Prevent aliases to be duplicated: no need to apply root aliases from
composer.json on packages currently locked, they should have their
aliases in the lock file, otherwise request an update.
This is necessary to allow the requiring of new packages which replace
packages currently locked without requiring explicitly listing them as
an argument, so simplifies the composer require command
This reduces code complexity while making partial updates more
predictable. This also allows composer require to successfully run a
partial update for a new package with transitive dependency updates.
You can specify a list of funding options each with a type and URL. The
type is used to specify the kind of funding or the platform through
which funding is possible.
The "composer install" can create the vendor/dir folders and be used as a script item on composer.json. Having another script running after it that relies on vendor/bir binaries (such as phpunit) will cause it to not find the binary. This fix addresses the issue by trying to append the path on each script iteration.
The only type of request job remaining was "install" which is really a
root requirement. The only other kind of input for the solver is now a
set of fixed packages.
Rules have been updated to account for only two kinds of former job
reason: FIXED or ROOT_REQUIRE. The job property has always been
redundant and has been removed, since reasonData suffices.
Problem reasons are always rules, so the unnecessary wrapping in an
array has been removed.
We now only ever generate a single rule per root require or fixed
package, so there is no need for the solver to special handle disabling
"jobs" anymore, the rule can just be disabled as usual.
For consistency special handling of rules for jobs in problems has been
integrated into the rule class like all other rule reasons. As part of
this change the error message for root requirements has been improved a
bit to make it clearer where the package installation request came from.
The word job has also been removed from operations, which are called
operations, not jobs.
Currently, preferred-install accepts the hash of patterns as the value in the composer.json. I've followed the same approach as used in extra and platform for letting the user define install preferences through CLI in the format: `composer config preferred-install my-organization/stable-package.dist`.
This includes two breaking changes:
- the hostname is not resolved in the case of an IP address.
- a hostname with a trailing period (FQDN) is not matched.
This brings the basic implementation in line with curl behaviour, with
the addition of full IP address and range matching (curl does not
differentiate between IP addresses host names).
The NO_PROXY environment variable can be set to either a comma-separated
list of host names that should not use a proxy, or single asterisk `*`
to match all hosts.
- Port numbers can be included by prefixing the port with a colon `:`.
- IP addresses can be used, but must be enclosed in square brackets
`[...]` if they include a port number.
- IP address ranges can specified in CIDR notation, separating the IP
address and prefix-length with a forward slash `/`.
We now output an update when only a source or dist ref changes even if
it's a stable version so in these cases the output needs to contain the
ref to make it clear to users what changed
This also changes the PRE/POST_PACKAGE_INSTALL/UPDATE/UNINSTALL events to have less information available on them, repositorySet, request and policy are gone
* github-composer/2.0: (63 commits)
Fix PSR warnings for optimized autoloader, refs #8397, refs #8403
Prepare 1.9.1 changelog
Output a hint that maybe you are not in the right directory, fixes#8404
Fix PSR warnings for optimized autoloader, refs #8397, refs #8403
Fix tests for PSR-fix in optimized autoloader, refs #8397
Fix tests for PSR-fix in optimized autoloader, refs #8397
Change PSR-fix for optimized autoloader to only warn for now, refs #8397
Fix output of dump-autoload command to avoid interfering with warnings, refs #8397
Remove credentials from git remotes in cache and vendor dirs
Avoid overwriting credentials with existing ones from git repos, refs #8293
Fix github auth to try https with pwd also, fixes#8356
Fix gitlab support for basic-auth fallback from ssh URLs
Avoid clearing the error output during removeDirectory execution, losing git error output, fixes#8351
Move test file parsing into try/catch block to avoid phpunit swallowing errors
make optimized autoloader respect PSR standards
Validate composer show with --tree and --path options set (#8390)
Don't show root warning for docker containers
Added phpdoc for ComposerAutoloaderInit$SHA1::getLoader() (#8393)
Validate schema name, type and version
Fix require command to allow working on network mounts, fixes#8231
...
The solver now only calculates a lock file transaction which does not
need to be sorted in order of dependencies. This is only necessary for
the local repo transaction generated without the solver during install
The pool builder tries to be minimal so it's fine for present/locked
packages not be assigned a solver/pool id. Adding a test to verify
correct creation of uninstall jobs
These special commands no longer (ab)use the partial update mechanism
but rather create a special install request for all current lock file
contents and later override any modified code references to the
originals. This leads to up to date remote metadata but no other
changes.
Commit: 149250ab92
ProcessExecutor::escape handled a false value inconsistently across
platforms, returning an emtpy string on Windows, otherwise `''`. This
is fixed to return `""` on Windows.
The GitDownloaderTest code has been appropriately updated.
This only removes the credentials if they are managed by composer auth.json or equivalent, if the credentials were present in the package URL to begin with they might remain
Refs #8293Fixes#3644Closes#3608
Particularly the test
tests/Composer/Test/Fixtures/installer/partial-update-downgrades-non-whitelisted-unstable.test
is interesting because it verifies that an older version will be
installed on update if the new one is only present in the installed repo
or vendor dir. This was the cause of a lot of weird edge cases and
unreliable update behavior in Composer v1
As a result some lock file packages are no longer in the pool, so the
former installed map, now present map cannot use package ids anymore
Need to revisit some more code later to simplify this, todo notes left
Ensures packages get loaded from locked repo correctly. We may not want
to support this particular use-case at all, but for now it fixes the
existing test, so we may want to revisit this later.
New approach is to use only the solved set of packages as input and then
to resolve with only the non-dev requirements and to mark everything as
dev that is not part of the result set, rather than transitioning a
temporary local repo state by uninstalling dev packages.
* github-composer/2.0: (48 commits)
Fix missing use/undefined var
Split up steps on VCS downloaders to allow doing network operations before touching the filesystem on GitDownloader, fixes#7903
Fix use statement
Deduplicate findHeaderValue code
Add install-path to the installed.json for every package, fixes#2174, closes#2424
Remove unnecessary config from phpstan
Make sure the directory exists and will not block installation later when downloading
Avoid wiping the whole target package if download of the new one fails, refs #7929
Only empty dir before actually installing packages, fixes#7929
Improve output when installing packages
Show best possible version in diagnose command
Remove extra arg
Allow path repos to point to their own source dir as install target, resulting in noop, fixes#8254
Fix use of decodeJson
Fix update mirrors to also update transport-options, fixes#7672
Fix updating or URLs to include dist type and shasum, fixes#8216
Fix origin computation
Improve handling of non-standard ports for GitLab and GitHub installs, fixes#8173
Load packages from the lock file for check-platform-reqs if no dependencies have been installed yet, fixes#8058
Fix error_handler return type declaration
...
Nils Adermann