Fast data entry possible?

[Lincoln A Baxter]

On Sat, 2011-04-23 at 21:52 -0700, John Ralls wrote:
> On Apr 23, 2011, at 8:57 PM, Eleanor Olds Batchelder wrote:
> 
> > If I assign a number to each account, will I be able to just enter
> the number, rather than three or four account names?  (This data entry
> burden could be a deal-breaker…)
> 
> No. That's been asked for, but never implemented. If it's only a few
> accounts, you could give them names with unique initials and put them
> at the top level while you catch up, then move them back into their
> places in the hierarchy and give them back their right names when
> you're done.
> 

Actually, there is. Though, as John says, not out of the box. It
requires "a little programming."

Or along the same lines as above, you could just import a converted QIF
file (from the CVS) several scripts exist to do this, and then move the
transactions from the accounts created by the descriptions by deleting
the accounts and telling GC where to move them to.  That's I a brought
my dads quicken data in to CG... but he had been pretty consistent with
is categories.

Or (picking up on my opening comment) One can convert the spreadsheet
data to a QIF file, where the "category" represents the other
(balancing) account.

I recently wrote a perl script to do this, from a CSV file downloaded
from SECU where my wife banks.  I could not get SECU's AQ banking
interface working (it just froze on getting the account list), and they
did not have a QFX download option (which pissed me off enough to write
the script).

Basically the script uses the transaction description in the downloaded
CSV file, to select a target account (as a QIF category).

It would take a bit of work to reduce the included modules down to just
what is needed (minimize dependencies... I used a personal module
library to read the CSV file), package for distribution. But it you are
interested I would consider it.  It's not fancy. The designated target
balancing accounts (categories in QIF terms) are hard coded, based on
regular expression matches in the description column of the downloaded
CSV file. It could be generalized, to reading a table of expressions
with target accounts as a configuration file, if there were general
interest.

On thing I think would really help GC's flexibility would the ability to
fork any system command (essentially as a "user exit"), from within any
import dialog, the command could be used preprocess a selected file (or
post process exports). On the export side (reports) however, the need is
now for less, because of the SQL backend, though that release is not yet
part of Ubuntu 10.04.  

By doing this (input user exit processing), and not limiting the
contributor learning scheme or cracking the code of GC there might be a
much larger library of contributed software.

A simple way to do this might be to modify the import dialogs, giving
them "advanced" options for specifying command to execute, the input
file to be passed to the command, and the output file import.  Today, it
already selects the file to import.  But if this were to work, it would
have the allow the specification of a file that does not yet exist... or
it it could to execute the specified script, (as a button press perhaps)
before the file to import is selected. (That might actually be better...
and less change).  We have to work a bit on the specifications if there
is interest... for instance, as I read this over, I think there would
have to be a way to view the output of the command, if any were
produced, (for debugging purposes).

Any interest?

Lincoln



> Regards,
> John Ralls
> 
> _______________________________________________
> gnucash-user mailing list
> gnucash-user at gnucash.org
> https://lists.gnucash.org/mailman/listinfo/gnucash-user
> -----
> Please remember to CC this list on all your replies.
> You can do this by using Reply-To-List or Reply-All.
>