Querying with Conditions

You can apply conditions to queries by using QueryCondition objects and then passing those objects to the find() or findById() methods. All filtering, except for filtering on expressions, takes place in MapR-DB before results are passed back to client applications.

NOTE: Because documents are not indexed, queries that use conditions require full table scans and can be CPU-intensive.

Creating a QueryCondition object

Suppose that you have a JSON table named Movies that contains documents about movies. You want to query the table to find all of the romantic comedy movies that were released in the year 2015 by a particular movie studio.

You could build a query condition like this:

  public QueryCondition buildQueryCondition() {
    return MapRDB.newCondition()
        .and()
          .is("studio", Op.EQUAL, "RomComs Unltd.")
          .is("release_date", Op.GREATER_OR_EQUAL, Values.parseDate("2015-01-01"))
          .is("release_date", Op.LESS_OR_EQUAL, Values.parseDate("2015-12-31"))
        .close()
      .build();
  }

The .and() requires that all of the conditions set by the nested .is() specifications must be true. The enum Op says to compare the value in the filter with the value in each document for the given field. The keyword after Op specifies the type of comparison to make.

QueryCondition objects are described further in the OJAI documentation.

Returning multiple whole documents by querying with conditions

Call the find() method of a Table object that represents the table. When calling the method, pass a QueryCondition object.

The result is a DocumentStream object, which is a collection of Document objects that contain the returned whole OJAI documents.

 public void findDoc(String tablePath, QueryCondition condition) {
    try(Table table = MapRDB.getTable(tablePath);
        DocumentStream documentStream = table.find(condition)) {
      for(Document document : documentStream) {
        System.out.println(document);
      }
    }
  }

Returning multiple partial documents by querying with conditions

Call the find() method of a Table object that represents the table. When calling the method, pass a QueryCondition object and the path or paths of the fields to include in the returned documents.

The result is a DocumentStream object, which is a list of Document objects that contain the returned partial OJAI documents.

 public void findDoc(String tablePath, QueryCondition condition, String[] fieldPaths) {
    try(Table table = MapRDB.getTable(tablePath);
        DocumentStream documentStream = table.find(condition, fieldPaths)) {
      for(Document document : documentStream) {
        System.out.println(document);
      }
    }
  }

Returning single whole documents by querying with conditions

Call the findById() method of a Table object that represents the table. When calling the method, pass the value of the _id field in the document and a QueryCondition object.

The result is a Document object that contains the returned whole JSON document.

 public void findDocById(String tablePath, String _id, QueryCondition condition) {
    try(Table table = MapRDB.getTable(tablePath)) {
      System.out.println(table.findById(_id, condition));
    }
  }

Returning single partial documents by querying with conditions

Call the findById() method of a Table object that represents the table. When calling the method, pass the value of the _id field in the document, a QueryCondition object, and the paths of the fields that you want to include in the returned document.

The result is a Document object that contains the returned partial JSON document.

 public void findDocById(String tablePath, String _id, QueryCondition condition, String fieldPath) {
    try(Table table = MapRDB.getTable(tablePath)) {
      System.out.println(table.findById(_id, condition, fieldPath));
    }
  }