ipython
it rise problems with permisions.
d
can you post a stack trace?
and are you importing directly in Python or using the YAML api?
o
I already erased all the thing (project and venv), because it was taking me so much time to solve it.
I was follwing the documentation at https://kedro.readthedocs.io/en/stable/kedro.extras.datasets.pandas.GBQQueryDataSet.html without pass th credentials in the .yaml but previously logged on GCP by CLI. So, when I ran 
kedro ipython
a issue with permision arise. And the same when I ran 
kedro run
. But if I run
Copy code
from kedro.extras.datasets.pandas import GBQQueryDataSet

sql = "SELECT * FROM dataset_1.table_a"

data_set = GBQQueryDataSet(sql, project='my-project')

sql_data = data_set.load()
changing the query to my proyect.dataset.table, the CLI credentials works well.
What I was trying to do is to connect with Google BigQuery with the credentials in CLI, without paste the credentials in gcp-conections\conf\local\credentials.yml
d
you’re not providing credentials to the 
GPQQueryDataSet
class
it accepts a credentials argument
o
Ok. So, it is mandatory to do it in that way: passing the credentials from the .yaml. I'll give it a try again.
d
no you can use the python api
but you do need to provide them
data_set = GBQQueryDataSet(sql, project='my-project')
this example only works if you’ve authenticated on an env level
credentials (*`Union`*[*`Dict`*[
str
, 
Any
], 
Credentials
, 
None
]) – Credentials for accessing Google APIs. Either 
google.auth.credentials.Credentials
object or dictionary with parameters required to instantiate 
google.oauth2.credentials.Credentials
. Here you can find all the arguments: https://google-auth.readthedocs.io/en/latest/reference/google.oauth2.credentials.html
o
If I add the call to 
google.auth.credentials.Credentials
in the node code in Python, the dataset class will find the credentials in the env. Y prefere something like that, that do not imlly to put the credentials in the. yaml
d
You can also bypass authentication in Kedro and do it on an environment level https://stackoverflow.com/questions/35159967/setting-google-application-credentials-for-bigquery-python-cli
❤️ 1
o
Thank you so much. It was solved with
Copy code
gcloud auth application-default login
So, it's no need to store the credentials in the .yaml nor to do anything else. The data now is ingested in 
kedro ipython
and when I run the pipeline with 
kedro run:+1:
d
💪
the credentials file is designed to give you a consistent way of doing things if you don’t have an environmental version of doing this
so this is the preferred pattern
o
I have no doubts about the consistency that Kedro give us. Thank you so much. Is becasue of that that I'm porting everything to Kedro framework. All this work around is just because of the IT department restrictions, due to security affairs.
K 1
d
awesome, shout if you have any other questions!
👍 1
3 Views
Open in Slack
PreviousNext
Powered by