Jump to content

Extension talk:Enscript

Add topic
From mediawiki.org

When I install and run this extension on Fedora 6, the Apache error log spits out:

sh: /enscript: No such file or directory

enscript is installed on the machine and is located at the default location:

/usr/bin/enscript

The version of Mediawiki is 1.8.5. Any idea why this is happening?

take-off on Enscript w/ Kate...

[edit]

Did something very similar, but using the perl port of Kate. Very clearly ripping off Yedidia, the php looks like:

<?php
#----------------------------------------------------------------------------
#    Extension initialization
#----------------------------------------------------------------------------
$KateVersion = '0.01';

global $wgExtensionCredits;
$wgExtensionCredits['parserhook'][] = array(
    'name'=>'Kate',
    'version'=>$KateVersion,
    'author'=>'xxxx',
    'url'=>'http://www.mediawiki.org/wiki/Extension:kate',
    'description' => 'Show code highlighted w/ Kate'
    );

$wgExtensionFunctions[] = "wfKate";

function wfKate() {
    global $wgParser;

    $wgParser->setHook( "Kate", "renderKate" );
}

function renderKate( $input , $argv, &$parser ) {
  $file['kate'] = "/usr/bin/perl /var/www/html/wiki/extensions/hlhtml.pl";
  $file['kate-options'] = $argv['lang'];

  $output="";

$descriptorspec = array(
   0 => array("pipe", "r"),  // stdin is a pipe that the child will read from
   1 => array("pipe", "w")  // stdout is a pipe that the child will write to
);
$cwd = '/tmp';
  $ph=proc_open($file['kate']." ".$file['kate-options'],$descriptorspec,$pipes,$
cwd);
  fwrite($pipes[0],$input);
  fclose($pipes[0]);
  $output=stream_get_contents($pipes[1]);
  fclose($pipes[1]);

  $return_value = proc_close($ph);

  $output = str_replace("<H1>(stdin)</H1>","",$output);
  return $output;
}

?>


Enscript issues with MW 1.11.2 and Enscript 1.6.4

[edit]

Had trouble getting things to work well with MediaWiki 1.11 and Enscript 1.6.4:

  • Enscript made a full set of headers and put "credits" after every bit of code
  • Odd errors appeared on the rendered page

The following hacked up version works OK but should probably be edited by someone who knows what she/he is doing. It would be really nice to have one footnote per page giving Enscript credit for generating the printout, but that's a bit beyond my skillset at this time...

<?php

$EnscriptVersion = '0.2';

#----------------------------------------------------------------------------
#    Extension initialization
#----------------------------------------------------------------------------
global $wgExtensionCredits;
$wgExtensionCredits['parserhook'][] = array
  (
  'name' => 'Enscript',
  'version' => $EnscriptVersion,
  'author' => 'Yedidia Klein',
  'url' => 'http://www.mediawiki.org/wiki/Extension:enscript',
  'description' => 'Show code highlighted w/ GNU Enscript'
  );

$wgExtensionFunctions[] = "wfEnscript";

#----------------------------------------------------------------------------

function wfEnscript ()
  {
  global $wgParser;
  $wgParser->setHook ("Enscript", "renderEnscript");
  }

#----------------------------------------------------------------------------

function renderEnscript ($input, $argv, &$parser)
  {
  $file['enscript'] = "/usr/bin/enscript";
  $file['enscript-options'] = " -q --language=html --color --style=msvc -E".$argv['lang']." -o -";

  $output = "";

  $descriptorspec = array
    (
    0 => array ("pipe", "r"),  // stdin is a pipe that the child will read from
    1 => array ("pipe", "w")   // stdout is a pipe that the child will write to
    );
  $cwd = '/tmp';
  $ph = proc_open ($file['enscript']." ".$file['enscript-options'], $descriptorspec, $pipes, $cwd);
  fwrite ($pipes[0], $input);
  fclose ($pipes[0]);
  $output = stream_get_contents ($pipes[1]);
  fclose ($pipes[1]);

  $return_value = proc_close ($ph);

  // Added by JR -- Knock off the HTML header, which screws things up pretty badly
  $ablank = "<! >";
  $output = preg_replace ('/<\!DOCTYPE.*>/', $ablank, $output);
  $output = preg_replace ('/<HTML>\n<HEAD>\n<TITLE>.*TITLE>\n<\/HEAD>/', $ablank, $output);
  $output = preg_replace ('/<BODY.*>/', $ablank, $output);
  $output = preg_replace ('/<ADDRESS.*ADDRESS>/', $ablank, $output);
  $output = preg_replace ('/<A.*>/', $ablank, $output);
  $output = str_replace ("<H1>(stdin)</H1>", $ablank, $output);
  $output = str_replace ("<HR>", $ablank, $output);
  $output = preg_replace ('/<PRE>\n/', "<PRE>", $output);
  $output = preg_replace ('/<\/BODY>\n<\/HTML>\n/', $ablank, $output);

  return $output;
  }

?>

HTML code cleanings

[edit]

Replace

$output = str_replace("<H1>(stdin)</H1>","",$output);

with

$output=preg_replace("/^.*<pre>/si","<pre>", $output);
$output=preg_replace("/<\/pre>.*$/si","</pre>", $output);

It will clear all unnecessary HTML tags from output (DOCTYPE, HTML, etc).