=================================================================== RCS file: /cvs/pta/pta_import.pl,v retrieving revision 1.26 retrieving revision 1.32 diff -u -p -r1.26 -r1.32 --- pta/pta_import.pl 2020/11/30 14:38:35 1.26 +++ pta/pta_import.pl 2021/06/04 13:50:06 1.32 @@ -21,15 +21,11 @@ use Getopt::Std qw(getopts); our ($opt_I); -my %banks = ( - "bbva_usa" => \&import_bbva_usa, - "capital_one_360" => \&import_capital_one_360, - "capital_one_credit" => \&import_capital_one_credit, - "chase_credit" => \&import_chase_credit, - "etrade_ira" => \&import_etrade_ira, - "optum_hsa" => \&import_optum_hsa, - "sparkasse_camt" => \&import_sparkasse_camt, - "wellsfargo" => \&import_wellsfargo, +my %date_formats = ( + 'MM/DD/YYYY' => 's#(\d+)/(\d+)/(\d+)#$3$1$2#', + 'MM/DD/YY' => 's#(\d+)/(\d+)/(\d+)#20$3$1$2#', + 'YYYY-MM-DD' => 's#(\d+)-(\d+)-(\d+)#$1$2$3#', + 'DD.MM.YY' => 's#(\d+)\.(\d+)\.(\d+)#20$3$2$1#', ); # === SUBROUTINES ===================================================== @@ -47,72 +43,91 @@ unless ($opt_I) { usage; } my $account_name = $opt_I; -$banks{$account_name} or die "unknown accountname: $account_name"; # Parse the configuration file. my $fn = "import/" . $account_name . ".txt"; open my $in, '<', $fn or die "$fn: $!"; -my ($cost_center_field, $csv_account, $date_field, $date_format, - $delim, $header, $quantity_field, $quote, @amount_fields, - @compiled, @description_fields, @ignored); +my ($cost_center_field, $csv_account, $date_field, $date_regex, + $delim, $header, $quantity_field, $quote, $subaccount, + @amount_fields, @compiled, @description_fields, @ignored); while (<$in>) { chomp; + s/\s+$//; next if /^(?:#|$)/; my $line = $_; - if (/^ACCOUNT\s+(\S+)$/) { - $csv_account and die "duplicate ACCOUNT line: $1"; + if (s/^ACCOUNT\s+//) { + $csv_account and die "duplicate ACCOUNT line: $_"; + /^(\d+)$/ or die "ACCOUNT parse error: $_"; $csv_account = $1; next; } if (s/^AMOUNT\s+//) { + @amount_fields and die "duplicate AMOUNT line: $_"; push @amount_fields, $1 - 1 while s/(\d+)\s*//; $_ eq '' or die "trailing garbage: AMOUNT ... $_"; next; } - if (/^COSTCENTER\s+(\d+)/) { + if (s/^COSTCENTER\s+//) { + $cost_center_field and + die "duplicate COSTCENTER line: $_"; + /^(\d+)$/ or die "COSTCENTER parse error: $_"; $cost_center_field = $1 - 1; next; } - if (/^DATE\s+(\d+)\s+(\S+)/) { - $date_field = $1 - 1; - $date_format = $2; + if (s/^DATE\s+//) { + $date_field || $date_regex and + die "duplicate DATE line: $_"; + s/^(\d+)\s+(\S+)$//; + $date_field = $1 or + die "DATE date_field parse error: $_"; + $date_field -= 1; + $date_regex = $date_formats{$2} + or die "unknown date format: $2"; next; } - if (/^DELIM\s+(\S)$/) { + if (s/^DELIM\s+//) { $delim and die "duplicate DELIM line: $1"; + /^([^|\^\$\*\+\?\(\)\[\]\{\}\\])$/ or + die "DELIM parse error: $_"; $delim = $1; next; } if (s/^DESCRIPTION\s+//) { - push @description_fields, $1 - 1 while s/(\d+)\s*//; - $_ eq '' or die "trailing garbage: DESCRIPTION ... $_"; - next; + @description_fields and die "duplicate DESCRIPTION line: $_"; + push @description_fields, $1 - 1 while s/(\d+)\s*//; + $_ eq '' or die "trailing garbage: DESCRIPTION ... $_"; + next; } - if (/^HEADER\s+(.