| Auto testing, a major part of software testing, has been much studied in this fieldat present. Most recently, automation of functional testing is the mainstream. Withrespect to automation of network testing, including automation of UI testing and unittesting, though such powerful independent testing tools as QTP，RFT，WinRunner areat hand as solution alternatives, business would find it hard to implement auto testingusing a given testing tool in practice. And here lie some of the reasons: theoretically,auto testing by means of testing tools saves you some work, while in actuality itseffectiveness and performance are not mutually agreeable with investment and output.On the other hand, the way the tools thereof remain consistent, while the situationwhere auto testing occurs is variable, such as the changing requirements, the case andscript etc. Any given testing tool can’t possibly meet these changes. Hence, in realtesting practice, enterprises could need an auto testing solution that best serves itsbusiness characteristics testing framework, to make real reliable and effectivesoftware auto testing.Drawing upon the aforementioned researches and applications, this thesis studiedthe technical know-how of software testing and testing framework, provides profoundinsight into the fundamentals and methods of auto testing, and features KDATF, thetesting framework which evolves from combined driven of keywords and data.Containing the creation of testing case and the execution thereof, KDATF,representing a comprehensive auto testing framework and covering the whole autotesting process, is intended for developing testing for relevant systems in the financialindustry. Highlighting the core function of KDATF framework, this thesis deeplyanalyzed and studied how to design it and its key technologies involved. The creationof testing case contains design of case template, establishment of elementary models,drawing up of component rules, creation of business components, creation of testingpoint and such key technologies as data driven, keywords driven, rules driven, casedissection technology. Data driven refers to making variable input and output on thecase template. Reading exterior data documents helps creating great number of testingcases. Keywords driven means describing testing cases in some designated key wordsother than the common as we use daily. Combined with the case dissection technology, it makes for testing scripts. Analyzing and dissecting the system to be tested, followedby drawing up of appropriate elements and rules, rules driven breeds testingenvironment and helps further with the creation of the testing case. Execution oftesting case involves the preliminary handling of script, the marking and execution thescript, as well as the drawing up of the testing report. The technologies included areASL source language, script marking, keywords driven and so on. An advancedscript-speak, ASL elevates script creation businesswise. Testing cases as described byASL proves easy to transform into scripts executable by auto testing tools.As far as the enterprises are concerned, KDATF doesn’t represent a concreteproduct, but a set of abstract components and the dynamics of these cases. Dependingon their actual business needs, these enterprises could tailor-make a testing platformwith KDATF framework.This thesis sketches out an auto testing platform intended forfinance systems, based on the KDATF it has hitherto highlighted. Employing RFT（Rational Function Tester）of IBM as the auto testing tool and designating the debtsregistry of a certain country as the target testing system, this platform has done asuccessful job of testing this system, and thereby confirmed the practicality and valueof the KDATF framework. |
PDF Full Text Request