type母座线接线图:谁能帮忙翻译一下这篇计算机文章?
来源:百度文库 编辑:神马品牌网 时间:2024/09/01 01:07:05
Test data should exercise every path of logic within a program and ideally it should exercise every combination of paths. In a program containing nine decision points, there would be possible logic paths taking 512 separate combinations of test data. If the program was designed as three modules each containing three decisions, however, the testing could be reduced as combinations of test data to test each of the paths in each of the modules, plus a few additional cases to test the communication between the modules.
This is a powerful argument for designing programs as a collection of modules rather than as single monolithic structures.
It is well to remember that it is variety rather than volume which is the requirement of test data. Unless a program is being tested specifically for its ability to handle a certain number of transactions in a given time or its ability to accumulate totals from a series of identical transaction types, there is no merit in providing multiple occurrences of identical data items. In the above example, if the program could handle the pattern A,B,C for one set of data, it could handle that pattern for any number of such sets and further A,B,C test patterns would prove nothing at all. In selecting test data, one must provide a sufficient variety of test data to exercise all paths of a program’s logic and all combinations of occurrences which the program must be able to handle.
This is a powerful argument for designing programs as a collection of modules rather than as single monolithic structures.
It is well to remember that it is variety rather than volume which is the requirement of test data. Unless a program is being tested specifically for its ability to handle a certain number of transactions in a given time or its ability to accumulate totals from a series of identical transaction types, there is no merit in providing multiple occurrences of identical data items. In the above example, if the program could handle the pattern A,B,C for one set of data, it could handle that pattern for any number of such sets and further A,B,C test patterns would prove nothing at all. In selecting test data, one must provide a sufficient variety of test data to exercise all paths of a program’s logic and all combinations of occurrences which the program must be able to handle.
每个测试数据应该在逻辑上,最好的节目应该尽一切结合的路子. 九个点决定了节目,可能会有不同的组合逻辑512路同时测试数据. 如果这一计划的目的是为每三个单元包括三个决定,但可以减少测试组合试验,试验数据的每一道每一单元,加上其他一些案例来考验沟通模式. 这是一个强有力的论据是设计一套节目单元,而不是铁板一块的单一结构. 这件事要记得有不同的要求,而不是量的试验数据. 除非节目特别受到考验,它能否处理一些交易的时候,或能够积累总额由一系列相同类型的交易,没有道理向多次发生相同项目的数据. 上述例子,如果该计划能够处理的模式,B、C为一组数据,它可以搞定任何方式多少套,加重、B、C模式试验证明不了什么忙. 在选择测试数据,就必须提供足够的各种测试数据全部实行了计划道路的逻辑,所有的节目组合出现必须能够处理.