shithub: werc

Info  •  Files  •  Log  •  Branches

Much better (I hope) get_html_title implementation, first try to find <title>, if that fails, get the first non-tag string in the file.

--- a/bin/werclib.rc

+++ b/bin/werclib.rc

@@ -98,12 +98,13 @@

 }

 fn get_html_title {

-    # H1 is not reliable because htmlroff doesn't use it :(

-    #desc=`{cat $1 | sed 32q | grep '<[Hh]1>' | sed 's/<[Hh]1>(.*)(<\/[Hh]1>|$)/\1/;s/<[^>]*>//g;1q'}

-    # Pick the first line of body  instead

-    desc=`{sed -n '/<[Bb][Oo][Dd][Yy]/,/./s/(<[^>]*>|$)//gp' < $1}

-    if(~ $#desc 0)

-        desc=`{sed 's/<[^>]*>//g; 1q' < $1}

+    t=`{sed -n '32q; s/^.*<[Tt][Ii][Tt][Ll][Ee]> *([^<]+) *(<\/[Tt][Ii][Tt][Ll][Ee]>.*)?$/\1/p' < $1}

+

+    # As a backup we might want to pick the first 'non-tag' text in the file with:

+    if(~ $"t '')

+        t=`{sed -n -e 's/^(<[^>]+>)*([^<]+).*/\2/p; 32q' < $1 | sed 1q}

+

+    echo $t

 }

 fn get_file_title {