I have a software test
I have a short period of time, and a lot of blind tests have almost no longer grow my test experience. There is always a little in front of each test, I don't know how I can effectively discover the defects in the software; It is not possible to release the test software that can be assured. I think many of the initial testing workers have similar feelings. We need to guide the theoretical knowledge of tests. The following is the harvest after I read some books that explain the test technology, and my current software industry in my country. In this area of understanding, I hope that I can give the test of peers.
First, the software tester is to find some software defects as early as possible and make sure it is closed.
Perhaps everyone will think that the work goal of the software tester is self-evident: It is to find software defects, but "software test" this book puts a more exact goals for software testers: as early as possible to find software defects, And make sure it is repaired. After reading the book and think carefully, I think this goal contains three major meanings:
1. The basic goal of software testers is to discover software defects.
I think it is necessary to emphasize this here again, because sometimes the product's development team wants to prove that software can run, not to find defects. In this case, the tester also lacks unremitting efforts to discover the spirit and enthusiasm of defects. So I think it is believed that it is believed that the basic goal of software testers is to discover software defects and is the primary condition for testing.
2. Software testers pursue is to identify software defects as early as possible.
Because of the repair cost of the software, with time, it will grow up to dozens of times, so the software tester should find out software defects as early as possible. For large software, while software development, there should be followed by the following tests. If the product has been developed, it will start testing, which is very likely to cause a large amount of time-consuming rework. How to find a defect as early as possible? "Software Test" This book introduces us some theoretical test methods: static black box test, dynamic black box test, static white box test, dynamic white box test; configuration test, compatibility test, easy-to-use test ... ..., how to effectively use these methods as soon as possible to discover software defects, need everyone to continue to explore, summarize, and continue to improve their testing capabilities in work practice.
3. Software testers must ensure that the software defects identified are closed.
Please note that we mentioned here that software testers must ensure that the software defects they find are closed, not software defects to be repaired. Our software testers need to maintain a normal heart to the software defects they find: It is not that we have worked hard to find every software defect is necessary to fix. It may be because there is not enough time, it is not a real software defect, and the repair of the risk is too large. The product development team decided not to fix some software defects.
In addition, the tester describes the software defects unclear, and the defects discovered by the software tester are ignored. Therefore, our tester must work hard in describing software defects: describe software defects with simple and clear statements; each piece of report is trying to make a software defect, avoiding multiple defects, which are ignored or forgotten; record The operational steps of the software defects have caused the defect to be reproduced.
Although our software tester needs to maintain a normal heart, but also must adhere to the principle of endless, track the processing results of each software defect, ensuring that software defects are closed. Turning off software defects can be defects to be repaired or decided not to fix. Whether the defects need to be repaired in the final person in charge of the software, check the defects to be closed.
Second, the most important thing for testing a software is to test its product function manual.
2 concept
Product function manual: Description of the functionality that the product ultimately needs. These features are ultimately determined customer needs, including the capabilities of some software must have.
In the process of standard software generation, the product function manual shall be determined before the system's summary design is performed after the user needs assessment meeting. 2 reasons
1. Many software defects are because the product function manual is not comprehensive, often changed;
2. Only the product function manual is read in detail, confirming the functionality that the product needs to be implemented can formulate practical testing options;
2 method
1. Referring to the demand manual, check whether the product described by the product function specification will be complete, accurate, consistent, and reasonably describe the functionality of the product, and ensure that these functions are tested.
2. Studying whether product manual meets the standards or norms of existing software design and development;
3. Study similar software, predict the final result of the product;
What should I do if the tester finds that the product manual does not meet the above points?
Test staff need to understand that our responsibility is to reflect the defect of the product, we don't need to correct the product, so the same discovery software, after discovering the defect of the product manual, should record them in detail, report The final person in this software is tracked on the processing of and this defect.
Note that the other discovered software is the same, the list of defects should be reported to the final person in charge of the software, not to relevant technical personnel or technical director, because the technician may have difficulty in the implementation of technology, refused to defective modify.
2 current executable
1. Many software do not have a written product manual before the development
At present, many software companies in our country are small hand workshops, there is no formal, complete, determined product manual before the program development, even if this is the case, the product manual exists, it exists in software In the mind of the designer, the head of the project, there is a software contour in their hearts, assuming the software will implement. Our testers can get this informal product manual in the same communication and constant inquiry, it is worth noting that we need to list them in writing after we get this information, and then to relevant personnel Ask, in the end, it can be complete, accurate, consistent, and reasonably describe the functionality of the product.
2. Test staff generally do not participate in the project in the initial stage of the project
There is currently a problem, the company generally does not allow software testers to participate in the project in the early days of the project, generally waiting to wait until the skin's prototype will only make software testers to test. In this case, the tester can try to figure out the product instructions through the prototype of the software already established, and then the same as the above session, ask the relevant personnel, to form a written complete, accurate, consistent, reasonable product Talk about the book. It is worth noting that when the tester runs the prototype of the software, some software defects often find some software defects. At this time, don't limit the cost of spending on these defects, so that the main task of the development of the product manual is, must remember: test One software is the most important thing is to test its product manual, and then develop specific test cases after the product manual is clear.
The above two points refer to the suggestion of the tester in less information on the company system, but I think it is better to ensure the quality of the software, maybe the entire operation process of the standard generation software is more effective: product manual is responsible People come hosting a set of places, because he grasps the direction of the product development; the meeting at the time of the product manual should be responsible for the person responsible for testing, so that they have a macro understanding of the product, I don't agree with the testers Earlier is limited to a project, if they feel nothing.
Third, the full test software is absolutely impossible and must be equivalent to the test.
2 concept
Equivalent allocation: Software has unlimited test cases, we must find ways to enter the similar input, output, operation into a group, to reduce unlimited test cases to the same effective small range, this process is called equivalent assignment . Boundary conditions: The edge condition of the operating limit of the software program, that is, if this boundary condition is exceeded, it may be taken out.
2 reasons
1. Great input
2. Too many output results
3. Software implementation is too much
4. Software instructions have no objective standards. From different perspectives, the standard of software defects is different.
2 method
1. Data test:
1) Determine the input boundary condition, test the data on both sides of the boundary line and the boundary line;
2) The boundary line may be 2 multiplines, default values, blank values, zero values, etc .; each software test problem is different, which may include different data in the format.
2. Status test (Software status refers to the current situation or mode of the software)
1) At least once each state;
2) Test the most common state conversion;
3) The least common branch between the test state;
4) Test all incorrect status and its return value;
5) Test random state transition.
2 current executable
If the excessive equivalent assignment of the quantity of the test case is increased, the risk of tested the shorter software defect will increase. For initial software testers, you must ask experienced testers to review the scheduled equivalents.
