shithub: werc

Info  •  Files  •  Log  •  Branches

New context var 'local_file' that will point to the on-disk-file that represents the main content of a given page request (if such exists, otherwise it is '' or ()).

Move title extracting functions from sitemap.tpl to werclib.rc.
Rewrite .md title extraction code to only match first lines when 'underlines' with =.

Use the changes above to generate better page titles when possible.

--- a/bin/corehandlers.rc

+++ b/bin/corehandlers.rc

@@ -85,17 +85,26 @@

 fn setup_handlers {

-    if(test -f $local_path.md)

-        handler_body_main=(md_handler $local_path.md)

-    if not if(test -f $local_path.tpl)

-        handler_body_main=(tpl_handler $local_path.tpl)

-    if not if(test -f $local_path.html)

-        handler_body_main=(html_handler $local_path.html)

+    if(test -f $local_path.md) {

+        local_file=$local_path.md

+        handler_body_main=(md_handler $local_file)

+    }

+    if not if(test -f $local_path.tpl) {

+        local_file=$local_path.tpl

+        handler_body_main=(tpl_handler $local_file)

+    }

+    if not if(test -f $local_path.html) {

+        local_file=$local_path.html

+        handler_body_main=(html_handler $local_file)

+    }

     # Global tpl (eg sitemap.tpl), should take precedence over txt handler!

     if not if(test -f tpl^$req_path^.tpl)

+        # XXX Should we set $local_file for global .tpls?

         handler_body_main=(tpl_handler tpl^$req_path^.tpl)

-    if not if(test -f $local_path.txt)

-        handler_body_main=(txt_handler $local_path.txt)

+    if not if(test -f $local_path.txt) {

+        local_file=$local_path.txt

+        handler_body_main=(txt_handler $local_file)

+    }

     # XXX Should check that $enabled_apps exist in $werc_apps?

     # XXX Should split init of apps that provide main handler (eg., blog) and apps that don't (eg., comments)?

--- a/bin/werc.rc

+++ b/bin/werc.rc

@@ -46,6 +46,7 @@

     req_path=`{echo -n $REQUEST_URI | sed 's/\?.*//; s!//+!/!g; s/%5[Ff]/_/g; s/'^$forbidden_uri_chars^'//g; s/\.\.*/./g; 1q'}

     req_url=$base_url^$req_path

     local_path=$sitedir$req_path

+    local_file=''

     ifs='/' { args=`{echo -n $req_path} }

     # Preload post args for templates where cgi's stdin is not accessible

@@ -103,13 +104,23 @@

             perm_redirect $t

     }

+    setup_handlers

+

+

     # Set Page title

-    if(~ $"pageTitle '')

+    if(! ~ $local_file '') {

+        t=`{get_file_title $local_file}

+        if(! ~ $"t '')

+            pageTitle=$t

+    }

+

+    dprint X^$"pageTitle^X

+

+    # XXX Is this never true? because we set pageTitle earlier based on url.

+    if(~ $"pageTitle '') 

         pageTitle=$"siteTitle' '$"siteSubTitle

     if not

         pageTitle=$"pageTitle' | '$"siteTitle' '$"siteSubTitle

-

-    setup_handlers

     for(h in $extraHttpHeaders)

         echo $h

--- a/bin/werclib.rc

+++ b/bin/werclib.rc

@@ -90,6 +90,36 @@

     sed -n '/^\* '$2': /p; /^\* '$2': /q; /^$/q' < $1

 }

+# File title extraction

+

+fn get_md_title {

+    #sed 's/^(................................................................[^ ]*).*$/\1/g; 1q' < $1 

+    sed -n -e '1N; /^.*\n===*$/N; /.*\n===*\n$/!b' -e 's/\n==*\n//p' < $1

+}

+

+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}

+}

+

+fn get_file_title {

+    if (~ $1 *.md)

+        get_md_title $1

+    if not if(~ $1 *.html)

+        get_html_title $1

+    if not if(~ $1 */) {

+        if(test -f $1/index.md)

+            get_md_title $1/index.md

+        if not if(test -f $1/index.html)

+            get_html_title $1/index.html

+    }

+}

+

+

 ##########################################################################

 ##########################################################################

 #app_blog_methods = ( _post index.rss )

--- a/tpl/sitemap.tpl

+++ b/tpl/sitemap.tpl

@@ -4,35 +4,6 @@

 tmpfile=/tmp/werc_sitemap_$pid.txt

 saveddf=$dirfilter

-fn get_md_title {

-    sed 's/^(................................................................[^ ]*).*$/\1/g; 1q' < $1 

-}

-

-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}

-}

-

-fn get_file_title {

-        

-    if(~ $1 */) {

-        if(test -f $1/index.md)

-            get_md_title $1/index.md

-        if not if(test -f $1/index.html)

-            get_html_title $1/index.html

-    }

-    if not if(~ $1 *.md)

-        get_md_title $1

-    if not if(~ $1 *.html)

-        get_html_title $1

-    if not

-        echo ''

-}

-

 fn listDir {

     d=$1

     dirfilter=$saveddf