*)$/) { + if (s/^HEADER\s+//) { $header and die "duplicate HEADER line: $1"; - $header = $1; + $header = $_; next; } - if (/^IGNORE\s+(.*)/) { - push @ignored, qr/$1/; + if (s/^IGNORE\s+//) { + push @ignored, qr/$_/; next; } - if (/^QUANTITY\s+(\d+)/) { + if (s/^QUANTITY\s+//) { + $quantity_field and die "duplicate QUANTITY line: $1"; + /^(\d+)$/ or die "QUANTITY parse error: $_"; $quantity_field = $1 - 1; next; } - if (/^QUOTE\s+(\S)$/) { + if (s/^QUOTE\s+//) { $quote and die "duplicate QUOTE line: $1"; + /^([^|\^\$\*\+\?\(\)\[\]\{\}\\])$/ or + die "QUOTE parse error: $_"; $quote = $1; next; } $delim or die "no DELIM line in $fn"; - s/^(.*)$delim\s+(\d+)\s+(\S+)// or + s/^(.*)$delim\s+(\d+)\s+(\S+)\s*(\S*)// or die "$fn import parse error: $line"; push @compiled, { re => [map { qr/$_/ } split /$delim/, $1], ac => $2, id => $3, + su => $4, }; } close $in; @@ -124,6 +139,7 @@ if (@ARGV) { } LINE: while () { chomp; + s/\s*$//; next if (/^$/); foreach my $ignore (@ignored) { next LINE if /$ignore/; @@ -137,10 +153,10 @@ LINE: while () { } my $copy_line = $line; my @fields; - if ($quote) { - push @fields, $1 while $copy_line =~ s/$quote([^$quote]*)$quote$delim?//; - } else { - @fields = split /$delim/, $line; + while ($copy_line =~ /./) { + $quote && $copy_line =~ s/^$quote(.*?)$quote(?:$delim|$)// + or $copy_line =~ s/^(.*?)(?:$delim|$)//; + push @fields, $1; } my $matches = 0; my ($account, $booking); @@ -154,23 +170,14 @@ LINE: while () { if ($matches) { $account = $selector->{ac}; $booking = $selector->{id}; + $subaccount = $selector->{su}; last; } } $matches or die "unmatched CSV line: $line"; my $date = $fields[$date_field] or die "date parse error: $line"; - if ($date_format eq "MM/DD/YYYY") { - $date =~ s#(\d+)/(\d+)/(\d+)#$3$1$2#; - } elsif ($date_format eq "MM/DD/YY") { - $date =~ s#(\d+)/(\d+)/(\d+)#20$3$1$2#; - } elsif ($date_format eq "YYYY-MM-DD") { - $date =~ s#(\d+)-(\d+)-(\d+)#$1$2$3#; - } elsif ($date_format eq "DD.MM.YY") { - $date =~ s#(\d+)\.(\d+)\.(\d+)#20$3$2$1#; - } else { - die "unhandled date format: $line"; - } + eval '$date =~ ' . $date_regex; foreach my $i (@amount_fields) { if (defined($fields[$i])) { $fields[$i] =~ s/,/./; @@ -188,12 +195,16 @@ LINE: while () { } $amount = sprintf "%.2f", $amount; my $description = join ' ', @fields[@description_fields]; - $date && $amount && $debit && $credit && $description + defined($date) && defined($amount) && defined($debit) && + defined($credit) && defined($description) or die "import parse error: $line"; $description =~ s/#//g; $description .= " [$fields[$cost_center_field]]" if $cost_center_field; $description .= " quantity $fields[$quantity_field]" if $quantity_field; + $description .= " ($subaccount)" + if $subaccount; + print "$date $booking $debit $credit $amount $description\n"; }