From: David Lecompte <david.lecompte@huawei.com>
Date: Wed, 3 Jun 2026 16:01:31 +0000 (+0100)
Subject: Remove CR and EN-Dash
X-Git-Url: https://nuage.metani.fr/gitweb/3GPP.git/commitdiff_plain/d788c1f2b3c65a888d59369fdd121ac20da1660b?hp=551b07f4893ca57b9d820c62bbdd2ad3b3660c5c

Remove CR and EN-Dash
---

diff --git a/bin/docx2asn.sh b/bin/docx2asn.sh
index 77d3b35..373abb9 100755
--- a/bin/docx2asn.sh
+++ b/bin/docx2asn.sh
@@ -47,6 +47,13 @@ while [ "$#" -gt 0 ]; do
 	    shift
 	    continue
 	fi
+
+	# On windows, soffice adds CR characters at line ends and produces EN-DASH in Windows-1252 enconding, i.e. 0x96
+	# Remove the CR at each line end and convert EN-DASH to normal minus
+	echo "${1%.docx}.txt"
+	LC_ALL=C sed -i $'s/\x96/-/g' "${1%.docx}.txt"
+	LC_ALL=C sed -i $'s/\r$//' "${1%.docx}.txt"
+
     fi
 
     if [[ -e "${1%.docx}.asn" ]] ; then
@@ -55,7 +62,7 @@ while [ "$#" -gt 0 ]; do
 	continue
     fi
 
-    sed -i 's/\r$//' "${1%.docx}.txt"
+       
     cat "${1%.docx}.txt" | awk '/^-- ASN1START$/,/^-- ASN1STOP$/' | grep -Evxe '-- ASN1ST(ART|OP)' | grep -v '^--' | awk 'NF || p; { p = NF }' p=0 > "${1%.docx}.asn"
 
     if [[ -e "${1%.docx}.asn" ]] ; then