Decode HTML using PHP instead of recode (adds PHP as a dependency, removes recode)

1 files changed, 4 insertions, 2 deletions

diff --git a/title.bash b/title.bash
index cc0f83c..7161950 100755
--- a/title.bash
+++ b/title.bash
@@ -20,6 +20,7 @@
 # v1.0  : if Content-Type is text/plain, just use the first line as the title
 # v1.1  : decode HTML in the title using recode (adds recode as a dependency)
 # v1.2  : deal with some cURL exit codes
+# v1.2.1: decode HTML using PHP instead of recode (adds PHP as a dependency, removes recode)
 
 OPTS='--location --insecure --silent --max-filesize 1048576 --max-time 10 -A "Mozilla/5.0 (X11; Linux x86_64; rv:36.0) Gecko/20100101 Firefox/36.0" -H "Accept-Encoding: gzip" -H "Accept: text/html"'
 
@@ -38,7 +39,7 @@ then
   echo "Usage: '!title <url>' where <url> is an http:// or https:// URL"
 elif [ "$url" == "-v" ] || [ "$url" == "--version" ]
 then
-  echo "blatitle version 1.2"
+  echo "blatitle version 1.2.1"
 else
   eval curl $OPTS -D /tmp/$TIME.header '$url' -o /tmp/$TIME.body.gz
   ret=$?
@@ -57,7 +58,8 @@ else
   then
     head -n1 /tmp/$TIME.body
   else
-    awk -vRS="</title>" '/<title>/{gsub(/.*<title>|\n+/,"");print;exit}' IGNORECASE=1 /tmp/$TIME.body | sed ':a;N;$!ba;s/\n//g' | sed -e 's/^[ \t]*//' | recode html..ascii
+    awk -vRS="</title>" '/<title>/{gsub(/.*<title>|\n+/,"");print;exit}' IGNORECASE=1 /tmp/$TIME.body | sed ':a;N;$!ba;s/\n//g' \
+      | sed -e 's/^[ \t]*//' | php -r 'while(($line=fgets(STDIN)) !== FALSE) echo html_entity_decode($line, ENT_QUOTES|ENT_HTML401);'
   fi
   rm /tmp/$TIME.header /tmp/$TIME.body
 fi