elasticsearch-annotations

Allows the creation of json mapping files based on class and field annotations.

This library can be used in any project which uses elasticsearch mappings for managing data in the search engine.

Tested using Java 1.7 & Elasticsearch 0.95

Set Up

Run 'mvn clean install' to build the library.

Add the packaged jar to your project to start using it.

Usage:

You can generate json string for the ElasticSearch engine for a class annotated using the library, as shown below:

// provide the name of the class which has been annotated using the library for which you need to 'get' the ElasticSearch mapping
ElasticsearchDocumentMetadata documentMetadata = ElasticsearchMapping.get(Tweet.class);
String mapping = documentMetadata.toMapping(); // Return the json mapping file

Example Class:

An example of a class which uses source field metadata, as shown on the ElasticSearch guide:

@ElasticsearchDocument(
        index = "my-index",
        type = "my_type",
        _elasticsearchFields = {
                @ElasticsearchField(
                        _fieldName = FieldName._SOURCE,
                        includes = {
                                "path1.*", "path2.*"
                        },
                        excludes = {
                                "path3.*"
                        }
                )
        }
)

public class myType {
    private String myValue;
}

Below is another example of a Tweet class, which shows some of the possible combinations of ElasticSearch field and type mappings which can be applied to a class.

@ElasticsearchDocument /** required */(
        index = TweetIndex.class,
//      type = "tweet"    /** optional, if not set it will use the simple class name in a lower hyphenated format */,
        _elasticsearchFields = {
                @ElasticsearchField(
                    _fieldName = FieldName._SOURCE,
                    enabled = BooleanValue.TRUE
                ),
                @ElasticsearchField(
                        _fieldName = FieldName._ALL,
                        enabled = BooleanValue.TRUE
                ),
                @ElasticsearchField(
                        _fieldName = FieldName._PARENT,
                        type = User.class
                ),
                @ElasticsearchField(
                        _fieldName = FieldName._TIMESTAMP,
                        enabled = BooleanValue.FALSE,
                        format = "dateOptionalTime"
                ),
                @ElasticsearchField(
                        _fieldName = FieldName._TTL,
                        enabled = BooleanValue.TRUE,
                        defaultValue = "1d"
                )
        }
)
public class Tweet {

    @ElasticsearchType(
            type = FieldType.STRING,
            index = "not_analyzed"
    )
    private String user;

    @ElasticsearchType(
            type = FieldType.DATE,
            format = "YYYY-MM-dd"
    )
    private String postDate;

    @ElasticsearchType(
            type = FieldType.STRING,
            store = BooleanValue.TRUE,
            index = "analyzed",
            null_value = "na",
            fields = {
                    @ElasticsearchMultiFieldType(_name = "raw", index = "not_analyzed", type = FieldType.STRING)
            }
    )
    private String message;

    @ElasticsearchType(
            type = FieldType.BOOLEAN
    )
    private Boolean hes_my_special_tweet;

    @ElasticsearchType(
            type = FieldType.INTEGER
    )
    private Integer priority;

    @ElasticsearchType(
            type = FieldType.FLOAT
    )
    private Float rank;

    public String getUser() {
        return user;
    }

    public void setUser(String user) {
        this.user = user;
    }

    public String getPostDate() {
        return postDate;
    }

    public void setPostDate(String postDate) {
        this.postDate = postDate;
    }

    public String getMessage() {
        return message;
    }

    public void setMessage(String message) {
        this.message = message;
    }

    public Boolean getHes_my_special_tweet() {
        return hes_my_special_tweet;
    }

    public void setHes_my_special_tweet(Boolean hes_my_special_tweet) {
        this.hes_my_special_tweet = hes_my_special_tweet;
    }

    public Integer getPriority() {
        return priority;
    }

    public void setPriority(Integer priority) {
        this.priority = priority;
    }

    public Float getRank() {
        return rank;
    }

    public void setRank(Float rank) {
        this.rank = rank;
    }
}

Output:

Above Tweet class annotations produces the following output when we try to 'get' the ElasticsearchMappings for the class:

First level of the json is a set of document metadata, followed by properties which provide the type metadata.

{
  "tweet": {
    "_ttl": {
      "default": "1d",
      "enabled": "true"
    },
    "_parent": {
      "type": "user"
    },
    "_source": {
      "enabled": "true"
    },
    "_timestamp": {
      "format": "dateOptionalTime",
      "enabled": "false"
    },
    "_all": {
      "enabled": "true"
    },
    "properties": {
      "postDate": {
        "format": "YYYY-MM-dd",
        "type": "date"
      },
      "hes_my_special_tweet": {
        "type": "boolean"
      },
      "rank": {
        "type": "float"
      },
      "message": {
        "null_value": "na",
        "index": "analyzed",
        "store": "true",
        "type": "string",
        "fields": {
          "raw": {
            "index": "not_analyzed",
            "type": "string"
          }
        }
      },
      "priority": {
        "type": "integer"
      },
      "user": {
        "index": "not_analyzed",
        "type": "string"
      }
    }
  }
}

Testing

The library has been unit tested with a combination of metadata usage, for example, saving to elasticsearch, inner class annotations, sub class annotations, geoshape annotations, etc.