force unicode/utf8 for html2text

ChangeLog

@@ -1,6 +1,7 @@
 ****************************************************************************
 # ChangeLog for lesspipe.sh						   #
 ****************************************************************************
+- force unicode/utf8 for html2text
 Version 2.14 Aug 16 2024
 - allow for colorizing dotfiles
 - add nvimpager as (preferred) colorizer 

README.md

@@ -456,7 +456,7 @@ STDOUT and the commands executed to STDERR.
 - sxw2txt              https://vinc17.net/software/sxw2txt (2010)
 - dtc                  https://git.kernel.org/cgit/utils/dtc/dtc.git (2023)
 - xmq                  https://github.com/libxmq/xmq/releases/latest (2024)
-- nvimpager            https://github.com/lucc/nvimpager (2025)
+- nvimpager            https://github.com/lucc/nvimpager (2024)
 
 ### 12.2 References
 - [1] http://www.greenwoodsoftware.com/less/	(less)

lesspipe.sh

@@ -835,13 +835,15 @@ handle_w3m () {
 
 ishtml () {
 	[[ $1 == - ]] && arg1=-stdin || arg1="$1"
+	htmlopt=--unicode-snob
+	has_cmd html2text && echo ""|html2text -utf8 2>/dev/null && htmlopt=-utf8
 	# 3 lines following can easily be reshuffled according to the preferred tool
 	has_cmd elinks && nodash "elinks -dump -force-html" "$1" && return ||
 	has_cmd w3m && handle_w3m "$1" && return ||
 	has_cmd lynx && lynx -force_html -dump "$arg1" && return ||
-	# different versions of html2text existing, therefore no encoding handling
+	# different versions of html2text existingi, force unicode
 	[[ "$1" == https://* ]] && return ||
-	has_cmd html2text && nodash html2text "$1"
+	has_cmd html2text && nodash html2text "$htmlopt" "$1"
 }
 
 # the main program