===================================================================
RCS file: /cvs/pta/pta_import.pl,v
retrieving revision 1.12
retrieving revision 1.29
diff -u -p -r1.12 -r1.29
--- pta/pta_import.pl	2020/11/17 15:31:02	1.12
+++ pta/pta_import.pl	2021/05/19 17:10:53	1.29
@@ -1,4 +1,4 @@
-#!/usr/bin/perl 
+#!/usr/bin/perl
 
 # Copyright (c) 2020 Freda Bundchen
 
@@ -21,147 +21,15 @@ use Getopt::Std qw(getopts);
 
 our ($opt_I);
 
+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  =====================================================
 
-sub import_account ($$$$) {
-	my ($account_name, $compiled_ref, $delim, $csv_account) = @_;
-	my ($account, $amount, $booking, $credit, $date, $debit,
-	    $description);
-	chomp(my $header = <>);
-	while (<>) {
-		next if (/^$/);
-		chomp;
-		my $line = $_;
-		my @fields = split /$delim/, $line;
-		my $matches = 0;
-		foreach my $selector (@$compiled_ref) {
-			$matches = 1;
-			for (my $i = 0; $i <= $#{$selector->{re}}; $i++) {
-				next if $fields[$i] =~ $selector->{re}[$i];
-				$matches = 0;
-				last;
-			}
-			if ($matches) {
-				$account = $selector->{ac};
-				$booking = $selector->{id};
-				last;
-			}
-		}
-		$matches or die "unmatched CSV line: $line";
-		if ($account_name eq "chase_credit") {
-			($date, $amount, $debit, $credit,
-			    $description) =
-			    import_chase_credit($account,
-			    $csv_account, $line);
-		} elsif ($account_name eq "capital_one_credit") {
-			($date, $amount, $debit, $credit,
-			    $description) =
-			    import_capital_one_credit($account,
-			    $csv_account, $line);
-		} elsif ($account_name eq "optum_hsa") {
-			($date, $amount, $debit, $credit,
-			    $description) =
-			    import_optum_hsa($account,
-			    $csv_account, $line);
-		} elsif ($account_name eq "sparkasse_camt") {
-			($date, $amount, $debit, $credit,
-			    $description) =
-			    import_sparkasse_camt($account,
-			    $csv_account, $line);
-		} else {
-			die "undefined format: $account_name";
-		}
-		unless ($date && $booking && $debit && $credit &&
-		    $amount && $description) {
-			die "import parse error: $line";
-		}
-		$description =~ s/#//g;
-		print "$date $booking $debit $credit $amount " .
-		    "$description\n";
-	}
-}
-
-sub import_chase_credit ($$$) {
-	my ($account, $csv_account, $line) = @_;
-	my ($trans_date, $post_date, $description, $category,
-	    $type, $amount) = split /,/, $line;
-	my ($debit, $credit);
-	$post_date =~ s#(\d+)/(\d+)/(\d+)#$3$1$2#; 
-	($amount, $debit, $credit) =
-	    get_accounts_by_amount_sign($amount, $account,
-	    $csv_account);
-	return ($post_date, $amount, $debit, $credit, $description);
-}
-
-sub import_capital_one_credit ($$$) {
-	my ($account, $csv_account, $line) = @_;
-	my ($trans_date, $post_date, $card_num,
-	    $description, $category, $csv_debit,
-	    $csv_credit) = split /,/, $line;
-	$post_date =~ s/(\d+)-(\d+)-(\d+)/$1$2$3/; 
-	my ($amount, $debit, $credit) = 
-	    get_accounts_by_csv_col($account, $csv_account,
-	    $csv_debit, $csv_credit);
-	return ($post_date, $amount, $debit, $credit, $description);
-}
-
-sub import_optum_hsa ($$$) {
-        my ($account, $csv_account, $line) = @_;
-        my ($date, $description, $amount,
-            $type) = split /,/, $line;
-	my ($debit, $credit);
-	$date =~ s/(\d+)-(\d+)-(\d+)/$1$2$3/;
-	$amount =~ s/\$//;
-	($amount, $debit, $credit) =
-	    get_accounts_by_amount_sign($amount, $account,
-	    $csv_account);
-	return ($date, $amount, $debit, $credit, $description);
-}
-
-sub import_sparkasse_camt ($$$) {
-        my ($account, $csv_account, $line) = @_;
-	my @fields;
-	$_ = $line;
-	push @fields, $1 while s/"([^"]*)";?//;
-	$_ eq "" or die "CAMT parse error before $_ in $line";
-	@fields == 17 or die "not 17 but @fields fields: $line";
-	$fields[1] =~ s/^(\d\d)\.(\d\d)\.(\d\d)$/20$3$2$1/
-	    or die "date parse error: $line";
-	$fields[14] =~ s/,/./;
-	return $fields[1],
-	    get_accounts_by_amount_sign($fields[14], $account, $csv_account),
-	    (join ' ', $fields[11], $fields[4]);
-}
-
-sub get_accounts_by_amount_sign ($$$) {
-	my ($amount, $account, $csv_account) = @_;
-	my ($debit, $credit);
-	if ($amount <= 0) {
-		$amount = substr $amount, 1;
-		$credit = $csv_account;	
-		$debit = $account;
-	} else {
-		$debit = $csv_account;	
-		$credit = $account;
-	}
-	return ($amount, $debit, $credit);
-}
-
-sub get_accounts_by_csv_col ($$$$) {
-	my ($account, $csv_account, $csv_debit, $csv_credit) = @_;
-	my ($amount, $debit, $credit);
-	if ($csv_debit eq "") {
-		$amount = $csv_credit;
-		$credit = $account;
-		$debit = $csv_account;
-	} else {
-		$amount = $csv_debit;
-		$credit = $csv_account;
-		$debit = $account;
-	}
-	return ($amount, $debit, $credit);
-}
-
 sub usage () {
 	printf STDERR "usage: %s -I accountname csvfilename\n", $0;
 	exit 1;
@@ -169,29 +37,89 @@ sub usage () {
 
