_This is a fork of yarn-deduplicate for yarn berry. _
Cleans up yarn.lock
by removing duplicates.
Yarn berry includes yarn dedupe
that currently only supports
"highest" strategy. Follow this issue: yarnpkg/berry#2297.
In the meantime, this is a fork of yarn-deduplicate that works with both highest and fewer (renamed "mostCommon") strategies as well as a new "fewerHighest" strategy which combines "mostCommon" and "highest".
yarn dlx yarn-berry-deduplicate
This will use the fewest then highest strategy to remove duplicated packages in yarn.lock
.
If you do not specify the yarn.lock path, it defaults to yarn.lock
.
Check all available options with:
yarn dlx yarn-berry-deduplicate --help
Note: you can apply multiple times deduplication and each time dedupe more. To prevent issues, try
running yarn dedupe
after each yarn-berry-deduplicate
.
yarn.lock
contains a list of all the dependencies required by your project (including transitive
dependencies), and the actual package version installed to satisfy those dependencies.
For the context of this project, a "duplicated package" is a package that appears on multiple nodes of the dependency tree with overlapping version ranges but resolved to different versions.
For example, imagine that your project directly depends on lodash
and babel
, and babel
depends
on lodash
as well. Specifically, your project depends on lodash@^1.0.0
and babel
depends on
lodash@^1.1.0
. Because how the resolution algorithm works in Yarn, you might end up with two
different copies of lodash
(for example, version 1.0.1
and 1.2.0
) in your project, even when
1.2.0
will suffice to satisfy both requirements for lodash
. That's a "duplicated package".
It is important to note that we do not consider duplicated packages when the version ranges don't
overlap. For example, if your project depends on underscore@^1.0.0
and underscore@^2.0.0
. Your
project will end up with two versions of underscore
, and yarn-deduplicate
won't change that.
When using yarn-deduplicate
remember that it will change your dependency tree. There are
certain code paths that now will run with a different set of dependencies. It is highly recommended
that you review each change to yarn.lock
. If the change is too big, use the flag --packages
to
deduplicate them gradually.
Yarn documentation seems to suggest this package shouldn't be necessary. For example, in https://classic.yarnpkg.com/en/docs/cli/dedupe/, it says
The dedupe command isn’t necessary.
yarn install
will already dedupe.
This is, however, not exactly true. There are cases where yarn will not deduplicate existing packages. For example, this scenario:
-
Install
libA
. It depends onlibB ^1.1.0
. At this point, the latest version oflibB
is1.1.2
, so it gets installed as a transitive dependency in your repo -
After a few days, install
libC
. It also depends onlibB ^1.1.0
. But this time, the latestlibB
version is1.1.3
.
In the above scenario, you'll end up with [email protected]
and [email protected]
in your repo.
Find more examples in:
--strategy <strategy>
highest
will try to use the highest installed version. For example, with the following
yarn.lock
:
library@^1.1.0:
version "1.2.0"
library@^1.2.0:
version "1.2.0"
library@^1.3.0:
version "1.3.0"
It will deduplicate library@^1.1.0
and library@^1.2.0
to 1.3.0
fewer
will try to minimize the number of installed versions by trying to deduplicate to the
version that satisfies most of the ranges first. For example, with the following yarn.lock
:
library@*:
version "2.0.0"
library@>=1.1.0:
version "3.0.0"
library@^1.2.0:
version "1.2.0"
It will deduplicate library@*
and library@>=1.1.0
to 1.2.0
.
Note that this may cause some packages to be downgraded. Be sure to check the changelogs between all versions and understand the consequences of that downgrade. If unsure, don't use this strategy.
It is not recommended to use different strategies for different packages. There is no guarantee that
the strategy will be honored in subsequent runs of yarn-deduplicate
unless the same set of flags
is specified again.
--packages <package1> <package2> <packageN>
Receives a list of packages to deduplicate. It will ignore any other duplicated package not in the
list. This option is recommended when the number of duplicated packages in yarn.lock
is too big to
be easily reviewed by a human. This will allow for a more controlled and progressive deduplication
of yarn.lock
.
--scopes <scope1> <scope2> <scopeN>
Receives a list of scopes to deduplicate. It will ignore any other duplicated package not in the
list. This option is recommended when deduplicating a large number of inter-dependent packages from
a single scope, such as @babel. This will allow for a more controlled and progressive deduplication
of yarn.lock
without specifying each package individually.
--exclude <package1> <package2> <packageN
--exclude-scopes <scope1> <scope2> <scopeN>
With these commands you can exclude certain packages/scopes from the deduplication process. This is specially useful if you want to apply a different strategy for a scope, for example.
By default, yarn-deduplicate
will only match pre-release versions if they share they share the
same major
, minor
and patch
versions (example: ^1.2.3-alpha.1
and 1.2.3-alpha.2
can be
deduplicated, but ^1.2.3
and 1.2.4-alpha.1
can't). This matches the behaviour of
semver.
To change this behaviour you can use the flag --includePrerelease
. This will treat all pre-release
versionas as if they were normal versions (^1.2.3
and 1.2.4-alpha.1
can be deduplicated).
This tool can be used as part of a CI workflow. Adding the flag --fail
will force the process to
exit with status 1 if there are duplicated packages. Example:
# Print the list of duplicated packages and exit with status 1
yarn dlx yarn-berry-deduplicate --list --fail
# Deduplicate yarn.lock and exit with status 1 if changes were required
yarn dlx yarn-berry-deduplicate --fail
Copyright (c) 2022 Sergio Cinos and others. Apache 2.0 licensed, see LICENSE.txt file.