bundler-site
bundler-site copied to clipboard
rubygems
Documentation freqently contradicts itself regarding remembered options
From https://bundler.io/v1.16/man/bundle-config.1.html#REMEMBERING-OPTIONS:
Flags passed to bundle install or the Bundler runtime, such as --path foo or --without production, are not remembered between commands.
https://bundler.io/v1.16/man/bundle-config.1.html#CONFIGURATION-KEYS
For instance, passing the --without flag to bundle install(1) prevents Bundler from installing certain groups specified in the Gemfile(5). [...] Additionally, subsequent calls to bundle install(1) remember this setting and skip those groups.
https://bundler.io/v1.16/man/bundle-install.1.html#OPTIONS:
--standalone[=<list>][...]
Using this option implicitly sets path, which is a [remembered option][REMEMBERED OPTIONS].
--with=<list>[...] If a group is given that is in the remembered list of groups given to --without, it is removed from that list.
--without=<list>[...] If a group is given that is in the remembered list of groups given to --with, it is removed from that list.
There may be more, but those are a few examples I observed.
From the behavior I've personally observed in Bundler, remembered options do indeed exist and both --deployment and --with are among them.
Yeah, options are remembered in Bundler 1, but in Bundler 2 they will not be -- this might be explain why they're inconsistent -- we might have gotten ahead of ourselves ;)
@segiddins do we think there's enough of an inconsistency that we should call it out in the docs? I wonder where we think we should add a note or something to make it clear to readers that options are remembered in Bundler 1.xx?
The problem is that the docs were writing are about Bundler 2, my guess is we accidentally bavkported something we shouldn’t have
@segiddins should we close this out? If we need to remove anything from v1.16, though, lmk!
The contradictions still exist in the v1.16 docs, so unless that documentation is no longer supported I wouldn't recommend closing this issue.
Sounds good @Ajedi32! We're in the process of finding a fix. Thanks for the quick response