Because the marketing department need to show the background and development of the students after 2 days to develop a display page code, submitted to the test seize finished testing the release. Rich black-box experience for himself, on the data chart test was the first time, become stretched.
After combat, where summary statistics about the functional test. Understand, for the wrong purposes only.
Since it is a data aspects of the test, in general, there are two main scenarios: no data and data.
1. No data
This situation can not be ignored as testers. When a new node means, without adding any child nodes, or a sub-node has no data, verification page is displayed properly. The test found that there 404not found the front end of a script error.
2. Data
There are test data, dividing each module out first, and then combined with the needs of each data block according to the case, the common use of equivalence classes, boundary value test method set out design. The purpose is to improve the scene coverage, improve test efficiency, ensure product quality. Indeed point, the pot back less and less scolded.
Simple example, Figure user amount of sex, for example, the development of the background design of gender have male and female, unknown, respectively, to write test cases
Input: New Gender all "M" / "F" / "unknown", Output: FIG subscribers displays "M" / "F" / "unknown" ratio was 100%
Input: new users are sex "male", "female", Output: FIG subscribers shows "male", "female" in proportions of 50%
Enter: New Gender as "male", "female", "unknown", the output respectively: user chart shows the amount of "male", "female" proportions of 33.3%.
FIG age of the user, in addition to the above uses equivalence class division, need to use a boundary value analysis point considerations of all ages, from the point and the point.