SADM/pest
1
0
Fork 0
mirror of https://github.com/pestphp/pest.git synced 2026-09-05 14:23:34 +02:00
Code Issues Packages Projects Releases Wiki Activity

chore: upgrades to phpunit:13.3

Browse Source
This commit is contained in:
nuno maduro
2026-08-10 10:16:30 +01:00
parent 2e4f8ad56f
commit ac48185736
14 changed files with 526 additions and 545 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/.snapshots/success.txt
+2 -228
View File
@@ -1226,7 +1226,7 @@
✓ it does not leak mock objects between retries
✓ it does not stop retrying when snapshot changes are absent
✓ it does not leak dynamic properties between retries
✓ it clears output buffer between retries when expectOutputString is used
! it clears output buffer between retries when expectOutputString is used → Creation of dynamic property PHPUnit\Framework\TestCase\OutputBuffer::$expectedString is deprecated
✓ it preserves output between retries when no output expectation is set
WARN Tests\Features\Helpers
@@ -1568,232 +1568,6 @@
PASS Tests\Features\Tia
✓ it does not run user hooks when replaying cached skipped and incomplete results
PASS Tests\Features\Tia\BranchShapes
✓ a branch name git allows is a branch key TIA can hold with dataset "slashes"
✓ a branch name git allows is a branch key TIA can hold with dataset "dots"
✓ a branch name git allows is a branch key TIA can hold with dataset "unicode"
✓ a branch name git allows is a branch key TIA can hold with dataset "digits"
✓ a branch name git allows is a branch key TIA can hold with dataset "underscores"
✓ a branch name git allows is a branch key TIA can hold with dataset "very long"
✓ a branch differing from the default only in case gets its own key
✓ a branch that only lives on the remote keeps its baseline
✓ a branch checked out in a worktree keeps its baseline
✓ deleting many branches reclaims every one of their baselines with dataset "sequential"
✓ deleting many branches reclaims every one of their baselines with dataset "parallel"
✓ a narrowed run does not reclaim anything
✓ a detached HEAD does not reclaim anything either
✓ the default branch baseline survives every branch that comes and goes
✓ a project below the git repository root refuses to run and writes nothing with dataset "sequential"
✓ a project below the git repository root refuses to run and writes nothing with dataset "parallel"
✓ a repository with no commits says so, and leaves plain runs alone
✓ a directory with no repository at all still asks for git
PASS Tests\Features\Tia\CompleteRunWriteTier
✓ a complete run prunes a deleted test with dataset "sequential"
✓ a complete run prunes a deleted test with dataset "parallel"
✓ a complete run records nothing for a test file the graph does not know
✓ a partial run records nothing for a test file the graph does not know
✓ a truncated run does not prune with dataset "sequential"
✓ a truncated run does not prune with dataset "parallel"
✓ a green bail run is complete
✓ --no-tia refreshes results without enabling tia with dataset "sequential"
✓ --no-tia refreshes results without enabling tia with dataset "parallel"
✓ a plain run refreshes the results it executed with dataset "sequential"
✓ a plain run refreshes the results it executed with dataset "parallel"
✓ a parallel replay keeps the recorded time of tests that did not run
✓ a run that never enables tia creates no graph with dataset "plain"
✓ a run that never enables tia creates no graph with dataset "filtered"
✓ a run that never enables tia creates no graph with dataset "parallel filtered"
✓ a test edit narrows to the affected file and replays the rest
✓ a parallel run merges worker results into the parent baseline
PASS Tests\Features\Tia\CoveragePiggyback
✓ a coverage report does not found a dependency graph with dataset "pest coverage"
✓ a coverage report does not found a dependency graph with dataset "phpunit coverage report"
✓ a coverage report does not found a dependency graph with dataset "parallel"
✓ a plain run after a coverage run records the whole project scope
✓ a coverage report leaves the edges of an existing graph alone
PASS Tests\Features\Tia\DefaultBranchReplay
✓ replays the default branch baseline on a new branch
✓ replays whatever the default branch is called with ('main')
