I think it would be reasonable to create a function like that and insert it
into the environment of a custom beanquery you build for yourself, or
perhaps just make a script.




On Thu, Aug 29, 2024, 18:04 DK <[email protected]> wrote:

> Thanks, the 10.00 fixed the rounding issue.
>
> In regard to the main issue. So I already use implicit_prices, but I
> wasn't aware of unique_prices. It indeed helped me identify some bugs
> caused by the import from GnuCash (or should I say, by GnuCash itself),
> where I'd have transaction like
>
> ```
> 2024-01-01 txn "Something"
>   Account1 10 EUR @@ 10 USD
>   Account2 10 EUR
> ```
>
> So the conversion is not needed. I removed these, and now flipping the
> initial transaction order, works correctly. I can't live with unique_prices
> on all the time, due to the fact that in real life, market prices are not
> unique per day (in fact, I had two ESPP batches that I sold on the same
> day, but one was sold at 313 USD while the other at 313.5 USD, hence
> unique_prices was complaining about it). But it, nevertheless, helpful to
> turn it on once in a while in order to identify bugs like this.
>
> This brings me, however, to another question. I conduct business in
> various currencies, but I have to file my tax reports in EUR. For some
> transaction, like the one above, I have explicit conversion rate. I bought
> a service for 10 USD, but my bank took 9.32 EUR based on their conversion
> rate. Hence, I'm obligated to report a loss of 9.32 EUR. I don't want the
> convert function to go and look at the closes rate on that date, I would
> like to have a function that converts the cost based on the price attached
> to a transaction, if there is one. If there is none, then I need to use ECB
> rate for that day in my reports. It would have been nice to have a convert
> function like that. On the other hand, maybe it's not the intended purpose
> of beancount, and instead I should get the original values to a python
> script and process them myself in the desired output for the tax
> authorities.
>
> Thanks for help!
>
> On Wednesday, August 28, 2024 at 2:09:06 PM UTC+2 [email protected] wrote:
>
>> On Wednesday, August 28, 2024 at 9:06:04 AM UTC+2 DK wrote:
>>
>> I recently switch to beancount, and I'm trying to set it up for both my
>> personal finances, and for my business.
>>
>> Consider the following transaction:
>>
>> ```
>> 2024-08-01 txn "Some Service" ""
>>     invoice: "xxx"
>>     Assets:Wise:EUR                           -9.32 EUR @@ 10 USD
>>     Expenses:MyComp:Operating:Software       10 USD
>>     Expenses:Misc:USD
>> ```
>> I purchased some service from a US company for 10 USD, but my Wise bank
>> account was charged with 9.32 EUR (The Expenses:Misc:USD is there because
>> apparently this transaction, as is, does not balance and is missing
>> 0.00000000000001 USD, or close to that, but that's a different issue).
>>
>>
>>
>> I think Martin considers this to be a bug, but you can work around it by
>> putting  10.00 instead of 10
>>
>> 2024-08-01 txn "Some Service" ""
>>     invoice: "xxx"
>>     Assets:Wise:EUR                           -9.32 EUR @@ 10.00 USD
>>     Expenses:MyComp:Operating:Software       10.00 USD
>>
>> check:
>>
>>
>> https://colab.research.google.com/drive/1laDSOEdxMC2b9yo3zX8e-gxyomM5Zuqm?usp=sharing
>>
>>
>>
>>
>> ```
>>
>>  Ideally, if I run the report now, I should get the same value, but I
>> don't! Instead, I get 10 USD as the value for the position, but the
>> converted cost of the position is now *8.46 EUR*, and even though it has
>> a price of 0.93 EUR attached to it.
>>
>>
>> This is because the  beanquery *convert *function is not using the
>> implicit convertion rate of the transaction (10 USD @@ 9.32 EUR), but the
>> price, derived from the latest price directive, which you have staying
>> elsewhere.
>> It kind of makes sense, because  when converting from one currency to
>> another you want to use an official rate, not the one used by a
>> specific merchant for a specific purchase.
>> you can use the implicit_prices
>> <https://github.com/beancount/beancount/blob/master/beancount/plugins/implicit_prices.py>plugin
>> though.
>>
>> """This plugin synthesizes Price directives for all Postings with a price
>> or
>> directive or if it is an augmenting posting, has a cost directive.
>> """
>> Even with the   implicit_prices
>> <https://github.com/beancount/beancount/blob/master/beancount/plugins/implicit_prices.py>plugin
>>  however,
>> you still can get a situation, that on a specific date you may have
>> different prices, but the convert function will use only one of it per day.
>>
>> To hunt such problems, there is  a unique_prices.py
>> <https://github.com/beancount/beancount/blob/master/beancount/plugins/unique_prices.py>
>> plugin
>>
>> In addition all of them are combined in the pedantic.py
>> <https://github.com/beancount/beancount/blob/master/beancount/plugins/pedantic.py>
>>  plugin
>>
>>
>>
>>
>> I tried to play with all kinds of parameters in the query, like with date
>> or without date, etc, but I still get 8.46 EUR for that transaction as
>> converted amount, hence the converted balance is also incorrect.
>>
>> I don't mind recording the transaction as in the first example, but what
>> I'm worried about is that the reports might turn out to be incorrect. I
>> randomly caught this transaction, but who knows how many there are/will be.
>>
>> Am I misunderstanding something?
>>
>> --
> 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 [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/beancount/32d17999-95e8-4613-ac33-370fc20dc41fn%40googlegroups.com
> <https://groups.google.com/d/msgid/beancount/32d17999-95e8-4613-ac33-370fc20dc41fn%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>

-- 
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 [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/beancount/CAK21%2BhPhLZbo42RefzZ7PSTs1uqHT7yocbU4SGRu6r618RCHRA%40mail.gmail.com.

Reply via email to