Difference between revisions of "User:Paxed/ReplaceCharsBlock"

From NetHackWiki
Jump to navigation Jump to search
m (todo, allow rules from another article, idea from User:Ilmari Karonen)
m (doc ruleset templates)
Line 76: Line 76:
 
</div>
 
</div>
 
|}
 
|}
 +
 +
Predefined replacement rulesets can be used with <tt><nowiki><replacecharsblock rules="ttymap"></nowiki></tt>, which would load rules from {{tl|ttymap}}. The lines after the first empty line in the ruleset template are ignored.
  
 
==TODO==
 
==TODO==
*Allow loading replacement rules from another article, use <nowiki><replacecharsblock rules="article_name"></nowiki>, which would load (by default) the fules from Template:article_name, as per <tt>Title::newFromText($titleText, NS_TEMPLATE)</tt>.
 
 
*Allow multiple replacement definitions for same char; use the replacements in the order they're defined, looping.
 
*Allow multiple replacement definitions for same char; use the replacements in the order they're defined, looping.
 
**Maybe also in random order, though that'll invalidate cache.
 
**Maybe also in random order, though that'll invalidate cache.

Revision as of 15:16, 1 February 2011

ReplaceCharsBlock extension

Adds a <replacecharsblock> -tag, which allows replacing (or translating, using php:strtr) characters and strings in the block with other characters or strings. Replacements are defined first, and the block of text to be replaced after an empty line.

Replacement format is replace_from=replace_to, one per line.

<replacecharsblock>
b=foo

abcb
</replacecharsblock>

afoocfoo

  • default as a replacement: used for any characters that don't have explicitly defined replacement. $1 in the replacement will be substituted with the character which is being replaced.
  • newline as a replacement: each newline character will be replaced with this.
  • Longer replacement strings will be replaced first, then shorter ones.
  • Replacements are case sensitive.

For example:

<div class='ttyscreen'>
<replacecharsblock>
k=[[kobold|{{red|k}}]]
k̂=[[kobold shaman|{{brightblue|k}}]]
f=[[cat|{{white|f}}]]
@={{white|@}}
|={{lightgray|&#124;}}
bar=[[foo|{{yellow|foo}}]]
default={{lightgray|$1}}
newline=<br>

------- ------- ------- -------
|...... |.BAR.. |...... |kkkk̂k.
|.....| |.bar.| |.....| |kkkkk|
|..k..| |.....| |..f..| |kkfkk|
|.kfk.| |..kk.| |.kkk.| |k...k|
|.k@k.| |..@fk| |..@..| |kk@kk|
|.....| |..kk.| |.....| |kkkkk|
-.----- -.----- -.----- -.-----
</replacecharsblock>
</div>

------- ------- ------- -------
|...... |.BAR.. |...... |kkkkk.
|.....| |.foo.| |.....| |kkkkk|
|..k..| |.....| |..f..| |kkfkk|
|.kfk.| |..kk.| |.kkk.| |k...k|
|.k@k.| |..@fk| |..@..| |kk@kk|
|.....| |..kk.| |.....| |kkkkk|
-.----- -.----- -.----- -.-----

Predefined replacement rulesets can be used with <replacecharsblock rules="ttymap">, which would load rules from {{ttymap}}. The lines after the first empty line in the ruleset template are ignored.

TODO

  • Allow multiple replacement definitions for same char; use the replacements in the order they're defined, looping.
    • Maybe also in random order, though that'll invalidate cache.
  • Exceptions? (eg. maybe one 'k' is different from all the other 'k' chars) Format, how to define?


Code for ReplaceCharsBlock.php

<?php
/**                                                                                                                                          
 * ReplaceCharsBlock -- replace chars in a block with strings.                                                                               
 *                                                                                                                                           
 */

if( !defined( 'MEDIAWIKI' ) ) {
        die( "This is not a valid entry point to MediaWiki.\n" );
}

$wgHooks['ParserFirstCallInit'][] = 'wfReplaceCharsBlock';

$wgExtensionCredits['parserhook'][] = array(
        'name' => 'ReplaceCharsBlock',
        'version' => '0.3',
        'author' => array(
                '[http://nethackwiki.com/wiki/User:Paxed Pasi Kallinen]',
                '[http://nethackwiki.com/wiki/User:Ilmari_Karonen Ilmari Karonen]',
        ),
        'url' => 'http://nethackwiki.com/wiki/User:Paxed/ReplaceCharsBlock',
        'description' => 'Replace characters in a block of text with strings.',
);

function wfReplaceCharsBlock( &$parser ) {
        $parser->setHook( 'replacecharsblock', 'renderCharReplacement' );
        return true;
}

function renderCharReplacement( $input, $argv, $parser, $frame ) {
        /*$parser->disableCache();*/

        list( $defs, $text ) = preg_split( '/\n\s*\n/', $input, 2 );
        if ( !isset( $text ) ) {
                $text = $defs;
                $defs = "";
        }

        $reps = array();
        $default = false;

        $defs = explode( "\n", $defs );
        foreach ( $defs as $i => $line ) {
                if ( !preg_match( '/\S/u', $line ) ) {
                        continue;
                }
                if ( !preg_match( '/^(\X+?)=(.+)$/u', $line, $match ) ) {
                        return "<strong class='error'>Replacecharsblock error on line $i: invalid definition " . htmlspecialchars( $line ) . "</strong>";
                }
                // wfDebug( __METHOD__ . ": mapping $match[1] to $match[2]\n" );                                                             
                if ( $match[1] === 'default' ) {
                        $default = $match[2];
                } else if ( $match[1] === 'newline' ) {
                    $reps["\n"] = $match[2];
                } else {
                        $reps[ $match[1] ] = $match[2];
                }
        }

        if ( $default !== false ) {
                preg_match_all( '/(?>[^\\pM\\pC]\\pM*)/u', $text, $chars );
                $chars = array_diff( array_unique( $chars[0] ), array_keys( $reps ) );
                foreach ( $chars as $char ) {
                        $reps[$char] = str_replace( '$1', $char, $default );
                        // wfDebug( __METHOD__ . ": adding default map from $char to $reps[$char]\n" );                                      
                }
        }

        $text = strtr( $text, $reps );
        return $parser->recursiveTagParse( $text, $frame );
}