Run the upgrader & linting tools

An initial (untested) run at a proper upgrade. Progress commit.
This commit is contained in:
Dylan Wagstaff 2017-11-23 12:56:44 +13:00
parent e6ddf0fe47
commit 1ed9a23a71
11 changed files with 750 additions and 657 deletions

View File

@ -2,62 +2,68 @@
namespace SilverStripe\ExternalLinks\Controllers; namespace SilverStripe\ExternalLinks\Controllers;
use Controller; use SilverStripe\Control\HTTP;
use HTTP; use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrackStatus;
use BrokenExternalPageTrackStatus; use SilverStripe\ExternalLinks\Jobs\CheckExternalLinksJob;
use CheckExternalLinksJob; use SilverStripe\ExternalLinks\Tasks\CheckExternalLinksTask;
use CheckExternalLinksTask; use SilverStripe\Control\Controller;
class CMSExternalLinks_Controller extends Controller
{
class CMSExternalLinks_Controller extends Controller { private static $allowed_actions = array('getJobStatus', 'start');
private static $allowed_actions = array('getJobStatus', 'start'); /*
/*
* Respond to Ajax requests for info on a running job * Respond to Ajax requests for info on a running job
* *
* @return string JSON string detailing status of the job * @return string JSON string detailing status of the job
*/ */
public function getJobStatus() { public function getJobStatus()
// Set headers {
HTTP::set_cache_age(0); // Set headers
HTTP::add_cache_headers($this->response); HTTP::set_cache_age(0);
$this->response HTTP::add_cache_headers($this->response);
->addHeader('Content-Type', 'application/json') $this->response
->addHeader('Content-Encoding', 'UTF-8') ->addHeader('Content-Type', 'application/json')
->addHeader('X-Content-Type-Options', 'nosniff'); ->addHeader('Content-Encoding', 'UTF-8')
->addHeader('X-Content-Type-Options', 'nosniff');
// Format status // Format status
$track = BrokenExternalPageTrackStatus::get_latest(); $track = BrokenExternalPageTrackStatus::get_latest();
if($track) return json_encode(array( if ($track) {
'TrackID' => $track->ID, return json_encode(array(
'Status' => $track->Status, 'TrackID' => $track->ID,
'Completed' => $track->getCompletedPages(), 'Status' => $track->Status,
'Total' => $track->getTotalPages() 'Completed' => $track->getCompletedPages(),
)); 'Total' => $track->getTotalPages()
} ));
}
}
/* /*
* Starts a broken external link check * Starts a broken external link check
*/ */
public function start() { public function start()
// return if the a job is already running {
$status = BrokenExternalPageTrackStatus::get_latest(); // return if the a job is already running
if ($status && $status->Status == 'Running') return; $status = BrokenExternalPageTrackStatus::get_latest();
if ($status && $status->Status == 'Running') {
return;
}
// Create a new job // Create a new job
if (class_exists('QueuedJobService')) { if (class_exists('QueuedJobService')) {
// Force the creation of a new run // Force the creation of a new run
BrokenExternalPageTrackStatus::create_status(); BrokenExternalPageTrackStatus::create_status();
$checkLinks = new CheckExternalLinksJob(); $checkLinks = new CheckExternalLinksJob();
singleton('QueuedJobService')->queueJob($checkLinks); singleton('QueuedJobService')->queueJob($checkLinks);
} else { } else {
//TODO this hangs as it waits for the connection to be released //TODO this hangs as it waits for the connection to be released
// should return back and continue processing // should return back and continue processing
// http://us3.php.net/manual/en/features.connection-handling.php // http://us3.php.net/manual/en/features.connection-handling.php
$task = CheckExternalLinksTask::create(); $task = CheckExternalLinksTask::create();
$task->runLinksCheck(); $task->runLinksCheck();
} }
} }
} }

View File

@ -4,38 +4,44 @@ namespace SilverStripe\ExternalLinks\Jobs;
use AbstractQueuedJob; use AbstractQueuedJob;
use QueuedJob; use QueuedJob;
use CheckExternalLinksTask;
use SilverStripe\ExternalLinks\Tasks\CheckExternalLinksTask;
if(!class_exists('AbstractQueuedJob')) return; if (!class_exists('AbstractQueuedJob')) {
return;
}
/** /**
* A Job for running a external link check for published pages * A Job for running a external link check for published pages
* *
*/ */
class CheckExternalLinksJob extends AbstractQueuedJob implements QueuedJob { class CheckExternalLinksJob extends AbstractQueuedJob implements QueuedJob
{
public function getTitle() { public function getTitle()
return _t('CheckExternalLiksJob.TITLE', 'Checking for external broken links'); {
} return _t('CheckExternalLiksJob.TITLE', 'Checking for external broken links');
}
public function getJobType() { public function getJobType()
return QueuedJob::QUEUED; {
} return QueuedJob::QUEUED;
}
public function getSignature() { public function getSignature()
return md5(get_class($this)); {
} return md5(get_class($this));
}
/**
* Check an individual page
*/
public function process() {
$task = CheckExternalLinksTask::create();
$track = $task->runLinksCheck(1);
$this->currentStep = $track->CompletedPages;
$this->totalSteps = $track->TotalPages;
$this->isComplete = $track->Status === 'Completed';
}
/**
* Check an individual page
*/
public function process()
{
$task = CheckExternalLinksTask::create();
$track = $task->runLinksCheck(1);
$this->currentStep = $track->CompletedPages;
$this->totalSteps = $track->TotalPages;
$this->isComplete = $track->Status === 'Completed';
}
} }

