Filter SpaceAfter from MISC column in conllu2korapxml
Change-Id: I8af47a6d1a7661d07893a46b1bb0056bd099a5a6
diff --git a/script/conllu2korapxml b/script/conllu2korapxml
index 32cfbcf..58f443c 100755
--- a/script/conllu2korapxml
+++ b/script/conllu2korapxml
@@ -301,6 +301,11 @@
}
$morpho .= qq( <f name="lemma">$parsed[2]</f>\n) if($parsed[2] ne "_" || $parsed[1] eq '_');
$morpho .= qq( <f name="msd">$parsed[5]</f>\n) if($parsed[5] ne "_");
+ # Filter SpaceAfter from MISC column
+ if ($parsed[9] ne "_") {
+ my @misc_parts = grep { !/^Spaces?After/ } split(/\|/, $parsed[9]);
+ $parsed[9] = @misc_parts ? join('|', @misc_parts) : '_';
+ }
if($parsed[9] ne "_") {
if ($parsed[9] =~ /[0-9.e]+/) {
$morpho .= qq( <f name="certainty">$parsed[9]</f>\n)