FileIndexer is a MediaWiki depecrecated extension that allows users to search within uploaded text files, such as PDFs, DOCs, TXTs, etc. This version is modified for use with internal-network/intranet use cases based on Mediawiki v1.35. Do NOT use on public wikis.
- Fix: Notice: Undefined variable: output in /var/www/html/WIKI/extensions/FileIndexer/SpecialFileIndexer.php on line 304
- Fix: Notice: Notice: Undefined property: OutputPage::$ in /var/www/html/WIKI/extensions/FileIndexer/SpecialFileIndexer.php on line 304
- Fix: Internal error [9be16b23a4c30900e0bc72e4] /index.php?title=Special:FileIndexer Error from line 304 of /var/www/html/WIKI/extensions/FileIndexer/SpecialFileIndexer.php: Call to a member function addWikiText() on null
- German/Dutch Comments > English
- Others(?)
- Upload changed file versions.
- Create valid extension.json settings per schema here: https://www.mediawiki.org/wiki/Manual:Extension.json/Schema#manifest_version
- Call extension using new methods:
wfLoadExtension( 'FileIndexer' );
To install FileIndexer, clone this repo to the extensions/ directory and add the following line to LocalSettings.php:
# FILEINDEXER EXTENSION TO ALLOW UPLOADED FILES TO BE INDEXED/SEARCHABLE
# HELPER "convPDFtoTXT" OCR use (?):
$wgFileIndexer_OCRisActive = true;
# EXTENSION CALL MEDIAWIKI v1.35:
# wfLoadExtension( 'FileIndexer' );
# OLD EXTENSION CALL METHOD:
require_once( "$IP/extensions/FileIndexer/FileIndexer.php" );
Install the required dependencies listed at FileIndexer.config.php (lines 59-66).
(per FileIndexer.config.php lines 59-66)
- convPDFtoTXT
- pdftotext
- iconv
- antiword
- xls2csv
- catppt
- strings
- unzip
- convPDFtoTXT:
(?) - pdftotext:
sudo apt-get install poppler-utils - iconv:
sudo apt install libc6 - antiword:
sudo apt install antiword - xls2csv:
sudo apt-get install catdoc - catppt:
sudo apt install catdoc - strings:
apt-get install binutils - unzip:
apt-get install unzip
- Search here for dependencies install method/packages: https://command-not-found.com/
Documentation missing, check the FileIndexer.config.php file directly.
By default, FileIndexer will try to use the command line tool "pdftotext" to index PDF files. However, by setting $wgFileIndexer_OCRisActive = true; in your LocalSettings before requiring the extension, the "convPDFtoTXT" tool will be used, which uses optical character recognition (OCR) to scan and index PDF files.
Once the extension has been installed and configured, log in as an admin and go to Special:FileIndexer to generate the index of the uploaded text files. After doing that the files will be searchable from the regular search box. New files are automatically indexed (?).