✓ replays whatever the default branch is called with ('master')
✓ replays whatever the default branch is called with ('trunk')
✓ replays whatever the default branch is called with ('develop')
✓ replays on a second new branch too
✓ writes nothing on a second run on the same branch
✓ replays on a branch whose name contains slashes
✓ replays again once back on the default branch
✓ replays on a new branch when tia is enabled by configuration
✓ a narrowed run on a new branch does not cost the fallback with dataset "sequential"
✓ a narrowed run on a new branch does not cost the fallback with dataset "parallel"
✓ replays inside a worktree on a new branch
PASS Tests\Features\Tia\DefaultBranchResolution
✓ a declared default branch beats autodetection
✓ a declared default branch that does not exist degrades to a full run
✓ a renamed default branch replays and writes under its new name
✓ the CI provider names the default branch where the checkout cannot
✓ GitLab names the default branch through its own variable
✓ a lone recorded baseline names the default branch
✓ a default branch nothing can name is refused rather than guessed
✓ an init.defaultBranch naming a branch that exists is still trusted
✓ a repository with no remote is refused rather than silently re-run
✓ a remote-less repository holding one baseline is not refused
✓ a declared default branch stands in for a missing remote
✓ tia still requires git
✓ a plain run outside a repository creates no baseline
✓ the default branch is resolved once per run, not once per test
PASS Tests\Features\Tia\DefaultBranchWriteTier
✓ narrows to the affected tests on a new branch
✓ filtered mode reads the fallback too
✓ filtered mode finds nothing to do on a clean green feature branch
✓ filtered mode falls back to a full replay when a cached failure cannot be located
✓ filtered mode finds nothing to do on the default branch itself
✓ a detached HEAD replays without minting a branch key
✓ a detached HEAD does not write into the default branch baseline with dataset "sequential"
✓ a detached HEAD does not write into the default branch baseline with dataset "parallel"
✓ the branch that ran gets its own key and the default branch keeps its baseline
✓ the fallback reaches parallel workers
PASS Tests\Features\Tia\FilteredMode
✓ re-runs a cached failure on a clean tree
✓ an explicit path turns filtered mode off
✓ filtered mode runs the whole suite when there is no baseline
✓ filtered mode finds nothing to do in parallel either
✓ a corrupt graph is reported and does not crash the run
✓ --parallel --retry is refused and leaves the graph alone
PASS Tests\Features\Tia\HostileState
✓ a graph mangled beyond use still lets the suite run with dataset "empty"
✓ a graph mangled beyond use still lets the suite run with dataset "truncated"
✓ a graph mangled beyond use still lets the suite run with dataset "not json"
✓ a graph mangled beyond use still lets the suite run with dataset "json scalar"
✓ a graph mangled beyond use still lets the suite run with dataset "json list"
✓ a graph mangled beyond use still lets the suite run with dataset "json null"
✓ a graph mangled beyond use still lets the suite run with dataset "empty object"
✓ a graph mangled beyond use still lets the suite run with dataset "nul bytes"
✓ a graph whose shape is wrong everywhere is repaired rather than trusted with dataset "sequential"
✓ a graph whose shape is wrong everywhere is repaired rather than trusted with dataset "parallel"
✓ a cached status this build cannot interpret is re-run, not replayed with dataset "below the range"
✓ a cached status this build cannot interpret is re-run, not replayed with dataset "one past the range"
✓ a cached status this build cannot interpret is re-run, not replayed with dataset "far past the range"
✓ a cached status this build cannot interpret is re-run, not replayed with dataset "huge"
✓ a cached status with no replay of its own does not fail the run with dataset "notice"
✓ a cached status with no replay of its own does not fail the run with dataset "deprecation"
✓ a cached status with no replay of its own does not fail the run with dataset "warning"
✓ a cached skip or todo replays with its message intact with dataset "skipped"