View File

@ -2,11 +2,13 @@
namespace SilverStripe\ExternalLinks\Model; namespace SilverStripe\ExternalLinks\Model;
use DataObject; use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrack;
use Member; use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrackStatus;
use Permission; use SilverStripe\Security\Member;
use Config; use SilverStripe\Security\Permission;
use SilverStripe\Core\Config\Config;
use SilverStripe\Control\HTTPResponse;
use SilverStripe\ORM\DataObject;
/** /**
* Represents a single link checked for a single run that is broken * Represents a single link checked for a single run that is broken
@ -14,66 +16,68 @@ use Config;
* @method BrokenExternalPageTrack Track() * @method BrokenExternalPageTrack Track()
* @method BrokenExternalPageTrackStatus Status() * @method BrokenExternalPageTrackStatus Status()
*/ */
class BrokenExternalLink extends DataObject { class BrokenExternalLink extends DataObject
{
private static $db = array( private static $db = array(
'Link' => 'Varchar(2083)', // 2083 is the maximum length of a URL in Internet Explorer. 'Link' => 'Varchar(2083)', // 2083 is the maximum length of a URL in Internet Explorer.
'HTTPCode' =>'Int' 'HTTPCode' =>'Int'
); );
private static $has_one = array( private static $has_one = array(
'Track' => 'BrokenExternalPageTrack', 'Track' => BrokenExternalPageTrack::class,
'Status' => 'BrokenExternalPageTrackStatus' 'Status' => BrokenExternalPageTrackStatus::class
); );
private static $summary_fields = array( private static $summary_fields = array(
'Created' => 'Checked', 'Created' => 'Checked',
'Link' => 'External Link', 'Link' => 'External Link',
'HTTPCodeDescription' => 'HTTP Error Code', 'HTTPCodeDescription' => 'HTTP Error Code',
'Page.Title' => 'Page link is on' 'Page.Title' => 'Page link is on'
); );
private static $searchable_fields = array( private static $searchable_fields = array(
'HTTPCode' => array('title' => 'HTTP Code') 'HTTPCode' => array('title' => 'HTTP Code')
); );
/** /**
* @return SiteTree * @return SiteTree
*/ */
public function Page() { public function Page()
return $this->Track()->Page(); {
} return $this->Track()->Page();
}
public function canEdit($member = false) { public function canEdit($member = false)
return false; {
} return false;
}
public function canView($member = false) { public function canView($member = false)
$member = $member ? $member : Member::currentUser(); {
$codes = array('content-authors', 'administrators'); $member = $member ? $member : Member::currentUser();
return Permission::checkMember($member, $codes); $codes = array('content-authors', 'administrators');
} return Permission::checkMember($member, $codes);
}
/** /**
* Retrieve a human readable description of a response code * Retrieve a human readable description of a response code
* *
* @return string * @return string
*/ */
public function getHTTPCodeDescription() { public function getHTTPCodeDescription()
$code = $this->HTTPCode; {
if(empty($code)) { $code = $this->HTTPCode;
// Assume that $code = 0 means there was no response if (empty($code)) {
$description = _t('BrokenExternalLink.NOTAVAILABLE', 'Server Not Available'); // Assume that $code = 0 means there was no response
} elseif( $description = _t('BrokenExternalLink.NOTAVAILABLE', 'Server Not Available');
($descriptions = Config::inst()->get('SS_HTTPResponse', 'status_codes')) } elseif (($descriptions = Config::inst()->get(HTTPResponse::class, 'status_codes'))
&& isset($descriptions[$code]) && isset($descriptions[$code])
) { ) {
$description = $descriptions[$code]; $description = $descriptions[$code];
} else { } else {
$description = _t('BrokenExternalLink.UNKNOWNRESPONSE', 'Unknown Response Code'); $description = _t('BrokenExternalLink.UNKNOWNRESPONSE', 'Unknown Response Code');
} }
return sprintf("%d (%s)", $code, $description); return sprintf("%d (%s)", $code, $description);
} }
} }

View File

@ -2,33 +2,37 @@
namespace SilverStripe\ExternalLinks\Model; namespace SilverStripe\ExternalLinks\Model;
use DataObject; use SilverStripe\CMS\Model\SiteTree;
use Versioned; use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrackStatus;
use SilverStripe\ExternalLinks\Model\BrokenExternalLink;
use SilverStripe\Versioned\Versioned;
use SilverStripe\ORM\DataObject;
/** /**
* Represents a track for a single page * Represents a track for a single page
*/ */
class BrokenExternalPageTrack extends DataObject { class BrokenExternalPageTrack extends DataObject
{
private static $db = array( private static $db = array(
'Processed' => 'Boolean' 'Processed' => 'Boolean'
); );
private static $has_one = array( private static $has_one = array(
'Page' => 'SiteTree', 'Page' => SiteTree::class,
'Status' => 'BrokenExternalPageTrackStatus' 'Status' => BrokenExternalPageTrackStatus::class
); );
private static $has_many = array( private static $has_many = array(
'BrokenLinks' => 'BrokenExternalLink' 'BrokenLinks' => BrokenExternalLink::class
); );
/** /**
* @return SiteTree * @return SiteTree
*/ */
public function Page() { public function Page()
return Versioned::get_by_stage('SiteTree', 'Stage') {
->byID($this->PageID); return Versioned::get_by_stage(SiteTree::class, 'Stage')
} ->byID($this->PageID);
}
} }

