A design method of test data:
When the test data structure, you need to look at the data source, the source of the data in general, there are three, one is based on an analysis of the needs of the system under test, for normal operations, abnormal situation, border and other information to build a complete data, also known as "made" data.
This includes not only the most basic data, such as: users, permissions, configuration, raw data, but also includes business data mentioned above. For relatively small systems and high feasible for large-scale systems can be complex.
Test Data Design Method Two:
The second way is to use the existing system, which has a similar system for the test against a production system upgrade or add functionality. This case the data is already running in a production environment exported. On this basis, re-organize the data,
Processing test data.
Test Data Design Method three:
Another way is to existing non-electronic business data entry into the system, while the business also completed the verification of test data accumulated. That side of the Test side accumulated data. But this accumulation of data tend to have certain limitations because
Already, by the occurrence of basic business data is correct, consistent, and may be missing data for some specific services (business infrequent). This requires an analysis of the test requirements, an additional new test data in order to complete coverage of business
Types of.
Test data applications:
1, the data should not be empty if there is a check;
2, the data of the default value Default value is correct;
3, reference data generated by other functions, whether real-time refresh;
4, after the system restart or shut down the page, the initial setting data, etc.
5, data length, type of control is reasonable, such as ID number, the actual business there will be a letter, and will appear in the final
One
Corresponding method:
Equivalence classes, boundary value, the scene method
Priority angle: Users
Design and maintenance of test data:
<ignore_js_op>
Test Data Summary:
1, not only in the "dirty" data verification environment, we must verify in a clean environment
2, it is necessary to cover each of the data types and business scenario
3, need to take into account the growth and changes in future products
4, not only to ensure that what can be the same to ensure that what does not
5, the data of the add, delete, change impact on historical data (add, delete, change, search)
without there → → → → No change has deleted →
without there → → → → No change has deleted →
6, simplified test execution time, improve testing efficiency
Test Data Example:
<ignore_js_op>
http://www.bcbxhome.com/bcbx/forum.php?mod=viewthread&tid=29&fromuid=27
(出处: 编测编学软件测试)