Extend the Formatizer to Recognize further Types of Files

The CombineFormatizer maintains a list of recognizers which are able to recognize certain file types, see List<FormatRecognizer> recognizerList in /src/main/java/de/unirostock/sems/cbext/Formatizer.java

Extending the List of Recognizers

• to extend the Formatizer you can simply pass another recognizer to the method Formatizer.addFormatRecognizer (FormatRecognizer recognizer)
• every Recognizer needs to extend the FormatRecognizer class, see /src/main/java/de/unirostock/sems/cbext/FormatRecognizer.java
• there are a few recognizers added by default, see /src/main/java/de/unirostock/sems/cbext/recognizer
  • BioPaxRecognizer is able to recognize BioPax files
  • CellMlRecognizer is able to recognize CellML files
  • SbgnRecognizer is able to recognize SBGN files
  • SbmlRecognizer is able to recognize SBML files
  • SbolRecognizer is able to recognize SBOL files
  • SedMlRecognizer is able to recognize SED-ML files
  • DefaultRecognizer is a fallback to recognize files by there mime type or extension
• every recognizer has a priority
• the default reconizer has a priority of 100:
  • if you want your recognizer to be asked first give it a higher priority
  • the priorities of recognizers shipped with this library
  • you can change the priority of a recognizer, for example if you know that there are mainly SBML files in your project:
    • SbmlRecognizer.priority = 200
    • afterwards you must resort the recognizers of the Formatizer using Formatizer.resortRecognizers ()
• you can remove all recognizers using Formatizer.removeRecognizers ()
• you can add all default recognizers using Formatizer.addDefaultRecognizers ()

Extending the FormatRecognizer class

The abstract class has 4 functions that you need to implement:

• int getPriority () must return a positive priority. Default recognizers have a priority of 100. Recognizers with a higher priority are asked first for a format. The first who knows a format wins this competition.
• URI getFormatByParsing (File file, String mimeType) you’ll get a files and it’s mime type, try to identify it’s format by looking into the file
• URI getFormatFromMime (String mime) you’ll get the mime type of a file, try to identify it’s format

• URI getFormatFromExtension (String extension) you’ll get the extension of a file, try to identify it’s format

• The format must always be a URI and should be an identifiers.org URL, or, if that does not exist, a purl.org url.
• If you can’t identify a format unambiguously you should return null and the next recognizer will be asked

An example of an extension can be found at PharmMLRecognizer There the Formatizer was extended to also recognize PharmML files.