This message was deleted.
# help
s
This message was deleted.
f
Did your data change?
d
I changed the file but the basic calculation is the same.
f
What's the link to your notebook?
Also, assuming this is DuckDB, are you editing on a different device today?
d
I uploaded the data as a json and am trying to work with the output from the data table
f
you have NaNs in your coordinates (empty strings in your source data)
🙌 1
As a stop-gap solution you can use the condition
Copy code
WHERE isfinite("longitude") AND "longitude" > 0.149
🙌 1
d
Thank you @Fabian Iwand. I wondered if it was to do with those which is why I added the WHERE clause. Strange thing is it worked yesterday. I must be doing something slightly different.
f
Did your data change?
☝️
I changed the file
☝️
😃 1
d
Yes I changed the file, but the field names and data types and method were the same as far as I could tell. Eventually if I look hard enough I'll probably work out that I've done something dopey and missed it.
f
right, but surely the contents of your file changed, probably introducing those rows with empty sets of coordinates?
d
That could have been it. I only changed it to add some metrics and order the columns to make more sense. I thought it was the same number of rows but you're probably right.
f
File attachments aren't versioned (yet), so you can go back to the first version where you got the query working and see it running (and failing) with the current file.
d
Yeah I tried going back and realised the original file wasn't available. I need to figure out how to make the new file work anyway, which hopefully you've solved for me (too much going on here so I haven't checked yet).