Skip to content

Runner: distribute files round-robin by size in parallel mode #1421

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
janedbal wants to merge 3 commits into
base: 4.x
Choose a base branch
from
+36 −21
Open

Runner: distribute files round-robin by size in parallel mode #1421

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
06ac35a
Runner: distribute files round-robin by size in parallel mode
janedbal Apr 28, 2026
687fea7
Runner: disable parallel for STDIN
janedbal Apr 28, 2026
a7f9f7f
ci: re-run
janedbal Apr 28, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
57 changes: 36 additions & 21 deletions src/Runner.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@
use PHP_CodeSniffer\Files\DummyFile;
use PHP_CodeSniffer\Files\File;
use PHP_CodeSniffer\Files\FileList;
use PHP_CodeSniffer\Files\LocalFile;
use PHP_CodeSniffer\Util\Cache;
use PHP_CodeSniffer\Util\Common;
use PHP_CodeSniffer\Util\ExitCode;
Expand Down Expand Up @@ -115,8 +116,11 @@ public function runPHPCS()

// Disable caching if we are processing STDIN as we can't be 100%
// sure where the file came from or if it will change in the future.
// Also disable parallel processing because STDIN is a single file
// held in memory and the workers can't see it across the fork.
if ($this->config->stdin === true) {
$this->config->cache = false;
$this->config->cache = false;
$this->config->parallel = 1;
}

$this->run();
Expand Down Expand Up @@ -397,18 +401,37 @@ private function run()
}
} else {
// Batching and forking.
$childProcs = [];
$numPerBatch = ceil($numFiles / $this->config->parallel);
$childProcs = [];

// Distribute files round-robin in size order so each batch
// ends up with a similar total workload.
$sortedPaths = [];
$todo->rewind();
while ($todo->valid() === true) {
$sortedPaths[] = $todo->key();
$todo->next();
}

$sizes = [];
foreach ($sortedPaths as $path) {
$size = @filesize($path);
if ($size === false) {
$size = 0;
}

for ($batch = 0; $batch < $this->config->parallel; $batch++) {
$startAt = ($batch * $numPerBatch);
if ($startAt >= $numFiles) {
break;
$sizes[$path] = $size;
}

usort(
$sortedPaths,
static function ($a, $b) use ($sizes) {
return ($sizes[$b] - $sizes[$a]);
}
);

$endAt = ($startAt + $numPerBatch);
if ($endAt > $numFiles) {
$endAt = $numFiles;
for ($batch = 0; $batch < $this->config->parallel; $batch++) {
if ($batch >= $numFiles) {
break;
}

$childOutFilename = tempnam(sys_get_temp_dir(), 'phpcs-child');
Expand All @@ -418,12 +441,6 @@ private function run()
} elseif ($pid !== 0) {
$childProcs[$pid] = $childOutFilename;
} else {
// Move forward to the start of the batch.
$todo->rewind();
for ($i = 0; $i < $startAt; $i++) {
$todo->next();
}

// Reset the reporter to make sure only figures from this
// file batch are recorded.
$this->reporter->totalFiles = 0;
Expand All @@ -437,12 +454,11 @@ private function run()
// Process the files.
$pathsProcessed = [];
ob_start();
for ($i = $startAt; $i < $endAt; $i++) {
$path = $todo->key();
$file = $todo->current();
for ($i = $batch; $i < $numFiles; $i += $this->config->parallel) {
$path = $sortedPaths[$i];
$file = new LocalFile($path, $this->ruleset, $this->config);

if ($file->ignored === true) {
$todo->next();
continue;
}

Expand All @@ -458,7 +474,6 @@ private function run()
$this->processFile($file);

$pathsProcessed[] = $path;
$todo->next();
}

$debugOutput = ob_get_contents();
Expand Down
Loading