✓ a cached skip or todo replays with its message intact with dataset "incomplete"
✓ a cached failure with a multi-line message re-runs rather than replaying the text
✓ a result pointing outside the project is not addressable and widens the run
✓ an edge pointing at a file id that does not exist is ignored
✓ a graph from a schema this build does not know is rebuilt, not read
✓ graph.json being a directory does not stop the run
✓ a state dir it cannot write to still replays
PASS Tests\Features\Tia\IssueStatuses
✓ a triggered issue is recorded as itself, not as a pass with dataset "sequential" / dataset "deprecation"
✓ a triggered issue is recorded as itself, not as a pass with dataset "sequential" / dataset "notice"
✓ a triggered issue is recorded as itself, not as a pass with dataset "sequential" / dataset "warning"
✓ a triggered issue is recorded as itself, not as a pass with dataset "parallel" / dataset "deprecation"
✓ a triggered issue is recorded as itself, not as a pass with dataset "parallel" / dataset "notice"
✓ a triggered issue is recorded as itself, not as a pass with dataset "parallel" / dataset "warning"
✓ a cached deprecation still fails the run that asked to fail on one with dataset "sequential"
✓ a cached deprecation still fails the run that asked to fail on one with dataset "parallel"
✓ replaying a cached issue does not downgrade it to a pass with dataset "sequential"
✓ replaying a cached issue does not downgrade it to a pass with dataset "parallel"
✓ a failure outranks an issue triggered on the way to it
✓ a skip outranks an issue triggered on the way to it
✓ a suppressed issue is not recorded
PASS Tests\Features\Tia\LivewireComponents
✓ a changed single-file component selects only the tests that rendered it, across workers
✓ a changed multi-file component asset selects the tests that rendered the component
✓ a deleted single-file component selects the tests that rendered it
✓ a Blade file with no generated view still falls back to the watch pattern
PASS Tests\Features\Tia\PartialRunWriteTier
✓ a filtered run rewrites only the test that ran
✓ a filtered run under --tia announces that tia does not apply
✓ a test suffix narrows the tier even though every test runs
✓ a dirty run narrows to the uncommitted test edit
✓ filtered mode yields to an explicit filter
✓ an env flag narrows exactly like the option it mirrors with ('PEST_TIA')
✓ an env flag narrows exactly like the option it mirrors with ('PEST_TIA_FILTERED')
✓ a partial run does not purge the graph even with --fresh
✓ --no-tia does not stop a partial run from refreshing its own entry
✓ two partial runs each keep the other entry
✓ a shard is a partial run
✓ a parallel partial run records the test that ran, like a sequential one
PASS Tests\Features\Tia\RemoteBaseline
✓ a published baseline is fetched instead of recorded locally
✓ a fetched baseline that will not decode is discarded rather than trusted
✓ a fetched baseline recorded against another tree is not used
✓ an artifact without a graph in it fails loudly
✓ a baseline that cannot be authenticated for fails loudly
✓ a workflow or artifact that is not there fails loudly
✓ a network failure warns and lets the suite run with dataset "querying the runs"
✓ a network failure warns and lets the suite run with dataset "downloading the artifact"
✓ no published baseline yet starts a cooldown, and a corrupt cooldown does not break the run
PASS Tests\Features\Tia\SelectionPaths
✓ a committed rename selects the tests that depended on the old path with dataset "sequential"
✓ a committed rename selects the tests that depended on the old path with dataset "parallel"
✓ an affected test file that is gone does not strand a filtered run with dataset "sequential"
✓ an affected test file that is gone does not strand a filtered run with dataset "parallel"
✓ a plain run reclaims the edge of a test file that is gone
✓ a changed view selects the test that rendered it
✓ a changed partial selects the test that rendered its ancestor with dataset "direct @include"
✓ a changed partial selects the test that rendered its ancestor with dataset "transitive @include"
✓ a changed partial selects the test that rendered its ancestor with dataset "x- component"
✓ a changed partial selects the test that rendered its ancestor with dataset "include cycle"
