If importing from PersonalChecking-*.csv, the default account should be “Assets:Checking:Personal”, and the output file should be “$LEDGERDIR/Personal/PersonalChecking.dat”.
How would that work with the default account that can be specified in the Ledger file itself?
I just noticed my bank sorted my statements backward, and of course CSV2L outputs them in that order. Sorting though is a PITA, given that I’ll have to buffer the output (not input, format can vary). I would prefer to sort on the date field in the output once it is normalized.
Maybe I’ll write out to a big hash, one node per date with an array for strings of output. Then I can just sort by keys and output the arrays.
This would impact the (nasty) way I do duplicate detection now…
I can grep a filehandle…
Perhaps the best way to handle this atm is to expect a sorted input, it wasn’t a big deal to sort my CSVs prior to importing.
Allow account matches to specify clear flag.