View File

@ -2,9 +2,11 @@
namespace SilverStripe\ExternalLinks\Model; namespace SilverStripe\ExternalLinks\Model;
use DataObject; use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrack;
use Versioned; use SilverStripe\ExternalLinks\Model\BrokenExternalLink;
use SilverStripe\CMS\Model\SiteTree;
use SilverStripe\Versioned\Versioned;
use SilverStripe\ORM\DataObject;
/** /**
* Represents the status of a track run * Represents the status of a track run
@ -14,121 +16,133 @@ use Versioned;
* @property int $TotalPages Get total pages count * @property int $TotalPages Get total pages count
* @property int $CompletedPages Get completed pages count * @property int $CompletedPages Get completed pages count
*/ */
class BrokenExternalPageTrackStatus extends DataObject { class BrokenExternalPageTrackStatus extends DataObject
{
private static $db = array( private static $db = array(
'Status' => 'Enum("Completed, Running", "Running")', 'Status' => 'Enum("Completed, Running", "Running")',
'JobInfo' => 'Varchar(255)' 'JobInfo' => 'Varchar(255)'
); );
private static $has_many = array( private static $has_many = array(
'TrackedPages' => 'BrokenExternalPageTrack', 'TrackedPages' => BrokenExternalPageTrack::class,
'BrokenLinks' => 'BrokenExternalLink' 'BrokenLinks' => BrokenExternalLink::class
); );
/** /**
* Get the latest track status * Get the latest track status
* *
* @return self * @return self
*/ */
public static function get_latest() { public static function get_latest()
return self::get() {
->sort('ID', 'DESC') return self::get()
->first(); ->sort('ID', 'DESC')
} ->first();
}
/** /**
* Gets the list of Pages yet to be checked * Gets the list of Pages yet to be checked
* *
* @return DataList * @return DataList
*/ */
public function getIncompletePageList() { public function getIncompletePageList()
$pageIDs = $this {
->getIncompleteTracks() $pageIDs = $this
->column('PageID'); ->getIncompleteTracks()
if($pageIDs) return Versioned::get_by_stage('SiteTree', 'Stage') ->column('PageID');
->byIDs($pageIDs); if ($pageIDs) {
} return Versioned::get_by_stage(SiteTree::class, 'Stage')
->byIDs($pageIDs);
}
}
/** /**
* Get the list of incomplete BrokenExternalPageTrack * Get the list of incomplete BrokenExternalPageTrack
* *
* @return DataList * @return DataList
*/ */
public function getIncompleteTracks() { public function getIncompleteTracks()
return $this {
->TrackedPages() return $this
->filter('Processed', 0); ->TrackedPages()
} ->filter('Processed', 0);
}
/** /**
* Get total pages count * Get total pages count
*/ */
public function getTotalPages() { public function getTotalPages()
return $this->TrackedPages()->count(); {
} return $this->TrackedPages()->count();
}
/** /**
* Get completed pages count * Get completed pages count
*/ */
public function getCompletedPages() { public function getCompletedPages()
return $this {
->TrackedPages() return $this
->filter('Processed', 1) ->TrackedPages()
->count(); ->filter('Processed', 1)
} ->count();
}
/** /**
* Returns the latest run, or otherwise creates a new one * Returns the latest run, or otherwise creates a new one
* *
* @return self * @return self
*/ */
public static function get_or_create() { public static function get_or_create()
// Check the current status {
$status = self::get_latest(); // Check the current status
if ($status && $status->Status == 'Running') { $status = self::get_latest();
$status->updateStatus(); if ($status && $status->Status == 'Running') {
return $status; $status->updateStatus();
} return $status;
}
return self::create_status(); return self::create_status();
} }
/* /*
* Create and prepare a new status * Create and prepare a new status
* *
* @return self * @return self
*/ */
public static function create_status() { public static function create_status()
// If the script is to be started create a new status {
$status = self::create(); // If the script is to be started create a new status
$status->updateJobInfo('Creating new tracking object'); $status = self::create();
$status->updateJobInfo('Creating new tracking object');
// Setup all pages to test // Setup all pages to test
$pageIDs = Versioned::get_by_stage('SiteTree', 'Stage') $pageIDs = Versioned::get_by_stage(SiteTree::class, 'Stage')
->column('ID'); ->column('ID');
foreach ($pageIDs as $pageID) { foreach ($pageIDs as $pageID) {
$trackPage = BrokenExternalPageTrack::create(); $trackPage = BrokenExternalPageTrack::create();
$trackPage->PageID = $pageID; $trackPage->PageID = $pageID;
$trackPage->StatusID = $status->ID; $trackPage->StatusID = $status->ID;
$trackPage->write(); $trackPage->write();
} }
return $status; return $status;
} }
public function updateJobInfo($message) { public function updateJobInfo($message)
$this->JobInfo = $message; {
$this->write(); $this->JobInfo = $message;
} $this->write();
}
/** /**
* Self check status * Self check status
*/ */
public function updateStatus() { public function updateStatus()
if ($this->CompletedPages == $this->TotalPages) { {
$this->Status = 'Completed'; if ($this->CompletedPages == $this->TotalPages) {
$this->updateJobInfo('Setting to completed'); $this->Status = 'Completed';
} $this->updateJobInfo('Setting to completed');
} }
} }
}

