From fde36ecccc9d90a0050f8dd3145d4125c4b65600 Mon Sep 17 00:00:00 2001
From: Jean-Francois Dockes <jfd@recoll.org>
Date: Thu, 26 Jan 2012 19:30:43 +0100
Subject: [PATCH] Handle garbled unrtf http-equiv header causing pbs with html5
 handler

---
 src/filters/rclrtf      | 4 +++-
 src/sampleconf/mimeconf | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/src/filters/rclrtf b/src/filters/rclrtf
index 24abafe4..abfe6e2f 100755
--- a/src/filters/rclrtf
+++ b/src/filters/rclrtf
@@ -83,6 +83,7 @@ checkcmds awk unrtf
 # output the result
 # The strange 'BEGIN' setup is to prevent 'file' from thinking this file
 # is an awk program
+# The thing about the charset is that unrtf outputs a garbled one.
 unrtf --nopict --html "$infile" 2> /dev/null |
 awk 'BEGIN'\
 ' {
@@ -90,10 +91,11 @@ awk 'BEGIN'\
 }
 /<\/head>/{
     if (gothead == 0) {
-        printf("<meta http-equiv=\"Content-Type\" content=\"text/html; charset=iso-8859-1\">\n")
+        printf("<meta http-equiv=\"Content-Type\" content=\"text/html; charset=UTF-8\">\n")
 	gothead = 1
     }
 }
+/<meta http-equiv=/{next}
 {
   print
 }
diff --git a/src/sampleconf/mimeconf b/src/sampleconf/mimeconf
index f47b2b4b..e985b5b0 100644
--- a/src/sampleconf/mimeconf
+++ b/src/sampleconf/mimeconf
@@ -99,7 +99,7 @@ message/rfc822 = internal
 text/calendar = execm rclics;mimetype=text/plain;charset=utf-8
 text/html  = internal 
 text/plain = internal 
-text/rtf = exec unrtf --nopict --html;charset=iso-8859-1;mimetype=text/html
+text/rtf = exec unrtf --nopict --html;mimetype=text/html
 text/x-c = internal
 text/x-c++ = internal
 text/x-c+ = internal