You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It would be more preferable if there was a process to release updated versions of tools which are low maintenance* and comparatively easier to build, such as curl
(* As opposed to something like Python, where, as I understand, the process is more involved.)
The text was updated successfully, but these errors were encountered:
There are three general types of builds in the repo, in increasing order of update efforts:
some have no minimal.diff, and these will be the easiest to update
some have a minimal.diff, the patches would need to updated when you update the version
some like python/gcc/emacs need a bit more checking when updating (actually python might become easier if we figure out an automatic script for handling the addition of packages)
What do you think is the best way to update each of the above?
For (1), I guess we could monitor the upstream packages for updates through GitHub actions and have the packages built automatically. Here, the updates should only be monitored for the top level packages that are shipped as binaries to users (e.g. monitoring updates to bash, as opposed to, say its depenency, ncurses) since users care more about them.
For (2), it may be better to upstream some of the changes if the maintainers are not opposed to their inclusion. (3) still requires manual checks for now.
It would be more preferable if there was a process to release updated versions of tools which are low maintenance* and comparatively easier to build, such as
curl
(* As opposed to something like Python, where, as I understand, the process is more involved.)
The text was updated successfully, but these errors were encountered: