Silent flag doesn't work on version 1.5.3 on Linux about grate

[erikbra]

There is a known bug in 1.5.3. I should probably pull it from all distributions. It doesn’t parse command line arguments correctly. Please use 1.5.4, the error should be fixed there (the error is related to .net 8 and more aggressive assembly trimming by default) man. 9. okt. 2023 kl. 10:09 skrev Jonas Soderholm ***@***.***

…

: *Describe the bug* The silent flag doesn't seem to have any effect when running grate from grate-linux-x64-self-contained-1.5.3.zip <https://github.com/erikbra/grate/releases/download/1.5.3/grate-linux-x64-self-contained-1.5.3.zip>. It works in grate-linux-x64-self-contained-1.5.2.zip <https://github.com/erikbra/grate/releases/download/1.5.2/grate-linux-x64-self-contained-1.5.2.zip> . *To Reproduce* 1. Open in ubuntu in WSL. 2. Copy the grate file from grate-linux-x64-self-contained-1.5.3.zip <https://github.com/erikbra/grate/releases/download/1.5.3/grate-linux-x64-self-contained-1.5.3.zip> into an empty folder and cd into it. 3. Run chmod +rwx grate. 4. Run ./grate -cs myconnection 5. You will see: Initializing connections. Running grate v1.5.3.0 against - . Looking in /home/xxx/grate for scripts to run. Please press enter when ready to kick... Try the same with grate-linux-x64-self-contained-1.5.2.zip <https://github.com/erikbra/grate/releases/download/1.5.2/grate-linux-x64-self-contained-1.5.2.zip> and it will run the script without the "Press enter prompt". *Expected behavior* The press enter prompt should not show with the silent flag *Screenshots* *Desktop (please complete the following information):* - OS: Ubuntu on WSL 2 in Windows 10 - Version: Grate version 1.5.3 *Additional context* I haven't tested if this is a problem on Windows. I wonder if the .Net 8 upgrade might have anything to do with this? — Reply to this email directly, view it on GitHub <#383>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAMNWQW2WOJLU6K3BSVTYBLX6OWKXAVCNFSM6AAAAAA5YMVBDWVHI2DSMVQWIX3LMV43ASLTON2WKOZRHEZTENJSGI2DMMY> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[wokket]

Just a thought for next time, we might want to consider taking a major version bump when we upgrade .Net versions commensurate with the likelyhood of bugs.

[magnusakselvoll]

@erikbra, it seems that version 1.5.4 is not released, at least not on the release page. Currently this bug affects the current version released.

[erikbra]

@magnusakselvoll - Sorry, I had stopped in the middle of the release process (probably got lost in too many open browser tabs), but it properly out there since 1. of Nov.

[erikbra]

Just a thought for next time, we might want to consider taking a major version bump when we upgrade .Net versions commensurate with the likelyhood of bugs.

Chicken!

No, sorry, just kidding! :) I partly agree. Or, at least test better. I don't know it changing .net versions should justify a major version number upgrade, as we compile it self-contained, and people shouldn't notice (if I test properly).

But, yes, I see your point.

I have it on my checklist to re-write the tests so that they run on the final, compiled, and trimmed product, calling it from the command-line, instead of instantiating the classes directly. This would make the tests catch these types of errors as well

[erikbra]

@mk0sojo Closing this, as the error is fixed in the newest version