User:Nonexyst/OCR-tools.js

Note: After saving, changes may not occur immediately. Click here to learn how to bypass your browser's cache.

Firefox / Safari: Hold Shift while clicking Reload, or press either Ctrl-F5 or Ctrl-R (Cmd-R on a Mac)
Google Chrome: Press Ctrl-Shift-R (Cmd-Shift-R on a Mac)
Internet Explorer: Hold Ctrl while clicking Refresh, or press Ctrl-F5
Opera: Clear the cache in Tools → Preferences

For details and instructions about other browsers, see Wikipedia:Bypass your cache.

Code that you insert on this page could contain malicious content capable of compromising your account. If you are unsure whether code you are adding to this page is safe, you can ask at the central discussion page, Scriptorium. The code will be executed when previewing this page under some skins, including Monobook. You can Purge this page in the interim if you wish to refresh the content sooner under another skin.

Documentation for this script can be added at User:Nonexyst/OCR-tools.

function ocr_1() {
	var textArea = $('#wpTextbox1');

	var txt = textArea.textSelection('getSelection');
	if (txt === '') {
		txt = textArea.textSelection('getContents');
		txt = ocr_1_inner(txt, 1);
		textArea.textSelection('setContents', txt);
	} else {
		txt = ocr_1_inner(txt, 0);
		textArea.textSelection('encapsulateSelection', {pre: txt, replace: true});
	}
}

function ocr_1_inner(sel, r) {
	// manipulate the text in the sel variable
	sel = sel.replace(/[ \t]+$/gm, '');  // remove trailing spaces
	sel = sel.replace(/^[ \t]+/gm, '');  // remove leading spaces
	sel = sel.replace(/[ \t]+/gm, ' ');  // remove multiple spaces

	sel = sel.replace(/(\S)[\-¬\xad]\n(\S)/gm, '$1$2'); // merge lines in the paragraph separated with dash into single line
	sel = sel.replace(/(\S)\n(\S)/gm, '$1 $2'); // merge other lines in the paragraph into single line

	if (r) { // remove leading and trailing line feeds
		sel = sel.replace(/\n/gm, '\x01'); // hide \n's

		sel = sel.replace(/^\x01+/gm, ''); // remove leading line feeds
		sel = sel.replace(/\x01+$/gm, ''); // remove trailing line feeds

		sel = sel.replace(/\x01/gm, '\n'); // restore hidden \n's
	}

	return sel;
}

function ocr_2() {
	var textArea = $('#wpTextbox1');

	txt = textArea.textSelection('getContents');
	txt = ocr_2_inner(txt);
	textArea.textSelection('setContents', txt);
}

function ocr_2_inner(sel) {
	// manipulate the text in the sel variable
	sel = sel.replace(/\n/gm, '\x01'); // hide \n's

	sel = sel.replace(/^(\x01*)([\-a-zA-Z0-9А-Яа-яёЁІіѢѣѲѳѴѵ]+)/gm, '$1{{hwe|$2|$2}}'); // 

	sel = sel.replace(/\x01/gm, '\n'); // restore hidden \n's

	return sel;
}

function ocr_3() {
	var textArea = $('#wpTextbox1');

	txt = textArea.textSelection('getContents');
	txt = ocr_3_inner(txt);
	textArea.textSelection('setContents', txt);
}

function ocr_3_inner(sel) {
	// manipulate the text in the sel variable
	sel = sel.replace(/\n/gm, '\x01'); // hide \n's

	sel = sel.replace(/([\-a-zA-Z0-9А-Яа-яёЁІіѢѣѲѳѴѵ]*[a-zA-Z0-9А-Яа-яёЁІіѢѣѲѳѴѵ])[-¬]?(\x01*)$/gm, '{{hws|$1|$1}}$2'); // 

	sel = sel.replace(/\x01/gm, '\n'); // restore hidden \n's

	return sel;
}

function ocr_4() {
	var textArea = $('#wpTextbox1');

	var txt = textArea.textSelection('getSelection');
	if (txt === '') {
		txt = textArea.textSelection('getContents');
		txt = ocr_4_inner(txt, 1);
		textArea.textSelection('setContents', txt);
	} else {
		txt = ocr_4_inner(txt, 0);
		textArea.textSelection('encapsulateSelection', {pre: txt, replace: true});
	}
}

function ocr_4_inner(sel) {
	// manipulate the text in the sel variable
	sel = sel.replace(/([“‘]) +/gm, '$1');
	sel = sel.replace(/ +([!%\):;,.?”’″′])/gm, '$1');
	sel = sel.replace(/ *— */gm, '—');
	sel = sel.replace(/[“”]/gm, '\"');
	sel = sel.replace(/[‘’]/gm, '\'');

	return sel;
}

jQuery( document ).ready( function( $ ) {

/* Check if view is in edit mode and that the required modules are available. Then, customize the toolbar … */
if ( $.inArray( mw.config.get( 'wgAction' ), [ 'edit', 'submit' ] ) !== -1 ) {
	console.log('page is in edit mode');
	mw.loader.using( 'ext.wikiEditor',  function () {

		console.log('adding toolbar buttons');
		
		////////////////////////////////////////////////////
		$( '#wpTextbox1' ).wikiEditor( 'addToToolbar', {
		///	'section': 'advanced',
			'section': 'proofreadpage-tools',
			'groups': {
				'OCR': {
					'label': 'Additional tools:'
				}
			}
		} );

		$( '#wpTextbox1' ).wikiEditor( 'addToToolbar', {
		///	'section': 'advanced',
			'section': 'proofreadpage-tools',
			'group': 'OCR',
			'tools': {
				'OCR-1': {
					//label: '',
					type: 'button',
					icon: '//upload.wikimedia.org/wikipedia/commons/thumb/8/8b/Merge-arrows.gif/32px-Merge-arrows.gif',
					action: {
						type: 'callback',
						execute: function(context){
							ocr_1();
		                }
		      		}
				},
		      	'OCR-2': {
					//label: '',
					type: 'button',
					icon: '//upload.wikimedia.org/wikipedia/commons/thumb/6/68/Merge-arrow.gif/32px-Merge-arrow.gif',
					action: {
						type: 'callback',
						execute: function(context){
							ocr_3();
		                }
		      		}
				},
		      	'OCR-3': {
					//label: '',
					type: 'button',
					icon: '//upload.wikimedia.org/wikipedia/commons/thumb/a/ae/Mergefrom.gif/32px-Mergefrom.gif',
					action: {
						type: 'callback',
						execute: function(context){
							ocr_2();
		                }
		      		}
				},
				'OCR-4': {
					//label: '',
					type: 'button',
					icon: '//upload.wikimedia.org/wikipedia/commons/d/dd/Dash.svg',
					action: {
						type: 'callback',
						execute: function(context){
							ocr_4();
		                }
		      		}
				}

			}
		} );
		////////////////////////////////////////////////////

	} );
}

} );

User:Nonexyst/OCR-tools.js

Navigation menu

Search