External Application Configuration
The external_applications.conf configuration file
Several components of CollectiveAccess employ external applications to perform various tasks. Typically these tasks relate to conversion and reformatting of uploaded media (images, video, audio, etc.) and indexing of text embedded in uploaded files.
The external_applications.conf file defines the locations of these applications on your server. If an application location is set incorrectly or the application is not installed then the functionality provided by the application will not be available within CollectiveAccess.
The locations you set should be absolute paths to the directory or executable (as specified below) in the standard format for your OS (Unix paths or Windows paths).
The following entries may be defined in this configuration file. Note that there are no default values for entries in external_applications.conf. You must define a value for all applications you wish to use.
|Entry||Description||Typical value (not default - just an example in Unix path format)|
|ghostscript_app||Path to Ghostscript binary ("gs" command) used to generate page images from PDF files||/usr/local/bin/gs|
|ffmpeg_app||Path to ffmpeg binary used to convert video and audio media||/usr/local/bin/ffmpeg|
|qt-faststart_app||Path to [ttp://ffmpeg.mplayerhq.hu qt-faststart] binary used to hint h.264/MPEG-4 video for streaming. qt-faststart is part of ffmpeg and located in the tools/ directory in the source tree.||/usr/local/bin/qt-faststart|
|dcraw_app||Path to dcraw binary used to convert various proprietary RAW formats produced by digital cameras||/usr/bin/dcraw|
|imagemagick_path||Path to directory containing ImageMagick binaries used to convert various image formats. Note that unlike the other entries in this file, imagemagick_path refers to a directory rather than a specific executable||/usr/local/bin|
|pdftotext_app||Path to pdftotext binary (part of the xpdf package from) used to extract text embedded in PDF files||/usr/local/bin/pdftotext|
|abiword_app||Path to abiword binary used to extract text from Microsoft Word files||/usr/local/bin/abiword|
|media_info_app||Path to MediaInfo binary used to extract metadata from media files. MediaInfo is optional and is used if present because it generally does a better job of extracting metadata than the methods built into CA and its media processing plugins.||/usr/local/bin/mediainfo|
|coreimagetool_app||Path to CoreImageTool binary used to process uploaded image files. CoreImage is a Mac OS X-specific API that provides optimized high-performance image processing capabilities. If you run CA on a Mac server you may see a significant performance boost if you run CoreImageTool (we have seen a doubling of performance as compared to ImageMagick in some cases).||/usr/local/bin/CoreImageTool|
|libreoffice_app||Path to LibreOffice 3.6 or better binary used to process uploaded Microsoft Word, Excel and PowerPoint files. If you wish to have Word, Excel and PowerPoint content indexed for search and previews generated, you must have LibreOffice installed. See this cookbook entry for more information.||/usr/bin/libreoffice|
|pdfminer_app||Path to the directory containing the PDFMiner script used to analyze uploaded PDF files. Like PDFToText, PDFMiner can extract text for indexing from PDF files. It can also extract page locations for individual words, enabling search-within-PDF functionality in CollectiveAccess. If you want to search within PDFs with term highlighting in CA PDFMiner must be installed.||/usr/local/bin|