Deserializing a big XML file is long with ride
Hello
I have a project that load and deserialize a big XML file (around 3 Go). I used to work with Visual Studio and it take 3 minutes to deserialize the data. I'm trying the exact same thing with rider but it takes more than 9 minutes.
Any idea why ?
Please sign in to leave a comment.
Hello @...,
Could you check, what MSBuild version (Use MSBuild version under Settings->Build, Execution, Deployment->Toolset and Build) are you using and the Target framework of this project?
MSBuild version is 16.0 and the Target framework is 4.5.2
In fact everything seems to be longer when I used Rider instead of VS. For example, the creation of automapper mappings takes 20s with VS but almost one minute with rider. I tried to run the project instead of debugging but it's almost the same results...
It's really strange...
Thank you for the update.
Please generate the log archive via Help->Collect Logs, use the Upload Service to send us the logs, and leave a comment here with your Upload ID.
Here we go : 2020_10_29_3JvxJLpN4wiabf8q
@... Thank you for the logs. It seems like Rider builds the project as expected and should emit the same executables as MSBuild would emit under other circumstances.
Could you check if Rider and your solution are trusted by/excluded from your AV software (either Windows Defender or other AV product)? You can refer to this guide to get the list of all directories that should be excluded from the real-time AV scans.
I already tried that yeah. That was the first thing I tried actually...
Hello @...,
Could you collect the backend snapshot (Profile Backend with Sampling) using the embedded profiler as described in this article? You can upload the snapshot using the Upload Service and leave a comment with the Upload ID in this thread or the corresponding issue on the tracker.
Thank you!
Hello
Here is the file : Upload ID "2020_11_25_5nK546EZXyt2zk3u"
Just for the record, I dit the exact same thing with VS2019, the process took less than two minutes. With Rider, it took a little bit more than 3 minutes.