Implementation of working alternatives on pacman #26482
Conversation
|
Please note that this is a temporary implementation for testing a beta version of Also, here is a PR where pacman packages are collected for all testing - termux-pacman/termux-packages#211 |
Maxython
force-pushed
the
alternatives-for-pacman
branch
7 times, most recently
from
b86b6d3 to
ee8b24f
Compare
| TERMUX_PKG_SHA256=61cbd445d1381b4b184bc7c4e2791f07a79f0f2807b7c600399d0d08e8cd28cf | ||
| TERMUX_PKG_REVISION=5 | ||
| TERMUX_PKG_SRCURL=(https://gitlab.archlinux.org/pacman/pacman/-/releases/v${TERMUX_PKG_VERSION}/downloads/pacman-${TERMUX_PKG_VERSION}.tar.xz | ||
| https://github.com/termux-pacman/pacman-switch/archive/refs/heads/main.zip) |
There was a problem hiding this comment.
I'm guessing you made this refs/heads/main for testing.
In a final version of this that should be pinned to a specific commit or release tag.
There was a problem hiding this comment.
Yes, you got it right. When pacman-switch is fully ready, I'll make a pinned release.
| echo " points=(${LINK[$name]}:${ALTERNATIVE[$name]}${DEPENDENTS[$name]})" | ||
| echo "}" | ||
| done | ||
| } > "${TERMUX_PREFIX_CLASSICAL}/share/pacman-switch/$(basename "${i//.alternatives/.sw}")" |
There was a problem hiding this comment.
What's wrong with the .alternatives file extension?
You're parsing those anyway.
Why invent your own ad-hoc plaintext format?
There was a problem hiding this comment.
Why invent your own ad-hoc plaintext format?
I did this to make it easier for pacman-switch to retrieve/parse data. Switcher files (.sw) are essentially shell scripts with functions that can be run to retrieve variables with data and then manipulate them immediately.
| ## | ||
| # Parse datas from the alternatives file and put the datas into the variables. | ||
| # . | ||
| # . | ||
| # **Parameters:** | ||
| # `path_to_alternatives_file` - The path to the package's `.alternatives` file. | ||
| # . | ||
| # **Returns:** | ||
| # returns their contents in the associative arrays: | ||
| # ${LINK[@]} ${[ALTERNATIVE[@]} ${DEPENDENTS[@]} ${PRIORITY[@]} | ||
| # . | ||
| # . | ||
| # termux_alternatives__parse_alternatives_file <path_to_alternatives_file> | ||
| ## |
There was a problem hiding this comment.
I think the rewritten function comment is overly verbose and loses some of the original information.
I'm also not sure why this got moved to a new function (pseudo-)namespace.
termux_parse_alternatives or termux_parse_alternatives_file should be plenty descriptive.
There was a problem hiding this comment.
I needed to move the function to utils so that it could be used not only in termux_step_update_alternatives but also in termux_step_install_switcher_files.
Regarding the function naming and its comments/description, this is a standard for scripts in utils (at least in scripts/utils/termux/). You can ask @agnostic-apollo for more details. Also, take a look at another script with the same standard - https://github.com/termux/termux-packages/blob/master/scripts/utils/termux/package/termux_package.sh
| for name in "${NAME[@]}"; do | ||
| # Not every entry will have dependents in its group | ||
| # but we need to initialize the keys regardless | ||
| : "${DEPENDENTS[$name]:=}" | ||
| done |
There was a problem hiding this comment.
Moving this into the function is a good call since it's a concern of the function and shouldn't need to be handled at the callsite.
Maxython
force-pushed
the
alternatives-for-pacman
branch
from
ee8b24f to
45d8e20
Compare
Maxython
force-pushed
the
alternatives-for-pacman
branch
from
45d8e20 to
994c3bf
Compare
Maxython
force-pushed
the
alternatives-for-pacman
branch
from
994c3bf to
caf1ad8
Compare
Maxython
force-pushed
the
alternatives-for-pacman
branch
from
caf1ad8 to
7493c3a
Compare
4 participants
For testing