<@U06C980PQTX>, <@U06CQGUPTPB> Sharing back this q...
# plugins-general
a
@ricky_renner, @hassan_syyid Sharing back this quick summary back to the channel in a new thread, since this original thread is now quite long and covering multiple topics. As discussed in the thread: 1. Meltano does not support 'settings write-back' by the tap/target - although we do have a ticket open to explore those requirements. 2. Meltano by default will generate a new config file each time the tap and target are invoked, cleaning up those files after execution for security purposes. 3. @ricky_renner has documented here that
tap-quickbooks
does support and/or expect this to handle a rotating
refresh_token
. 4. To my knowledge, we don't (yet) have a way to run meltano
elt
with a self-managed config file. For past use cases, we generally found there was some workaround in the tap itself. So far, just
tap-jira
and
tap-quickbooks
have been identified. I believe
tap-jira
had a workaround so it dropped off the priority. @hassan_syyid mentioned that the
tap-quickbooks
setting called
is_sandbox
may be a workaround but I can't speak to that myself.
New 🧵
r
@aaronsteers the
is_sandbox
setting did not make a difference unfortunately because although the tap itself edits the config file, Meltano is not picking up on those changes.
a
Thanks for confirming.
n
I’ve been running tap-quickbooks as an extract only till now, would love the ability to run
meltano elt
incorporating a config override for the refresh token. Currently i persist the token file, and grab it at execution time, and persist it again once the tap has updated the refresh token.
Copy code
# get the latest refresh token
gsutil cp <gs://cloudbuild_352875507036099891800/config.json> .

# run tap using persisted config 
tap-quickbooks --config config.json --catalog config/catalog.json > quickbooks.json

# save refresh token before parsing data
gsutil cp config.json <gs://cloudbuild_352875507036099891800>
a
@nigel_vining and @ricky_renner - Would one of you be interested+available to join #C01QS0RV78D tomorrow by chance? Or on another Wednesday? I added a few thoughts on coming up with a viable spec here in the issue: https://gitlab.com/meltano/meltano/-/issues/2710#note_801310978
r
@aaronsteers I'd love to join one of the office hours. I don't think I can do noon tomorrow though
n
Thanks @aaronsteers, same as @ricky_renner, i would be happy to join an office hours, but would need to be next week for me. That feature is very topical as i am about to move away from running the legacy tap extract pattern and leverage meltano elt 🙂
a
Great! Let's plan to discuss next week then, and I'm the meanwhile we can keep the conversation going in async, here and in the issue.