search

AppImage / appimagetool

A low-level tool to generate an AppImage from an existing AppDir
58 stars 11 forks source link

Download classic runtime by default #33

Open TheAssassin opened 11 months ago

TheAssassin commented 11 months ago

The static runtime is still considered experimental and not nearly as battle-tested as the "classic" variant from AppImageKit.

For users who wish to download the static variant instead, an environment variable was added which changes the new default behavior.

This commit allows me to use this new appimagetool in linuxdeploy-plugin-appimage, which would serve as a field test for the tool itself. I do not want to make people use the static variant until it is fully compatible with everything and the specification changes have been made.

TheAssassin commented 11 months ago

Nope. I explained the rationale very well. The vast majority of AppImages is built with regular appimagetool. This tool can't just switch to some experiment without prior notice.

It was you who wanted to have no unnecessary changes before a 1.0 release.

TheAssassin commented 11 months ago

People who change to this repo do so because they want the static runtime.

Show me. This repository has not been advertised anywhere on purpose. At least not by me. Which is in line with what we discussed recently.

probonopd commented 11 months ago

appimagetool is free to use the old appimagetool or opt out of the static runtime. Although I thought we wanted to get rid of the old dynamically linked runtime asap.

probonopd commented 11 months ago

Show me.

See the upvotes on https://github.com/AppImage/AppImageKit/issues/877#issuecomment-1588584063 - people can't wait to get the static runtime due to Ubuntu not shipping libfuse2 anymore.

TheAssassin commented 11 months ago

The old runtime is not maintained anymore, not recommended anymore.

It is of course still maintained actively, being part of the core AppImage repositories. Just because there are no changes doesn't mean that it is not maintained. Please stop spreading misleading information.

upvotes

are not a representative statistic measure in anyway. Just looking at the download counts proves you wrong.

probonopd commented 11 months ago

This PR requires discussion.

probonopd commented 8 months ago

@TheAssassin have you found a way to update AIL so that this PR is no longer needed?