View File

@ -2,13 +2,12 @@
namespace SilverStripe\ExternalLinks\Reports; namespace SilverStripe\ExternalLinks\Reports;
use SS_Report; use SilverStripe\Core\Convert;
use Convert; use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrackStatus;
use BrokenExternalPageTrackStatus; use SilverStripe\ORM\ArrayList;
use ArrayList; use SilverStripe\View\Requirements;
use Requirements; use SilverStripe\Forms\LiteralField;
use LiteralField; use SilverStripe\Reports\Report;
/** /**
* Content side-report listing pages with external broken links * Content side-report listing pages with external broken links
@ -16,78 +15,86 @@ use LiteralField;
* @subpackage content * @subpackage content
*/ */
class BrokenExternalLinksReport extends SS_Report { class BrokenExternalLinksReport extends Report
{
/** /**
* Returns the report title * Returns the report title
* *
* @return string * @return string
*/ */
public function title() { public function title()
return _t('ExternalBrokenLinksReport.EXTERNALBROKENLINKS', "External broken links report"); {
} return _t('ExternalBrokenLinksReport.EXTERNALBROKENLINKS', "External broken links report");
}
public function columns() { public function columns()
return array( {
"Created" => "Checked", return array(
'Link' => array( "Created" => "Checked",
'title' => 'External Link', 'Link' => array(
'formatting' => function($value, $item) { 'title' => 'External Link',
return sprintf( 'formatting' => function ($value, $item) {
'<a target="_blank" href="%s">%s</a>', return sprintf(
Convert::raw2att($item->Link), '<a target="_blank" href="%s">%s</a>',
Convert::raw2xml($item->Link) Convert::raw2att($item->Link),
); Convert::raw2xml($item->Link)
} );
), }
'HTTPCodeDescription' => 'HTTP Error Code', ),
"Title" => array( 'HTTPCodeDescription' => 'HTTP Error Code',
"title" => 'Page link is on', "Title" => array(
'formatting' => function($value, $item) { "title" => 'Page link is on',
$page = $item->Page(); 'formatting' => function ($value, $item) {
return sprintf( $page = $item->Page();
'<a href="%s">%s</a>', return sprintf(
Convert::raw2att($page->CMSEditLink()), '<a href="%s">%s</a>',
Convert::raw2xml($page->Title) Convert::raw2att($page->CMSEditLink()),
); Convert::raw2xml($page->Title)
} );
) }
); )
} );
}
/** /**
* Alias of columns(), to support the export to csv action * Alias of columns(), to support the export to csv action
* in {@link GridFieldExportButton} generateExportFileData method. * in {@link GridFieldExportButton} generateExportFileData method.
* @return array * @return array
*/ */
public function getColumns() { public function getColumns()
return $this->columns(); {
} return $this->columns();
}
public function sourceRecords() { public function sourceRecords()
$track = BrokenExternalPageTrackStatus::get_latest(); {
if ($track) return $track->BrokenLinks(); $track = BrokenExternalPageTrackStatus::get_latest();
return new ArrayList(); if ($track) {
} return $track->BrokenLinks();
}
return new ArrayList();
}
public function getCMSFields() { public function getCMSFields()
Requirements::javascript('externallinks/javascript/BrokenExternalLinksReport.js'); {
$fields = parent::getCMSFields(); Requirements::javascript('externallinks/javascript/BrokenExternalLinksReport.js');
$fields = parent::getCMSFields();
$reportResultSpan = '</ br></ br><h3 id="ReportHolder"></h3>'; $reportResultSpan = '</ br></ br><h3 id="ReportHolder"></h3>';
$reportResult = new LiteralField('ResultTitle', $reportResultSpan); $reportResult = new LiteralField('ResultTitle', $reportResultSpan);
$fields->push($reportResult); $fields->push($reportResult);
$button = '<button id="externalLinksReport" type="button">%s</button>'; $button = '<button id="externalLinksReport" type="button">%s</button>';
$runReportButton = new LiteralField( $runReportButton = new LiteralField(
'runReport', 'runReport',
sprintf( sprintf(
$button, $button,
_t('ExternalBrokenLinksReport.RUNREPORT', 'Create new report') _t('ExternalBrokenLinksReport.RUNREPORT', 'Create new report')
) )
); );
$fields->push($runReportButton); $fields->push($runReportButton);
return $fields; return $fields;
} }
} }

