From 33ce867b1bd4abb8fc9dde049507937182b9c6e3 Mon Sep 17 00:00:00 2001
From: Daniel Friesel <derf@finalrewind.org>
Date: Mon, 18 Nov 2019 20:09:07 +0100
Subject: csv2json, json2json: Perform consistency checks

---
 README.md       |  6 +++---
 share/csv2json  |  4 ++--
 share/json2json | 30 ++++++++++++++++++++++++++++++
 3 files changed, 35 insertions(+), 5 deletions(-)

diff --git a/README.md b/README.md
index d53b36b..42cd267 100644
--- a/README.md
+++ b/README.md
@@ -73,9 +73,9 @@ located in `share/stations.json`. There are two recommended editing methods.
 Automatic method, e.g. to incorporate changes from Open Data sources:
 
 * modify stations.json with a script in any JSON-aware language you like
-* run ./json2json in the share diretcory. This transforms stations.json into
-  its canonical format, which simplifies tracking of changes and reduces diff
-  size
+* run `./json2json` in the share diretcory. This performs consistency checks and
+  transforms stations.json into its canonical format, which simplifies tracking
+  of changes and reduces diff size
 
 Manual method:
 
diff --git a/share/csv2json b/share/csv2json
index 9e6340c..2bac49f 100755
--- a/share/csv2json
+++ b/share/csv2json
@@ -50,8 +50,8 @@ for my $line (@csv_lines) {
 
 my $have_duplicates = 0;
 my @names           = map { $_->{name} } @stations;
-my @ds100           = map { $_->{ds100} } @stations;
-my @uic_ids         = map { $_->{uic} } @stations;
+my @ds100           = map { $_->{ds100} } sort { $a->{ds100} cmp $b->{ds100} } @stations;
+my @uic_ids         = map { $_->{uic} } sort { $a->{uic} <=> $b->{uic} } @stations;
 
 for my $i ( 1 .. $#names ) {
 	if ( $names[ $i - 1 ] eq $names[$i] ) {
diff --git a/share/json2json b/share/json2json
index 5cad10f..189ee2c 100755
--- a/share/json2json
+++ b/share/json2json
@@ -10,5 +10,35 @@ use JSON;
 my $json_str = read_file('stations.json');
 my $stations = JSON->new->utf8->decode($json_str);
 @{$stations} = sort { $a->{name} cmp $b->{name} } @{$stations};
+
+my $have_duplicates = 0;
+my @names           = map { $_->{name} } @{$stations};
+my @ds100           = map { $_->{ds100} } sort { $a->{ds100} cmp $b->{ds100} } @{$stations};
+my @uic_ids         = map { $_->{uic} } sort { $a->{uic} <=> $b->{uic} } @{$stations};
+
+for my $i ( 1 .. $#names ) {
+	if ( $names[ $i - 1 ] eq $names[$i] ) {
+		say "Duplicate station name: $names[$i]";
+		$have_duplicates = 1;
+	}
+}
+for my $i ( 1 .. $#ds100 ) {
+	if ( $ds100[ $i - 1 ] eq $ds100[$i] ) {
+		say "Duplicate DS100 code: $ds100[$i]";
+		$have_duplicates = 1;
+	}
+}
+for my $i ( 1 .. $#uic_ids ) {
+	if ( $uic_ids[ $i - 1 ] == $uic_ids[$i] ) {
+		say "Duplicate UIC ID: $uic_ids[$i]";
+		$have_duplicates = 1;
+	}
+}
+
+if ($have_duplicates) {
+	say "Thank you for your contribution.";
+	say "Please remove duplicate entries before opening a pull request.";
+}
+
 my $json_out = JSON->new->utf8->canonical->pretty->encode($stations);
 write_file( 'stations.json', $json_out );
-- 
cgit v1.2.3