This will fix if a package will be upgraded from some extras to all
extras. Previously, it will not work because some dependencies of 'all'
are already installed, so the feature 'all' is already installed.
Now, it will use a list of all extras to iterate over them separatly. This
will result it will never install all extras via `[all]`, but rather
`[foo,bar]`.
As the previous detection took the wrong values, this explorer now
checks if packages for an extra are installed or not. If not, the extra
is not installed.
Based on the information of the explorer, it will install the package
again with the absent extras.
Closes#853, see issue for full description / discussion.
Short summary:
- There was about 6.53% chances of `--renewal-hook` not being applied
- Using --automatic-renewal in one cert and not in another was an error.
- It was not possible to use different hooks for different certificates.
- FreeBSD support was utterly broken.
This is a poor implementation of optional dependencies for pip packages.
It ensures to install them if the package will be installed, but does
not take into account if they must be added/removed after the package is
already installed. Also, it will not be autoremoved, as all dependencies
will not be removed.
Previously, cdist would silently swallow the error (no invalid config file was
generated).
Reason: `set -e` does not exit if a command in a sub-command group fails,
it merely returns with a non-zero exit status.
e.g. the following snippet does not abort the script if sshd -t returns with a
non-zero exit status:
set -e
cmp -s old new || {
# check config file and update it
sshd -t -f new \
&& cat new >old
}
or compressed:
set -e
false || { false && true; }
echo $?
# prints 1
As discussed in the chat, this type now supports a broader list of OSes
which it supports backports for. Because of this, it was renamed to
something more generic. "apt" should fit in.
Because the sourced explorer can't be detected by shellcheck, it will be
completely disabled. Changing the path to /etc/os-release isn't
deterministic either.
The shellcheck wiki page suggests to use `source=/dev/null` instead of
`disable=SC1090`, but it was choosen to completely avoid that check ..
Because the function already exists, it will be used for the file to be
changed, too. Therefor, no quotes are required for that value.
The prefix and suffix match was also improved: There is no regex check
any more (the regex did checked the whole line); instead it will do it
simple.
This changes the here-document to do not interpret any shell-things. It
also single-quotes some more strings that are printed to code-remote.
Fixes#838
