Hassan_BallanChampion Level 3Joined 6 years ago435 Posts138 LikesLikes received46 SolutionsView All Badges
ContributionsMost RecentMost LikesSolutionsRe: If statement For this scenario, don’t branch on the dropdown value itself—branch on what appears in the UI after the selection. Recommended approach: Select Yes or No from the UOB dropdown Wait for the page to update (WaitProperty or short Delay) Use If Object or If…Then with a Property Checkpoint Example logic: After selecting Yes → If Object (DependentField.Exists = True) → Fill the additional fields After selecting No → Verify the dependent fields do not exist and continue Important note: When checking that an object exist, TestComplete will wait up to its default timeout (typically 10 seconds) before continuing. To avoid unnecessary delays, it’s a good practice to lower the check operation timeout to 1 second. Best keywords to use: If Object Property Checkpoint (Exists or Visible) WaitProperty This is the most reliable way to handle dynamic fields in TestComplete keyword tests. 🤖 AI-assisted response 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution Re: SessionCreator returns exitcode: 7 Failed to create user session Even with the same org policy, two Windows 11 PCs can still behave differently. GPO defines intent, not the effective state at runtime, and SessionCreator is very sensitive to that state. Common reasons one PC works while the other fails: Policy application timing / drift Updates, reboots, and logon state (locked vs unlocked) can leave machines with different effective settings. Local Security Policy differences Local overrides (secpol.msc) like Log on locally, Log on as a batch job, or hidden Deny entries can block interactive session creation on one PC only. Credential Guard / LSA protection These may be enabled by updates on one machine and not the other, breaking password-based session creation and causing exit code 7. Session state at task start If one PC has an active/unlocked session (or lingering session), SessionCreator can work. A fully locked console often fails on Win11 now. Task Scheduler execution context Tasks that look identical in the UI can differ in XML (logon type, session, SID). Quick isolation steps: Compare rsop.msc or gpresult /h outputs Check secpol.msc → User Rights Assignment Verify Credential Guard / LSA status Run query user at task time Export & diff Task Scheduler task XML Test with auto-login + unlocked session Bottom line: Windows 11 has tightened interactive session rules. One machine likely crossed a security threshold the other hasn’t yet, which explains why SessionCreator fails only on the locked PC. 🤖 AI-assisted response 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution Re: page.PagePicture - Screenshot duplication AI currently are good search tools and have good reasoning when you ask the right questions. This behavior isn’t just a random timing quirk, it's an inherent technology limitation. To confirm, you may try the documentation provided example as is in a loop to see if it fails. If it does not, the intermittent root cause would be the highly dynamic web apps "Oracle Cloud" re-rendering while scrolling. The page.PagePicture(0,0,-1,-1) works by scrolling the page and stitching viewport captures together to produce what looks like a full-page image. That approach can struggle on dynamic, virtual-scrolling content (like Oracle Cloud pages), because portions of the page are re-rendered or recycled while TestComplete is scrolling. When that happens, you can see duplicate top content and a truncated bottom instead of a complete screenshot. PagePicture Method (Page Objects) Remarks To capture an image of the whole page, call PagePicture without any parameters. The method will automatically scroll the page and combine the images of hidden areas. SmartBear’s documentation for PagePicture notes that the method automatically scrolls and combines images and that sticky elements or scrolling behavior can affect the final result. Practical suggestions that may help: Use page.Picture() to capture only the visible viewport and avoid full-page stitching entirely. Manually scroll via JavaScript (window.scrollTo(...)) and capture multiple viewports yourself. Disable browser hardware acceleration — this can reduce frame re-use that might cause repeated captures. Break the screenshot into sections and stitch yourself if a true full-page is required. Unfortunately, there’s no TestComplete setting that guarantees totally consistent full-page captures for highly dynamic web apps; this is an inherent limitation of the current scrolling-and-stitching approach. 🤖 AI-assisted response 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution Re: WebModule issues with Virtual Machine This usually happens when TestComplete intermittently fails to attach to the page DOM and instead attaches to the SmartBear browser extension context. Since Edge is Chromium-based, the extension runs under a chrome-extension://…/BackgroundPage.html URL. When this context becomes active (more common on VMs), Object Spy/Object Mapper highlights the entire browser window instead of individual web elements. Recording may still work intermittently because it uses a more tolerant capture mechanism. Match VM and host display settings: Even with DPI set to 100%, ensure the host and VM use the same screen resolution and avoid dynamically resizing the VM window, as resolution changes can cause TestComplete to misidentify web elements. Isolation steps: Try identifying elements on a simple public site like w3schools.com. If it fails there too, the issue is environment-related (VM/Edge/extension). If it works there, check whether your web app uses supported technologies Supported Web Browsers and Technologies. Close all Edge instances, reopen Edge first, fully load the page, then start TestComplete. Use a clean Edge profile with only the SmartBear extension enabled. Verify VM settings (disable enhanced graphics/3D acceleration, avoid dynamic DPI or resolution changes). Suggestions / next steps: Prefer recording + Name Mapping over Object Spy as a workaround. If possible, test in Chrome with the SmartBear extension. If the issue reproduces on public sites, raise a SmartBear support incident and include: VM type (Hyper-V/VMware/Citrix) TestComplete and Edge versions Confirmation that the URL switches to chrome-extension://… Screenshots or a short repro video 🤖 AI-assisted response 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution Re: log message issues in testcomplete Since this works on your manager’s machine but not yours, I’d recommend isolating environmental differences: Confirm the project path is identical on both stations (same local path). Create a brand-new TestComplete project on both machines and verify log navigation in both JavaScript and ScriptCPP to confirm whether the issue is ScriptCPP-specific. Validate the OS version/build on both machines (e.g., Windows 11 24H2 vs older). If they differ, raise a SmartBear support ticket, as this may be OS-related. Clear TestComplete caches on the affected machine: Close TestComplete, then delete %LOCALAPPDATA%\SmartBear\ %APPDATA%\SmartBear\ Reopen TestComplete and retest. These steps should help determine whether the issue is OS-related, cache-related, or specific to ScriptCPP rather than the project itself. 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution Re: Microsoft Access Database Engine 2016 Redistributable is no longer supported This forum is for community questions and it does not seem that you have a question. You will need to upvote existing request or submit a new feature request in the TestComplete Feature Requests | SmartBear Community if not already requested. 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution Re: execution plan properties description blanking out Merry Christmas and Happy New Year, and dhundley wish you the best on your retirement. Re: log message issues in testcomplete Also you stated messages in plural does not bring you back to the code, hinting that none of the messages, is that correct or is it only some of the messages? If some of the messages that should give you a hint to why. Re: Testcomplete closed itself It usually happens because TestComplete itself crashed, even though it may look like it just closed. In addition to checking TestComplete Logs also check Windows Event Viewer and contact support providing additional details https://support.smartbear.com/testcomplete/docs/general-info/getting-support.html 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution Re: log message issues in testcomplete If you ask ChatGPT it would provide you with a lot of reasons to why this could be expected based on your situation. 👍 Found it helpful? Click Like ✅ Issue resolved? Click Mark as Solution