View File

@ -2,200 +2,229 @@
namespace SilverStripe\ExternalLinks\Tasks; namespace SilverStripe\ExternalLinks\Tasks;
use BuildTask;
use Debug;
use BrokenExternalPageTrack;
use DOMNode; use DOMNode;
use BrokenExternalLink;
use Config;
use BrokenExternalPageTrackStatus;
use Injector;
use DB;
class CheckExternalLinksTask extends BuildTask {
private static $dependencies = array(
'LinkChecker' => '%$LinkChecker'
);
/**
* @var bool
*/
protected $silent = false;
/** use SilverStripe\Dev\Debug;
* @var LinkChecker use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrack;
*/ use SilverStripe\ExternalLinks\Model\BrokenExternalLink;
protected $linkChecker; use SilverStripe\Core\Config\Config;
use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrackStatus;
use SilverStripe\Core\Injector\Injector;
use SilverStripe\ORM\DB;
use SilverStripe\Dev\BuildTask;
protected $title = 'Checking broken External links in the SiteTree'; class CheckExternalLinksTask extends BuildTask
{
protected $description = 'A task that records external broken links in the SiteTree'; private static $dependencies = array(
'LinkChecker' => '%$LinkChecker'
);
protected $enabled = true; /**
* @var bool
*/
protected $silent = false;
/** /**
* Log a message * @var LinkChecker
* */
* @param string $message protected $linkChecker;
*/
protected function log($message) {
if(!$this->silent) Debug::message($message);
}
public function run($request) { protected $title = 'Checking broken External links in the SiteTree';
$this->runLinksCheck();
}
/**
* Turn on or off message output
*
* @param bool $silent
*/
public function setSilent($silent) {
$this->silent = $silent;
}
/** protected $description = 'A task that records external broken links in the SiteTree';
* @param LinkChecker $linkChecker
*/
public function setLinkChecker(LinkChecker $linkChecker) {
$this->linkChecker = $linkChecker;
}
/** protected $enabled = true;
* @return LinkChecker
*/
public function getLinkChecker() {
return $this->linkChecker;
}
/** /**
* Check the status of a single link on a page * Log a message
* *
* @param BrokenExternalPageTrack $pageTrack * @param string $message
* @param DOMNode $link */
*/ protected function log($message)
protected function checkPageLink(BrokenExternalPageTrack $pageTrack, DOMNode $link) { {
$class = $link->getAttribute('class'); if (!$this->silent) {
$href = $link->getAttribute('href'); Debug::message($message);
$markedBroken = preg_match('/\b(ss-broken)\b/', $class); }
}
// Check link public function run($request)
$httpCode = $this->linkChecker->checkLink($href); {
if($httpCode === null) return; // Null link means uncheckable, such as an internal link $this->runLinksCheck();
}
/**
* Turn on or off message output
*
* @param bool $silent
*/
public function setSilent($silent)
{
$this->silent = $silent;
}
// If this code is broken then mark as such /**
if($foundBroken = $this->isCodeBroken($httpCode)) { * @param LinkChecker $linkChecker
// Create broken record */
$brokenLink = new BrokenExternalLink(); public function setLinkChecker(LinkChecker $linkChecker)
$brokenLink->Link = $href; {
$brokenLink->HTTPCode = $httpCode; $this->linkChecker = $linkChecker;
$brokenLink->TrackID = $pageTrack->ID; }
$brokenLink->StatusID = $pageTrack->StatusID; // Slight denormalisation here for performance reasons
$brokenLink->write();
}
// Check if we need to update CSS class, otherwise return /**
if($markedBroken == $foundBroken) return; * @return LinkChecker
if($foundBroken) { */
$class .= ' ss-broken'; public function getLinkChecker()
} else { {
$class = preg_replace('/\s*\b(ss-broken)\b\s*/', ' ', $class); return $this->linkChecker;
} }
$link->setAttribute('class', trim($class));
}
/** /**
* Determine if the given HTTP code is "broken" * Check the status of a single link on a page
* *
* @param int $httpCode * @param BrokenExternalPageTrack $pageTrack
* @return bool True if this is a broken code * @param DOMNode $link
*/ */
protected function isCodeBroken($httpCode) { protected function checkPageLink(BrokenExternalPageTrack $pageTrack, DOMNode $link)
// Null represents no request attempted {
if($httpCode === null) return false; $class = $link->getAttribute('class');
$href = $link->getAttribute('href');
$markedBroken = preg_match('/\b(ss-broken)\b/', $class);
// do we have any whitelisted codes // Check link
$ignoreCodes = Config::inst()->get('CheckExternalLinks', 'IgnoreCodes'); $httpCode = $this->linkChecker->checkLink($href);
if(is_array($ignoreCodes) && in_array($httpCode, $ignoreCodes)) return false; if ($httpCode === null) {
return; // Null link means uncheckable, such as an internal link
}
// Check if code is outside valid range // If this code is broken then mark as such
return $httpCode < 200 || $httpCode > 302; if ($foundBroken = $this->isCodeBroken($httpCode)) {
} // Create broken record
$brokenLink = new BrokenExternalLink();
$brokenLink->Link = $href;
$brokenLink->HTTPCode = $httpCode;
