Setting Types to Uploaded Data for Bulk Import to Big Query

App Building
1

Hi
I'm working on a bulk update for Big Query.
I have a button to upload and parse a csv.
The results populate a table.
Using 'table1' widget, I manually set the table types to what I need.
When I run the upload it is rejected saying 'type is string and not BIGNUMERIC'

But, if I make the columns editable, change all the numeric values a bit, and change the bulk import from {{ table1.data }} to {{ table1.recordUpdates }} everything works.

So it seems {{ table1.data }} is not respecting the types I set in columns 'table1' widget.
Is there a way to refresh the table data to update the types?
I checked various {{ table1.xxxxx }} which returned an array - no success. And if I don't make an update to each row, then {{ table1.recordUpdates }} won't populate all the rows.

2

Hey @James_76 :wave: Changing the column type in the table doesn't actually reflect back in the table1.data property :pensive: I think your best bet is to do this using JS, actually!

Something like:

{{ table1.data.map(row => ({...row, id: parseFloat(row.id)})) }}

Would that work for you?

3

Hi!
Thanks for the reply.
Where would I do the mapping?
As a transformer in the insert bulk records query, or somewhere in table1.
If in table1 will the reference name change?

Are there any tutorials on setting data types?

4

Found my solution in retool fundamentals lesson 5 :grinning:
But I'm using Big Query and trying to push nulls, so now facing this issue:

1 Like
5

:pensive: :pensive: darn. Let me check in on that request again!

6

So the mapping does not work consistently, it only works if the result is an integer. If I have decimals it will not convert. I tried slicing the string to cap the decimals - this fails too, interesting if I slice (0,1) it works as the result is an int.

I tried importing BigNumber and BigDecimal libraries, but could not get either to work.
Is there any way to get this package

loaded into retool - it will give me access to the bignumeric type :pray: