From 181df7fbaeaa3ae070c724436c5a5f62fbcb9a32 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?P=C5=99emysl=20Eric=20Janouch?= Date: Thu, 4 Nov 2021 08:31:47 +0100 Subject: [PATCH] czech-wordnet.pl: fix to work on non-GNU systems --- dicts/czech-wordnet.pl | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/dicts/czech-wordnet.pl b/dicts/czech-wordnet.pl index d626e5a..d1d32f9 100755 --- a/dicts/czech-wordnet.pl +++ b/dicts/czech-wordnet.pl @@ -4,11 +4,14 @@ use warnings; use strict; +# GNU Gzip can unpack a ZIP file, but not the BSD one, and unzip can't use stdin +my $zipcat = qx/command -v bsdtar/ ? 'bsdtar -xOf-' : 'zcat'; + my $base = 'https://lindat.cz/repository/xmlui'; my $path = 'handle/11858/00-097C-0000-0001-4880-3'; open(my $doc, '-|', "curl -Lo- '$base/bitstream/$path/Czech_WordNet_1.9_PDT.zip'" - . ' | zcat | iconv -f latin2 -t UTF-8') or die $!; + . " | $zipcat | iconv -f latin2 -t UTF-8") or die $!; # https://nlp.fi.muni.cz/trac/deb2/wiki/WordNetFormat but not quite; # for terminology see https://wordnet.princeton.edu/documentation/wngloss7wn