$brokenLink->TrackID = $pageTrack->ID;
$brokenLink->StatusID = $pageTrack->StatusID; // Slight denormalisation here for performance reasons
$brokenLink->write();
}
/** // Check if we need to update CSS class, otherwise return
* Runs the links checker and returns the track used if ($markedBroken == $foundBroken) {
* return;
* @param int $limit Limit to number of pages to run, or null to run all }
* @return BrokenExternalPageTrackStatus if ($foundBroken) {
*/ $class .= ' ss-broken';
public function runLinksCheck($limit = null) { } else {
// Check the current status $class = preg_replace('/\s*\b(ss-broken)\b\s*/', ' ', $class);
$status = BrokenExternalPageTrackStatus::get_or_create(); }
$link->setAttribute('class', trim($class));
}
// Calculate pages to run /**
$pageTracks = $status->getIncompleteTracks(); * Determine if the given HTTP code is "broken"
if($limit) $pageTracks = $pageTracks->limit($limit); *
* @param int $httpCode
* @return bool True if this is a broken code
*/
protected function isCodeBroken($httpCode)
{
// Null represents no request attempted
if ($httpCode === null) {
return false;
}
// Check each page // do we have any whitelisted codes
foreach ($pageTracks as $pageTrack) { $ignoreCodes = Config::inst()->get('CheckExternalLinks', 'IgnoreCodes');
// Flag as complete if (is_array($ignoreCodes) && in_array($httpCode, $ignoreCodes)) {
$pageTrack->Processed = 1; return false;
$pageTrack->write(); }
// Check value of html area // Check if code is outside valid range
$page = $pageTrack->Page(); return $httpCode < 200 || $httpCode > 302;
$this->log("Checking {$page->Title}"); }
$htmlValue = Injector::inst()->create('HTMLValue', $page->Content);
if (!$htmlValue->isValid()) continue;
// Check each link /**
$links = $htmlValue->getElementsByTagName('a'); * Runs the links checker and returns the track used
foreach($links as $link) { *
$this->checkPageLink($pageTrack, $link); * @param int $limit Limit to number of pages to run, or null to run all
} * @return BrokenExternalPageTrackStatus
*/
public function runLinksCheck($limit = null)
{
// Check the current status
$status = BrokenExternalPageTrackStatus::get_or_create();
// Update content of page based on link fixes / breakages // Calculate pages to run
$htmlValue->saveHTML(); $pageTracks = $status->getIncompleteTracks();
$page->Content = $htmlValue->getContent(); if ($limit) {
$page->write(); $pageTracks = $pageTracks->limit($limit);
}
// Once all links have been created for this page update HasBrokenLinks // Check each page
$count = $pageTrack->BrokenLinks()->count(); foreach ($pageTracks as $pageTrack) {
$this->log("Found {$count} broken links"); // Flag as complete
if($count) { $pageTrack->Processed = 1;
// Bypass the ORM as syncLinkTracking does not allow you to update HasBrokenLink to true $pageTrack->write();
DB::query(sprintf(
'UPDATE "SiteTree" SET "HasBrokenLink" = 1 WHERE "ID" = \'%d\'',
intval($pageTrack->ID)
));
}
}
$status->updateJobInfo('Updating completed pages'); // Check value of html area
$status->updateStatus(); $page = $pageTrack->Page();
return $status; $this->log("Checking {$page->Title}");
} $htmlValue = Injector::inst()->create('HTMLValue', $page->Content);
if (!$htmlValue->isValid()) {
continue;
}
private function updateCompletedPages($trackID = 0) { // Check each link
$noPages = BrokenExternalPageTrack::get() $links = $htmlValue->getElementsByTagName('a');
->filter(array( foreach ($links as $link) {
'TrackID' => $trackID, $this->checkPageLink($pageTrack, $link);
'Processed' => 1 }
))
->count();
$track = BrokenExternalPageTrackStatus::get_latest();
$track->CompletedPages = $noPages;
$track->write();
return $noPages;
}
private function updateJobInfo($message) { // Update content of page based on link fixes / breakages
$track = BrokenExternalPageTrackStatus::get_latest(); $htmlValue->saveHTML();
if($track) { $page->Content = $htmlValue->getContent();
$track->JobInfo = $message; $page->write();
$track->write();
} // Once all links have been created for this page update HasBrokenLinks
} $count = $pageTrack->BrokenLinks()->count();
$this->log("Found {$count} broken links");
if ($count) {
// Bypass the ORM as syncLinkTracking does not allow you to update HasBrokenLink to true
DB::query(sprintf(
'UPDATE "SiteTree" SET "HasBrokenLink" = 1 WHERE "ID" = \'%d\'',
intval($pageTrack->ID)
));
}
}
$status->updateJobInfo('Updating completed pages');
$status->updateStatus();
return $status;
}
private function updateCompletedPages($trackID = 0)
{
$noPages = BrokenExternalPageTrack::get()
->filter(array(
'TrackID' => $trackID,
'Processed' => 1
))
->count();
$track = BrokenExternalPageTrackStatus::get_latest();
$track->CompletedPages = $noPages;
$track->write();
return $noPages;
}
private function updateJobInfo($message)
{
$track = BrokenExternalPageTrackStatus::get_latest();
if ($track) {
$track->JobInfo = $message;
$track->write();
}
}
} }

