From 55d0f53f7aa6298cde818cdb637d3d6f7b29ecf1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?P=C5=99emysl=20Eric=20Janouch?= Date: Thu, 7 Oct 2021 17:21:49 +0200 Subject: [PATCH] Abandon the idea of sts for slovnik-cizich-slov We'd need something like inlining the type sequence within the data, which doesn't seem particularly elegant. --- dicts/slovnik-cizich-slov.sh | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/dicts/slovnik-cizich-slov.sh b/dicts/slovnik-cizich-slov.sh index bc6bd9a..343b98f 100755 --- a/dicts/slovnik-cizich-slov.sh +++ b/dicts/slovnik-cizich-slov.sh @@ -1,7 +1,6 @@ #!/bin/sh -e # Slovník cizích slov, see https://slovnik-cizich-slov.abz.cz/web.php/o-slovniku -# TODO: Skipping the optional pronunciation field, tabfile can't handle it yet, -# but could be made to accept a lowercase sametypesequence +# XXX: skipping the /optional/ pronunciation field, tabfile can't handle that curl -Lo- https://slovnik-cizich-slov.abz.cz/export.php | \ iconv -f latin2 -t UTF-8 | perl -CSD -F\\\| -le ' print "$_\t" . $F[2] =~ s/\\/\\\\/gr =~ s/; /\\n/gr for split(", ", $F[0])