Apache UIMA Java SDK JSON CAS Support

Clone this repo:
  1. cbe763b Merge pull request #25 from apache/no-issue-clean-up-spec by Richard Eckart de Castilho · 9 weeks ago main
  2. 7334ae5 No issue: Comment out alternative sections, future work and clarify UTF-16 code units as the offset counting strategy by Richard Eckart de Castilho · 9 weeks ago
  3. 47f0d5a Merge pull request #23 from apache/release/20-Apache-UIMA-Java-SDK-JSON-CAS-IO-0.5.0 by Richard Eckart de Castilho · 1 year, 2 months ago
  4. ced609b [maven-release-plugin] prepare for next development iteration by Richard Eckart de Castilho · 1 year, 2 months ago
  5. dc58202 [maven-release-plugin] prepare release uimaj-io-json-0.5.0 by Richard Eckart de Castilho · 1 year, 2 months ago rel/uimaj-io-json-0.5.0

JSON serialization of the Apache UIMA CAS

The JSON CAS I/O implementation for use with the UIMA Java SDK allows to serialize UIMA CAS data to JSON and to de-serialize the data back from JSON again, loading it into a CAS object. The aim of this library is to facilitate the data interoperability of UIMA data across different platform and programming languages. For example, the implementation contains functionality to deal with the different character offset counting strategies used by different languages such as Java and Python.

Example

Here is a small example of a JSON CAS document from the unit test data. It is not meant to be a meaningful example, just one that shows some of the elements of the JSON format and how they are encoded. Please refer to the specification for an in-depth description.

{
  "%TYPES" : { },
  "%FEATURE_STRUCTURES" : [ {
    "%ID" : 1,
    "%TYPE" : "uima.cas.Sofa",
    "sofaNum" : 1,
    "sofaID" : "_InitialView",
    "mimeType" : "text",
    "sofaString" : "This is a test"
  }, {
    "%ID" : 2,
    "%TYPE" : "uima.tcas.Annotation",
    "@sofa" : 1,
    "begin" : 0,
    "end" : 4
  }, {
    "%ID" : 3,
    "%TYPE" : "uima.tcas.Annotation",
    "@sofa" : 1,
    "begin" : 5,
    "end" : 7
  }, {
    "%ID" : 4,
    "%TYPE" : "uima.tcas.Annotation",
    "@sofa" : 1,
    "begin" : 8,
    "end" : 9
  }, {
    "%ID" : 5,
    "%TYPE" : "uima.tcas.Annotation",
    "@sofa" : 1,
    "begin" : 10,
    "end" : 14
  }, {
    "%ID" : 6,
    "%TYPE" : "uima.tcas.DocumentAnnotation",
    "@sofa" : 1,
    "begin" : 0,
    "end" : 14,
    "language" : "x-unspecified"
  } ],
  "%VIEWS" : {
    "_InitialView" : {
      "%SOFA" : 1,
      "%MEMBERS" : [ 2, 3, 4, 5, 6 ]
    }
  }
}

Reading and writing JSON CAS

Serializing a CAS to JSON

import org.apache.uima.json.jsoncas2.JsonCas2Serializer

CAS cas = ...;
new JsonCas2Serializer().serialize(cas, new File("cas.json"));

De-serializing a CAS from JSON

import org.apache.uima.json.jsoncas2.JsonCas2Deserializer;

CAS cas = ...; // The CAS must already be prepared with the type system used by the CAS JSON file
new JsonCas2Deserializer().deserialize(new File("cas.json"), cas);

Format specification

For the format specification, please refer here.

Other UIMA JSON CAS implementations

A Python-based implementation of the UIMA JSON CAS format is available as part of the third-party DKPro Cassis library.