View File

@ -4,51 +4,57 @@ namespace SilverStripe\ExternalLinks\Tasks;
use SS_Cache; use SS_Cache;
/** /**
* Check links using curl * Check links using curl
*/ */
class CurlLinkChecker implements LinkChecker { class CurlLinkChecker implements LinkChecker
{
/** /**
* Return cache * Return cache
* *
* @return Zend_Cache_Frontend * @return Zend_Cache_Frontend
*/ */
protected function getCache() { protected function getCache()
return SS_Cache::factory( {
__CLASS__, return SS_Cache::factory(
'Output', __CLASS__,
array('automatic_serialization' => true) 'Output',
); array('automatic_serialization' => true)
} );
}
/** /**
* Determine the http status code for a given link * Determine the http status code for a given link
* *
* @param string $href URL to check * @param string $href URL to check
* @return int HTTP status code, or null if not checkable (not a link) * @return int HTTP status code, or null if not checkable (not a link)
*/ */
public function checkLink($href) { public function checkLink($href)
// Skip non-external links {
if(!preg_match('/^https?[^:]*:\/\//', $href)) return null; // Skip non-external links
if (!preg_match('/^https?[^:]*:\/\//', $href)) {
return null;
}
// Check if we have a cached result // Check if we have a cached result
$cacheKey = md5($href); $cacheKey = md5($href);
$result = $this->getCache()->load($cacheKey); $result = $this->getCache()->load($cacheKey);
if($result !== false) return $result; if ($result !== false) {
return $result;
}
// No cached result so just request // No cached result so just request
$handle = curl_init($href); $handle = curl_init($href);
curl_setopt($handle, CURLOPT_RETURNTRANSFER, TRUE); curl_setopt($handle, CURLOPT_RETURNTRANSFER, true);
curl_setopt($handle, CURLOPT_CONNECTTIMEOUT, 5); curl_setopt($handle, CURLOPT_CONNECTTIMEOUT, 5);
curl_setopt($handle, CURLOPT_TIMEOUT, 10); curl_setopt($handle, CURLOPT_TIMEOUT, 10);
curl_exec($handle); curl_exec($handle);
$httpCode = curl_getinfo($handle, CURLINFO_HTTP_CODE); $httpCode = curl_getinfo($handle, CURLINFO_HTTP_CODE);
curl_close($handle); curl_close($handle);
// Cache result // Cache result
$this->getCache()->save($httpCode, $cacheKey); $this->getCache()->save($httpCode, $cacheKey);
return $httpCode; return $httpCode;
} }
} }

View File

@ -2,19 +2,17 @@
namespace SilverStripe\ExternalLinks\Tasks; namespace SilverStripe\ExternalLinks\Tasks;
/** /**
* Provides an interface for checking that a link is valid * Provides an interface for checking that a link is valid
*/ */
interface LinkChecker { interface LinkChecker
{
/**
* Determine the http status code for a given link /**
* * Determine the http status code for a given link
* @param string $href URL to check *
* @return int HTTP status code, or null if not checkable (not a link) * @param string $href URL to check
*/ * @return int HTTP status code, or null if not checkable (not a link)
public function checkLink($href); */
public function checkLink($href);
} }

