steven_yetter
Contributor
Joined 3 years ago
User Profile
User Widgets
Contributions
Specifying Test Plan Execution without using the TestPlan GUI
We have a test Plan that is getting long. We are using a single suite and project for the application under test. We don't see a way to have multiple test plans. How do we tell Test Complete to execute portions of the Execution Plan. We have the ability to tell Test Complete to execute a test on the command line, but doing that for all the tests but not all of the tests would be a really long command line.319Views1like2CommentsRe: Can I get all my test names for a project into a file?
Going back to the subject of this thread. Your TestComplete GUI doesn't have a list of tests to run? Your code has the list of tests? Your code is one script? Your log messages are in your script? Can you send a screen shot of TestComplete?811Views0likes0CommentsRe: Can I get all my test names for a project into a file?
xml files don't have to have the extension .xml. The header can tell Windows what the XML file is, what icon to put on it, and what to execute. Below are some examples of TestComplete files that are xml files under the covers. The pjs file is xml. Right click on it and open with notepad++. The pjs.tcLS file is xml. The tcTgExtender file is xml The tckDTest file is xml but that is keyword tests The .mds file is xml. I think that is a project.821Views0likes0CommentsLogging Test Results
In the Log I get green/pass check and red/fail log entries. Eg. Pass: The property checkpoint passed: Visible equals True. Fail: Unable to verify a value with the property checkpoint because the target object does not exist. Is there a way to customize the Pass message above so that the log. "The IsoSpeak caption of the splash screen is visible" Or some other text so that we identify what was visible. To change the Pass message do you have to disable the existing logging for a checkpoint. On the Fail, is there a way to distinguish between target object does not exist vs. the value of a property does not match. For object does not exist, I would try it as a fatal error and stop execution, but on a value not matching, I would report a failed test and continue running more tests. There is an option on Playback to continue on error or stop on error. What constitutes an error? Is a non-matching checkpoint value an error. I read a list of object types that causes an error. If a test app raises a warning message that only exists under a certain condition, how do we test for warning message object if object does not exist is an error. I want not having the warning to be success and continue.263Views0likes1CommentRe: "Common" or "Lib" Project
I saw that article about code from other projects? What about using keyword tests from other projects? Project Test-CSA Run IsoSpeak. (and verify Splash Windows contents for user name etc.) Select Admin Mode Select Load Project parameter C-SA Path (and verify Project Summary for C-SA) Select Project-Info (and verify experiments for C-SA ) Select Data Processing (and verify experments data processing status for C-SA) Select Export and verify Message box of export type buttons Select Score Export (and verify file browser) Enter path to export to for C-SA Project Test-I-IM Run IsoSpeak. (and verify Splash Windows contents for user name etc.) Select Admin Mode Select Load Project parameter I-IM Path (and verify Project Summary for I-IM Select Project-Info (and verify experiments for I-IM) Select Data Processing (and verify experments data processing status for I-IM) Select Export and verify Message box of export type buttons Select Score Export (and verify file browser) Enter path to export to for I-IM Select Admin Mode is a step common to both projects. A test project is for a chip type like C-SA and C- is a chip family. We have approximately 5 chip families. There is no way to have Select Admin Mode keyword test in a library project, without converting the keyword test to script?343Views0likes0Comments"Common" or "Lib" Project
Test Project Library If I have a parameterized test that is used by several other Test Projects in my TestComplete suite, should I put that test with others in its own TestComple Project that doesn't gets run directly along with other similar tests and have a "library" project of common keyword tests? For example, I have test, "Load_spkpro[spkpro_name]" where spkpro_name is the name of an app-specific project. I have test projects Test_IM_spkpro, Test_Q_VLD_spkpro, Test_Z_TT_spkpro, Test_C_SA_spkpro and so on. Each spkpro project in the application has different functionality, different graphs, different tables, different operations, and so really different tests. But before the project is loaded and even when the project is loaded and verified, there are common steps and checkpoints that work for all spkpro project types. Should Load_spkpro be in a separate project with other common tests and treated like a library that doesn't get run directly? What's the approach in TestComplete for having a library of tests that get executed from multiple projects? Let me know if any of the above is not clear and I'll elaborate some more.425Views1like3CommentsCan I get all my test names for a project into a file?
I am renaming all our tests with a different naming scheme. To give examples I would like to get all the names into a text file. Is there a way to highlight all tests and copy the names? Also, the edit process is going to be tedious. Is there a way to bypass the TestComplete user interface, and use an editor to do a bulk change?Solved976Views0likes9Comments