script na vytažení všech použitých tagů pasáží

honzaflash · Oct 7, 2021 · 1732064 · 1732064
1 parent 285101b
commit 1732064
Showing 1 changed file with 32 additions and 0 deletions.
diff --git a/perl_scripts/gather-tags.pl b/perl_scripts/gather-tags.pl
@@ -0,0 +1,32 @@
+#!/usr/bin/perl
+
+my $usage =
+    "usage: ./gather-tags.pl TWINE_HTML [-l]\n" .
+    "    prints all existing passage tags\n";
+
+
+if (not defined $ARGV[0]) {
+    print $usage;
+    exit 1;
+}
+
+open(HTML, '<', $ARGV[0]) or die "couldn't open the file: $ARGV[0]";
+
+my %tags;
+while (<HTML>) {
+    if ($_ =~ /<tw-passagedata [^>]* tags="([^"]*)"/) {
+        for my $word (split(/\s+/, $1)) {
+            $tags{$word} = 1;
+        }
+    }
+}
+
+for (keys %tags) {
+    if ($ARGV[1] eq "-l") {
+        print "$_\n";
+    } else {
+        print "$_, ";
+    }
+}
+
+