On Tue, 28 Apr 2009 08:24:34 -0400, Mike or Penny Novack <stepbystepf...@mtdata.com> wrote: > marcus.wolsc...@googlemail.com wrote: > You would first have to look at "what rule does the bank follow" with of > course no expectation that any two banks would be following the same > rule. That should be enough to convince you that nothing done at the > Gnucash end could guarantee the same order except the ability to define > a complex set of rules for order.
In HBCI it is guaranteed that processed orders keep their position in a day. New orders are appended to the day. That end is not the issue. I have found the ordering to be stable enough to help my users in compating days with many orders to find that one transaction that differs by enforcing the same ordering. As I am creating and modifying the gnucash-transactions on the XML-level I can alter them however I want. > When choosing a bank, an important consideration is whether for I am not choosing a bank here. I am asking a technical question about how gnucash orders it's transactions. The rest is my problem and I know very well how to solve it. > Keep in mind that for checks coming back to the bank, their arrival and I have never done a check in my live. Never seen one. This is europe. Welcome to the world of fast inter-bank transactions without any fees and online-banking with lists of onetime-passcodes (tan). > entry into the bank's records is random with respect to when written and > entered in GnuCash. I am not using the aqbanking. Did I say that I do? I have my own HBCI-code to do that and it does a "little" more then just import everything it gets as transactions of 2 splits. > The only thing that would be the same in both places > is the check number and the date written. There is no check-number. The greatest deficiency in HBCI is that there is no primary key. However I am adding an identifier that gnucash ignores to the imported transactions to clearly identify a transaction as belonging to a given HBCI-transaction when doing a synchronize again later. I can modify: * the timestamp stored int he date -field of gnucash-transactions * the order they have in the file * their uuid (easy on create, more difficult on updates) Gnucash is a nice tool to work with. I have done so for years and would not want to miss it. But for complex, automated tasks such as this I am using my own program configured with lots of plugins and scripts. Marcus _______________________________________________ gnucash-devel mailing list gnucash-devel@gnucash.org https://lists.gnucash.org/mailman/listinfo/gnucash-devel