✓ a changed Inertia page selects the test that rendered its component
✓ a changed shared JS module selects the tests of the pages that import it
✓ a changed frontend runtime file selects every Inertia test
PASS Tests\Features\Tia\StateReclamation
✓ a detached HEAD does not purge the graph on structural drift with dataset "sequential"
✓ a detached HEAD does not purge the graph on structural drift with dataset "parallel"
✓ a detached HEAD does not purge the graph with --fresh either with dataset "sequential"
✓ a detached HEAD does not purge the graph with --fresh either with dataset "parallel"
✓ a detached HEAD leaves an unreadable graph for a checkout that can rebuild it
✓ a cached failure whose test file was deleted stops widening later runs with dataset "sequential"
✓ a cached failure whose test file was deleted stops widening later runs with dataset "parallel"
✓ a complete run reclaims the entry and the edge of a deleted test file with dataset "sequential"
✓ a complete run reclaims the entry and the edge of a deleted test file with dataset "parallel"
✓ a pruned result does not come back from the fallback with dataset "sequential"
✓ a pruned result does not come back from the fallback with dataset "parallel"
✓ the fallback still reaches a branch that has never run a test file
✓ a branch that git no longer knows loses its baseline
✓ an unknown cached status is re-run rather than replayed as a failure with dataset "unknown"
✓ an unknown cached status is re-run rather than replayed as a failure with dataset "future"
✓ an unknown cached status is re-run rather than replayed as a failure with dataset "garbage"
✓ a cached notice, deprecation or warning does not replay as a failure with dataset "notice"
✓ a cached notice, deprecation or warning does not replay as a failure with dataset "deprecation"
✓ a cached notice, deprecation or warning does not replay as a failure with dataset "warning"
✓ a malformed baseline entry cannot break the run with dataset "sequential"
✓ a malformed baseline entry cannot break the run with dataset "parallel"
✓ a run torn down mid-file does not prune the tests it never reached with dataset "sequential"
✓ a run torn down mid-file does not prune the tests it never reached with dataset "parallel"
✓ a fatal error mid-file is a test error, not a truncation with dataset "sequential"
✓ a fatal error mid-file is a test error, not a truncation with dataset "parallel"
✓ a green complete run leaves the graph exactly as it found it with dataset "bail"
✓ a green complete run leaves the graph exactly as it found it with dataset "stop-on-failure"
✓ a green complete run leaves the graph exactly as it found it with dataset "compact"
✓ a green complete run leaves the graph exactly as it found it with dataset "parallel bail"
✓ a green complete run leaves the graph exactly as it found it with dataset "parallel one process"
✓ a green complete run leaves the graph exactly as it found it with dataset "parallel more processes than files"
✓ --tia --no-tia is a plain run that still refreshes what it executed with dataset "sequential"
✓ --tia --no-tia is a plain run that still refreshes what it executed with dataset "parallel"
✓ --fresh on a partial run neither purges nor prunes with dataset "sequential"
✓ --fresh on a partial run neither purges nor prunes with dataset "parallel"
✓ a second green run on a feature branch writes nothing at all with dataset "sequential"
✓ a second green run on a feature branch writes nothing at all with dataset "parallel"
✓ a graph whose recorded commit is gone is re-anchored, not warned about forever with dataset "sequential"
✓ a graph whose recorded commit is gone is re-anchored, not warned about forever with dataset "parallel"
PASS Tests\Features\Ticket
✓ it may be associated with an ticket #1, #2
✓ nested → it may be associated with an ticket #1, #4, #5, #6, #3
@@ -2435,4 +2209,4 @@
✓ pass with dataset with ('my-datas-set-value')
✓ within describe → pass with dataset with ('my-datas-set-value')
Tests: 1 deprecated, 4 warnings, 5 incomplete, 2 notices, 40 todos, 34 skipped, 1760 passed (3983 assertions)
Tests: 2 deprecated, 4 warnings, 5 incomplete, 2 notices, 40 todos, 34 skipped, 1561 passed (3405 assertions)