Skripty na chroustání Rejstříku škol umí aktuální formát

db/skoly/parse-all

 #!/bin/bash
 set -e
 
-rm -rf extra/parsed
-mkdir extra/parsed
+rm -rf extra/skoly/parsed
+mkdir extra/skoly/parsed
 
-for src in extra/html/*.html ; do
-	dst=extra/parsed/$(basename $src .html).tsv
+for src in extra/skoly/html/*.html ; do
+	dst=extra/skoly/parsed/$(basename $src .html).tsv
 	echo -n "$src -> "
-	./rejskol-parse <$src >$dst
+	db/skoly/rejskol-parse <$src >$dst
 	wc -l $dst
 done

db/skoly/rejskol-download

@@ -8,7 +8,7 @@ my $mech = WWW::Mechanize->new(autocheck => 1, strict_forms => 1);
 $mech->get('https://rejstriky.msmt.cz/rejskol/VREJVerejne/VerejneRozhrani.aspx');
 $mech->form_id('form1');
 
-mkdir 'extra/html';
+mkdir 'extra/skoly/html';
 download_type('B');	# Základní školy
 download_type('C');	# Střední školy
 exit 0;
@@ -59,7 +59,7 @@ sub download_region {
 	sleep 1;
 	my $resp = $mech->click_button(id => 'btnVybrat');
 
-	open my $f, '>:utf8', "extra/html/$type-$nuts.html";
+	open my $f, '>:utf8', "extra/skoly/html/$type-$nuts.html";
 	print $f $resp->decoded_content;
 	close $f;
 

db/skoly/rejskol-parse

@@ -18,10 +18,10 @@ my $div_with_tables = $divs[1] // die;
 my @last_main = ();
 my $tab_count = 0;
 for my $table ($div_with_tables->find_by_tag_name('table')) {
-	my $style = $table->attr('style');
-	my $main = 0;
-	if ($tab_count == 0 || defined $style) {
-		$main = 1;
+	my $style = $table->attr('style') // "";
+	my $main = 1;
+	if ($tab_count == 1 || $style =~ m{#f0f8ff}) {
+		$main = 0;
 	}
 	my $last_out = "";
 	for my $tr ($table->find_by_tag_name('tr')) {