Included page: .FitNesse.FullReferenceGuide.UserGuide.WritingAcceptanceTests.SliM.SetUp (edit)

setup

import
fitnesse.slim.test
fitnesse.fixtures


The rows in a query table represent the expected results of a query.
You can specify them precisely if you like as in the following:

Query:employees hired before 10-Dec-1980
company number employee number first name last name hire date
4808147 9942 Bill Mitchell 19-Dec-1966
4808147 1429 Bob Martin 10-Oct-1975

Or you can leave cells blank and allow them to be filled in:

Query:employees hired before 10-Dec-1980
employee number first name last name hire date
1429
8832

The code for the fixture is:
package fitnesse.slim.test;

import java.util.Date;
import java.util.List;

import static java.util.Arrays.asList;

public class EmployeesHiredBefore {

  public EmployeesHiredBefore(Date date) {
  }

  public void table(List<List<String>> table) {
    // optional function
  }

  public List<List<List<String>>> query() {

    return
      asList( // table level
        asList( // row level
          asList("company number", "4808147"), // cell column name, value
          asList("employee number", "1429"),
          asList("first name", "Bob"),
          asList("last name", "Martin"),
          asList("hire date", "10-Oct-1974")
        ),
        asList(
          asList("company number", "5123122"),
          asList("employee number", "8832"),
          asList("first name", "James"),
          asList("last name", "Grenning"),
          asList("hire date", "15-Dec-1979")
        )
      );
  }
}

Note the list function simply builds an ArrayList from it's arguments. It's in the ListUtility class

The first thing to notice is the Query: in the first cell of the table. This tells the Slim table processor that this is a query table. Next notice the constructor argument. (See Constructor Arguments). The column headers are field names. The fixture class must have a query method that returns a list of rows. Each row is a list of fields. Each field is a two-element list composed of the field name and it's string value.

Each row in the table is checked to see if there is a match in the query response. The fields are matched left to right. If the leftmost field matches, then the row is considered to be "found". Fields that don't matche are marked in error as in the 10-Oct-1974 field above. A cell that is left blank in the table will be filled in from the result and counted as ignored. If the first cell of a row cannot be matched, then the row is considered missing. If there is an unmatched row remaining in the query response, it is added to the table and marked surplus. The order of the rows is irrelevant.

If a table method is declared in the fixture it will be called before the query function is called. It will be passed a list of rows which are themselves lists of cells. The rows and cells represent the all but the first row of the table. This is the same format at the doTable method of Table table, and the table method of Decision table.

In analogy to decision tables you can define comment columns by prefixing the header with the hash symbol.

Query:employees hired before 10-Dec-1980
last name hire date # comment
Grenning
Martin we hired him first