DBI: cannot run query without specifying dataset

[czeildi]

I get an error if I do not specify dataset in the second case:

library(DBI)
library(bigrquery)

billing <- bq_test_project() # replace this with your project ID 
sql <- "SELECT year, month, day, weight_pounds FROM `publicdata.samples.natality`"

con <- dbConnect(
    bigrquery::bigquery(),
    project = "publicdata",
    dataset = "samples",
    billing = billing
)
dbGetQuery(con, sql, n = 10)
dbDisconnect(con)

con <- dbConnect(
    bigrquery::bigquery(),
    project = "publicdata",
    billing = billing
)
dbGetQuery(con, sql, n = 10)
# Error: dataset is not a string (a length one character vector).
dbDisconnect(con)

I used my own billing project, otherwise ran the same code.

sorry for no reprex, it somehow ran for a lot of time without returning

[czeildi]

related: project is a required argument and you must provide a valid project id (although it can be different from both billing and any project actually used in your sql queries later)

as you can run a query using tables from different projects in Google Big Query I do not see the need for this argument to be required

[hadley]

Looks like I missed this case in my tests; dataset is definitely supposed to be optional.

[czeildi]

Regarding my previous comment on needing project:

I can see now that billing is optional: so it is enough to populate project with billing project and than there's technically no superfluous information. However, if possible I would still prefer the more explicit version: specify the billing explicitly (this is unique) and the project should receive the billing as default value and can be specified otherwise by user

[hadley]

I agree that would be optimal, but I think it would be too much of a change to an existing function.