 # === MAIN PROGRAM =====================================================
 
-my ($csv_account, $fn, $in, $account_name, $delim, @compiled);
 getopts 'I:' or usage;
-if ($opt_I) {
-	$account_name = $opt_I;
-	$fn = "import_" . $account_name . ".txt";
-	open $in, $fn or die "$fn: $!";
-} else {
+unless ($opt_I) {
+	warn "The option -I is required.";
 	usage;
 }
+my $account_name = $opt_I;
+
+# 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_regex,
+    $delim, $header, $quantity_field, $quote, @amount_fields,
+    @compiled, @description_fields, @ignored);
 while (<$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 = $1;
 		next;
 	}
-	if (/^DELIM\s+(\S)$/) {
+	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 (s/^COSTCENTER\s+//) {
+		$cost_center_field and
+		    die "duplicate COSTCENTER line: $_";
+		/^(\d+)$/ or die "COSTCENTER parse error: $_";
+		$cost_center_field = $1 - 1;
+		next;
+	}
+	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 (s/^DELIM\s+//) {
 		$delim and die "duplicate DELIM line: $1";
+		/^([^|\^\$\*\+\?\(\)\[\]\{\}\\])$/ or
+		    die "DELIM parse error: $_";
 		$delim = $1;
 		next;
 	}
+	if (s/^DESCRIPTION\s+//) {
+		@description_fields and die "duplicate DESCRIPTION line: $_";
+		push @description_fields, $1 - 1 while s/(\d+)\s*//;
+		$_ eq '' or die "trailing garbage: DESCRIPTION ... $_";
+		next;
+	}
+	if (s/^HEADER\s+//) {
+		$header and die "duplicate HEADER line: $1";
+		$header = $_;
+		next;
+	}
+	if (s/^IGNORE\s+//) {
+		push @ignored, qr/$_/;
+		next;
+	}
+	if (s/^QUANTITY\s+//) {
+		$quantity_field and die "duplicate QUANTITY line: $1";
+		/^(\d+)$/ or die "QUANTITY parse error: $_";
+		$quantity_field = $1 - 1;
+		next;
+	}
+	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
 	    die "$fn import parse error: $line";
@@ -203,4 +131,74 @@ while (<$in>) {
 }
 close $in;
 $csv_account or die "no ACCOUNT line in $fn";
-import_account($account_name, \@compiled, $delim, $csv_account);
+
+# Parse the CSV file from the bank.
+if (@ARGV) {
+	open STDIN, '<', $ARGV[0] or die "$ARGV[0]: $!";
+}
+LINE: while (<STDIN>) {
+	chomp;
+	s/\s*$//;
+	next if (/^$/);
+	foreach my $ignore (@ignored) {
+		next LINE if /$ignore/;
+	}
+	my $line = $_;
+	if ($header) {
+		$line eq $header
+		    or die "expected HEADER $header\nbut got $line";
+		undef $header;
+		next;
+	}
+	my $copy_line = $line;
+	my @fields;
+	if ($quote) {
+		push @fields, $1 while $copy_line =~ s/$quote([^$quote]*)$quote$delim?//;
+	} else {
+		@fields = split /$delim/, $line;
+	}
+	my $matches = 0;
+	my ($account, $booking);
+	foreach my $selector (@compiled) {
+		$matches = 1;
+		for (my $i = 0; $i <= $#{$selector->{re}}; $i++) {
+			next if $fields[$i] =~ $selector->{re}[$i];
+			$matches = 0;
+			last;
+		}
+		if ($matches) {
+			$account = $selector->{ac};
+			$booking = $selector->{id};
+			last;
+		}
+	}
+	$matches or die "unmatched CSV line: $line";
+	my $date = $fields[$date_field] or
+	    die "date parse error: $line";
+	eval '$date =~ ' . $date_regex;
+	foreach my $i (@amount_fields) {
+		if (defined($fields[$i])) {
+			$fields[$i] =~ s/,/./;
+			$fields[$i] =~ s/\$//;
+			$fields[$i] = "-$1" if $fields[$i] =~ /^\((\d+\.\d+)\)/;
+		}
+	}
+	my $debit = $csv_account;
+	my $credit = $account;
+	my $amount = $fields[$amount_fields[-1]] || -$fields[$amount_fields[0]];
+	if ($amount < 0) {
+		$amount *= -1;
+		$credit = $csv_account;
+		$debit = $account;
+	}
+	$amount = sprintf "%.2f", $amount;
+	my $description = join ' ', @fields[@description_fields];
+	$date && $amount && $debit && $credit && $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 $quantity_field;
+	print "$date $booking $debit $credit $amount $description\n";
+}