[pmwiki-users] Include specific lines of text on a page

[Hans]

Saturday, January 19, 2008, 7:37:40 PM, Vince Administration wrote:

> Hans, here is a start. Your approach extracted a row containing a  
> given string. The syntax below
>>> (:grep "regex" wikifilename/pattern . :)
> is meant to say
> :grep  means extract
> "regex' means each line satisfying a criterion, so instead of just  
> containing a given string, you could have
> wild cards, etc, like in php's search replace functions.  The :pipe
> etc would just concatenate those.

Thanks! I also had an email from Peter where he explained his ideas to
me further. i then concocted another markup expression definition for
{(grep 'PATTERN' PAGENAME )}, where PAGENAME can take wiki wildcard
patterns. Here it is:

# {(grep TEXTPATTERN PAGENAME)}
$MarkupExpr['grep'] = 'Grep($pagename, @$args[0], @$args[1])';
function Grep($pagename, $expr, $src) {
        if ($expr=='' || $src=='') return '';
        $grp = PageVar($pagename, '$Group');
        //check for group.name pattern
        if (strstr($src,'.')) 
                $pat = $src;
        else $pat = $grp.".".$src;
        //make preg pattern from wildcard pattern
        $prpat = GlobToPCRE(FixGlob($pat));
        //make list from preg name pattern
        $sourcelist =   ListPages("/$prpat[0]/"); 
        $newrows = array();
        //process each source page in turn
        foreach($sourcelist as $source) {
                if ($source==$pagename) return '';
                $page = RetrieveAuthPage($source, 'read', true);
                $text = $page['text'];
                $textrows = explode("\n",$text);
                foreach ($textrows as $row) {
                        if (preg_match("/$expr/", $row)) {
                                $newrows[] = $row;
                        }
                }
        }
        return implode("\n", $newrows);
}


The preg_match pattern in
                        if (preg_match("/$expr/", $row)) {
                                $newrows[] = $row;
                        }
is still not working quite well.
any suggestions for improvement will be much appreciated.


  ~Hans