using the File → Insert, i have a simple line model which inserted into ANY LO 2026 project (e.g. the plain paper tabloid template) immediately crashes and launches the report dialog. checking the windows event log, i see this:
Application: LayOut.exe
Framework Version: v4.0.30319
Description: The process was terminated due to an unhandled exception.
Exception Info: System.AccessViolationException
at .layout.ContentInsertControllerBase.InsertContentFiles(layout.ContentInsertControllerBase*, layout.Page*, layout.LayerDefinition*, std.vector<layout::ContentInsertControllerBase::InsertContentParams,std::allocatorlayout::ContentInsertControllerBase::InsertContentParams >, atlast.geometry.Point2)
at .layout.ContentInsertController.InsertContentFiles(layout.ContentInsertController*, layout.Page*, layout.LayerDefinition*, std.vector<layout::ContentInsertControllerBase::InsertContentParams,std::allocatorlayout::ContentInsertControllerBase::InsertContentParams >, atlast.geometry.Point2)
at .LOQtEntryPoint(HINSTANCE__, HINSTANCE__, Char*, Int32)
at .wWinMain(HINSTANCE__, HINSTANCE__, Char*, Int32)
at .wWinMainCRTStartup()
if i repeat this same procedure with LO 2025 and LO 2024 - no issue loading it. fwiw - the SU model “validation check” shows no errors.
also note this happens whether i’m in “user” or “admin” mode and the permissions on the source folder is full control for authenticated users.
the 2026 install, since the configuration settings are broken (i.e. they don’t let you use the directories you would like to use, only the ones the installer sets), the templates etc are all in the default folders. the tabloid plain paper template is a stock template. in the error message you’ll see an unhandled exception on the atlast.geometry - so my guess is the 26.1.256 checks are either a) indirections using the qt_metacast (and failing) or b) something else is broken or simply not properly handled.
i’m glad it’s working on your system. i am concerned though since the file inserts in 2025 and 2024 without crashing. so this strongly suggests something is amiss in 2026… and based on the number of people complaining about the crashes on opening or inserting, it’s perhaps time for the dev team to get serious about getting proper error handling in place (firstly) and secondly better testing on systems which do not have “magic” installed on them.
i’ve spent 45 years developing and architecting systems, almost 99.999% of the time improper error handling and poor qa are the main contributing factors. many times this is driven by marketing of new “features” no one asked for, or someone thought would reduce support and operation costs. the reality is, poor programming and qa driven by illogical prioritization of features without expanding the teams to build and maintain it is simply how products end up going belly up…
as a note - SU 26.2.243 does not (yet) crash on open (which 26.1.256 did) and the LO 26.2.243 allows me to insert the simple line model without crashing (so far).