[GNC] import csv transactions with split

David Solet dsolet at gmail.com
Mon Sep 20 12:27:25 EDT 2021


David, thank you for your response. That is not the problem, since there is
a screen where one links the name of the account in the csv file--whatever
it is--with the actual account in gnucash. However, I found the problem; it
is with the content of the csv file. In my csv file, a value for
customer_id (linked to gnucash's Description field) is on every line.
Description needs to be on the first line only. Also, in my file, a value
for descr (linked to gnucash's Memo field) is on every line. It needs to be
on every line *except* the first line. With those changes, the import
worked exactly as I wanted it to. I'm leaving this response in case this
comes up for another user.
David

On Sun, Sep 19, 2021 at 9:41 PM David H <hellvee at gmail.com> wrote:

> I don't use the csv importer but it seems to be complaining about not
> having a transfer account - "New, UNBALANCED (need acct to transfer
> $($$$.cc))!  While you do have a text description in the account column
> when I do a csv import in gnucash format the account names are in the full
> format eg "Expenses:Credit Card:Fees".  Does that help ?
>
> Cheers David H.
>
> On Mon, 20 Sept 2021 at 13:14, David Solet <dsolet at gmail.com> wrote:
>
>> Hello,
>> My version = 4.6 (Windows). I want to import a csv file with split
>> transactions. The following text lines from the csv file represent one
>> transaction (not real) that took place on 9/19/2021. This is a simplified
>> example. An actual transaction would likely have more splits. Column heads
>> are in the first line. The first line starts with a "," because the first
>> column head is blank--that is not a typo.
>> ,acct,customer_id,amt,descr,depdate
>> 1,a1net,a1custtl,155.28,cus_JwZTuBkST5GCko_7/29/2021,9/19/2021
>> 2,dues,cus_GtJPUDZyyQ,-40,cus_GtJPUDZyyQ_8/14/2021,
>> 3,dues,cus_IuTPxTysqQ,-40,cus_IuTPxTysqQ_8/1/2021,
>> 4,dues,cus_JwZTuBkST5,-40,cus_JwZTuBkST5_7/29/2021,
>> 5,dues,cus_K2IckcwXuD,-40,cus_K2IckcwXuD_8/13/2021,
>> 6,fees,cus_GtJPUDZyyQ,1.18,cus_GtJPUDZyyQ_8/14/2021,
>> 7,fees,cus_IuTPxTysqQ,1.18,cus_IuTPxTysqQ_8/1/2021,
>> 8,fees,cus_JwZTuBkST5,1.18,cus_JwZTuBkST5_7/29/2021,
>> 9,fees,cus_K2IckcwXuD,1.18,cus_K2IckcwXuD_8/13/2021,
>>
>> After examining previously exported csv transactions with splits from my
>> gnucash file, I assumed date needed to be in the first line only. I
>> attempted to import this as one transaction with multiple splits. I
>> checked
>> the multi-split box and identified:
>> acct = account
>> depdate = date
>> amt = deposit
>> customer_id = description
>> descr = memo
>>
>> After choosing next, I linked the account field with existing accounts.
>> The
>> account "a1net" is the bank. "dues" is membership dues (type = income).
>> "fees" is credit card fees  (type = expense). However, at the Match
>> Transactions screen, my choices did not complete as expected. All of the
>> splits showed as "New, UNBALANCED...". The screen looked like this:
>>
>> I will point out that there are three separate splits that are linked to
>> "Membership Fees" and three linked to "Credit card fees PRS" in this one
>> transaction. a1net is the net bank deposit. I want to preserve that level
>> of detail for documentation, and that kind of detail can be preserved when
>> transactions like this are manually entered. Is it possible to do this
>> with
>> a csv import? Thank you for your help.
>> David
>> _
>>
>


More information about the gnucash-user mailing list