View File

@ -1,148 +1,165 @@
<?php <?php
class ExternalLinksTest extends SapphireTest { use SilverStripe\ExternalLinks\Tasks\LinkChecker;
use SilverStripe\Core\Injector\Injector;
use SilverStripe\ExternalLinks\Tasks\CheckExternalLinksTask;
use SilverStripe\ExternalLinks\Model\BrokenExternalPageTrackStatus;
use SilverStripe\i18n\i18n;
use SilverStripe\Reports\Report;
use SilverStripe\ExternalLinks\Reports\BrokenExternalLinksReport;
use SilverStripe\Dev\SapphireTest;
protected static $fixture_file = 'ExternalLinksTest.yml'; class ExternalLinksTest extends SapphireTest
{
protected $extraDataObjects = array( protected static $fixture_file = 'ExternalLinksTest.yml';
'ExternalLinksTestPage'
); protected $extraDataObjects = array(
'ExternalLinksTestPage'
);
protected $illegalExtensions = array( protected $illegalExtensions = array(
'SiteTree' => array('Translatable') 'SiteTree' => array('Translatable')
); );
public function setUpOnce() { public function setUpOnce()
if (class_exists('Phockito')) { {
Phockito::include_hamcrest(false); if (class_exists('Phockito')) {
} Phockito::include_hamcrest(false);
}
parent::setUpOnce(); parent::setUpOnce();
} }
public function setUp() { public function setUp()
parent::setUp(); {
parent::setUp();
// Check dependencies // Check dependencies
if (!class_exists('Phockito')) { if (!class_exists('Phockito')) {
$this->skipTest = true; $this->skipTest = true;
return $this->markTestSkipped("These tests need the Phockito module installed to run"); return $this->markTestSkipped("These tests need the Phockito module installed to run");
} }
// Mock link checker // Mock link checker
$checker = Phockito::mock('LinkChecker'); $checker = Phockito::mock(LinkChecker::class);
Phockito::when($checker) Phockito::when($checker)
->checkLink('http://www.working.com') ->checkLink('http://www.working.com')
->return(200); ->return(200);
Phockito::when($checker) Phockito::when($checker)
->checkLink('http://www.broken.com/url/thing') // 404 on working site ->checkLink('http://www.broken.com/url/thing') // 404 on working site
->return(404); ->return(404);
Phockito::when($checker) Phockito::when($checker)
->checkLink('http://www.broken.com') // 403 on working site ->checkLink('http://www.broken.com') // 403 on working site
->return(403); ->return(403);
Phockito::when($checker) Phockito::when($checker)
->checkLink('http://www.nodomain.com') // no ping ->checkLink('http://www.nodomain.com') // no ping
->return(0); ->return(0);
Phockito::when($checker) Phockito::when($checker)
->checkLink('/internal/link') ->checkLink('/internal/link')
->return(null); ->return(null);
Phockito::when($checker) Phockito::when($checker)
->checkLink('[sitetree_link,id=9999]') ->checkLink('[sitetree_link,id=9999]')
->return(null); ->return(null);
Phockito::when($checker) Phockito::when($checker)
->checkLink('home') ->checkLink('home')
->return(null); ->return(null);
Phockito::when($checker) Phockito::when($checker)
->checkLink('broken-internal') ->checkLink('broken-internal')
->return(null); ->return(null);
Phockito::when($checker) Phockito::when($checker)
->checkLink('[sitetree_link,id=1]') ->checkLink('[sitetree_link,id=1]')
->return(null); ->return(null);
Phockito::when($checker) Phockito::when($checker)
->checkLink(Hamcrest_Matchers::anything()) // anything else is 404 ->checkLink(Hamcrest_Matchers::anything()) // anything else is 404
->return(404); ->return(404);
Injector::inst()->registerService($checker, 'LinkChecker'); Injector::inst()->registerService($checker, LinkChecker::class);
} }
public function testLinks() { public function testLinks()
// Run link checker {
$task = CheckExternalLinksTask::create(); // Run link checker
$task->setSilent(true); // Be quiet during the test! $task = CheckExternalLinksTask::create();
$task->runLinksCheck(); $task->setSilent(true); // Be quiet during the test!
$task->runLinksCheck();
// Get all links checked // Get all links checked
$status = BrokenExternalPageTrackStatus::get_latest(); $status = BrokenExternalPageTrackStatus::get_latest();
$this->assertEquals('Completed', $status->Status); $this->assertEquals('Completed', $status->Status);
$this->assertEquals(5, $status->TotalPages); $this->assertEquals(5, $status->TotalPages);
$this->assertEquals(5, $status->CompletedPages); $this->assertEquals(5, $status->CompletedPages);
// Check all pages have had the correct HTML adjusted // Check all pages have had the correct HTML adjusted
for($i = 1; $i <= 5; $i++) { for ($i = 1; $i <= 5; $i++) {
$page = $this->objFromFixture('ExternalLinksTestPage', 'page'.$i); $page = $this->objFromFixture('ExternalLinksTestPage', 'page'.$i);
$this->assertNotEmpty($page->Content); $this->assertNotEmpty($page->Content);
$this->assertEquals( $this->assertEquals(
$page->ExpectedContent, $page->ExpectedContent,
$page->Content, $page->Content,
"Assert that the content of page{$i} has been updated" "Assert that the content of page{$i} has been updated"
); );
} }
// Check that the correct report of broken links is generated // Check that the correct report of broken links is generated
$links = $status $links = $status
->BrokenLinks() ->BrokenLinks()
->sort('Link'); ->sort('Link');
$this->assertEquals(4, $links->count()); $this->assertEquals(4, $links->count());
$this->assertEquals( $this->assertEquals(
array( array(
'http://www.broken.com', 'http://www.broken.com',
'http://www.broken.com/url/thing', 'http://www.broken.com/url/thing',
'http://www.broken.com/url/thing', 'http://www.broken.com/url/thing',
'http://www.nodomain.com' 'http://www.nodomain.com'
), ),
array_values($links->map('ID', 'Link')->toArray()) array_values($links->map('ID', 'Link')->toArray())
); );
// Check response codes are correct // Check response codes are correct
$expected = array( $expected = array(
'http://www.broken.com' => 403, 'http://www.broken.com' => 403,
'http://www.broken.com/url/thing' => 404, 'http://www.broken.com/url/thing' => 404,
'http://www.nodomain.com' => 0 'http://www.nodomain.com' => 0
); );
$actual = $links->map('Link', 'HTTPCode')->toArray(); $actual = $links->map('Link', 'HTTPCode')->toArray();
$this->assertEquals($expected, $actual); $this->assertEquals($expected, $actual);
// Check response descriptions are correct // Check response descriptions are correct
i18n::set_locale('en_NZ'); i18n::set_locale('en_NZ');
$expected = array( $expected = array(
'http://www.broken.com' => '403 (Forbidden)', 'http://www.broken.com' => '403 (Forbidden)',
'http://www.broken.com/url/thing' => '404 (Not Found)', 'http://www.broken.com/url/thing' => '404 (Not Found)',
'http://www.nodomain.com' => '0 (Server Not Available)' 'http://www.nodomain.com' => '0 (Server Not Available)'
); );
$actual = $links->map('Link', 'HTTPCodeDescription')->toArray(); $actual = $links->map('Link', 'HTTPCodeDescription')->toArray();
$this->assertEquals($expected, $actual); $this->assertEquals($expected, $actual);
} }
/** /**
* Test that broken links appears in the reports list * Test that broken links appears in the reports list
*/ */
public function testReportExists() { public function testReportExists()
$reports = SS_Report::get_reports(); {
$reportNames = array(); $reports = Report::get_reports();
foreach($reports as $report) { $reportNames = array();
$reportNames[] = $report->class; foreach ($reports as $report) {
} $reportNames[] = $report->class;
$this->assertContains('BrokenExternalLinksReport',$reportNames, }
'BrokenExternalLinksReport is in reports list'); $this->assertContains(
} BrokenExternalLinksReport::class,
$reportNames,
'BrokenExternalLinksReport is in reports list'
);
}
} }

View File

@ -1,5 +1,7 @@
<?php <?php
use SilverStripe\Dev\TestOnly;
class ExternalLinksTestPage extends Page implements TestOnly class ExternalLinksTestPage extends Page implements TestOnly
{ {
private static $db = array( private static $db = array(