Closed NickleDave closed 1 year ago
It's a malformed row or identifier - I can see it in your logs:
ERROR:rse.utils.urls:Cannot find endpoint https://api.github.com/repos/tree/v0.6.5.
But the client should still skip and provide an error message. Give this branch a try. https://github.com/rseng/rse/pull/79. I'm going for a run - back after!
Thank you for starting that branch but I'm not quite sure I understand your reasoning for skipping a malformed row.
Wouldn't that just fail silently without the user not knowing they'd lost a record?
Maybe it would be helpful here to have a clearer error message. Something that prevents us from ending up at the cryptic 'NoneType' object has no attribute 'items'
Yes I added a warning.
It’s up to the user to run it again, but if they have just used up extensive numbers of their git api limit and I throw an error and they lose the parsed data, that’s a very frustrating outcome.
Speaking from personal experience.
if they have just used up extensive numbers of their git api limit and I throw an error and they lose the parsed data, that’s a very frustrating outcome.
Ah, this makes sense.
Could there be a finally
block that writes the parsed data even if there's an exception raised?
Could there be a finally block that writes the parsed data even if there's an exception raised?
Nvm, I see there's nesting that makes this non-trivial
Passing out for the night but will return to try that branch
Fixed by #79
Description
I'm back to trying to run
rse import
on an edited copy of the google-sheet from @rhine3's bioacoustics-software repo.It now crashes with this
Full traceback below. I think
rse
makes it through all theurl
s now and the source of this crash is something else? Unless I'm reading the traceback wrong somehowWhat I Did