infer_tolerance_from_cost and inferred_tolerance_multiplier

[Filippo Valsorda]

I have a brokerage account that receives fractional shares. I use a single transaction to enter the income and taxes from the paystub, and the deposit into the account. I use infer_tolerance_from_cost because the fractional amount of shares is more coarse than the USD values.

It looks like the fractional amount is rounded down in the brokerage account, so I tried setting inferred_tolerance_multiplier to 2, but it did not have any effect. Looking at the code, inferred_tolerance_multiplier does not apply to the cost_tolerance.

option "infer_tolerance_from_cost" "TRUE"
option "inferred_tolerance_multiplier" "2"

Am I missing something, or would it be useful to make inferred_tolerance_multiplier apply to the cost tolerance?

Thank you,
Filippo

[Martin Blais]

I'd need a simple example of your input to fully answer this.

Also, you can try "bean-doctor context" on your transaction to view the calculated tolerances.

--
You received this message because you are subscribed to the Google Groups "Beancount" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beancount+...@googlegroups.com.
To post to this group, send email to bean...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/beancount/1537749222.3543211.1518074208.080CDA04%40webmail.messagingengine.com.
For more options, visit https://groups.google.com/d/optout.

[Filippo Valsorda]

Here's an example file, bean-check and bean-doctor output.

https://gist.github.com/FiloSottile/45093ce512bed582c15d352523206608

I believe it's just a matter of applying inferred_tolerance_multiplier to cost_tolerances in beancount.core.interpolate.infer_tolerances()

P.S. Please keep me CC'd, as I am not subscribed to the mailing list.