@@ -27,178 +27,178 @@ |
||
27 | 27 | * |
28 | 28 | */ |
29 | 29 | class tx_crawler_domain_process_manager { |
30 | - /** |
|
31 | - * @var $timeToLive integer |
|
32 | - */ |
|
33 | - private $timeToLive; |
|
34 | - /** |
|
35 | - * @var integer |
|
36 | - */ |
|
37 | - private $countInARun; |
|
30 | + /** |
|
31 | + * @var $timeToLive integer |
|
32 | + */ |
|
33 | + private $timeToLive; |
|
34 | + /** |
|
35 | + * @var integer |
|
36 | + */ |
|
37 | + private $countInARun; |
|
38 | 38 | |
39 | - /** |
|
40 | - * @var integer |
|
41 | - */ |
|
42 | - private $processLimit; |
|
39 | + /** |
|
40 | + * @var integer |
|
41 | + */ |
|
42 | + private $processLimit; |
|
43 | 43 | |
44 | - /** |
|
45 | - * @var $crawlerObj tx_crawler_lib |
|
46 | - */ |
|
47 | - private $crawlerObj; |
|
44 | + /** |
|
45 | + * @var $crawlerObj tx_crawler_lib |
|
46 | + */ |
|
47 | + private $crawlerObj; |
|
48 | 48 | |
49 | - /** |
|
50 | - * @var $queueRepository tx_crawler_domain_queue_repository |
|
51 | - */ |
|
52 | - private $queueRepository; |
|
49 | + /** |
|
50 | + * @var $queueRepository tx_crawler_domain_queue_repository |
|
51 | + */ |
|
52 | + private $queueRepository; |
|
53 | 53 | |
54 | - /** |
|
55 | - * @var tx_crawler_domain_process_repository |
|
56 | - */ |
|
57 | - private $processRepository; |
|
54 | + /** |
|
55 | + * @var tx_crawler_domain_process_repository |
|
56 | + */ |
|
57 | + private $processRepository; |
|
58 | 58 | |
59 | - /** |
|
60 | - * @var $verbose boolean |
|
61 | - */ |
|
62 | - private $verbose; |
|
59 | + /** |
|
60 | + * @var $verbose boolean |
|
61 | + */ |
|
62 | + private $verbose; |
|
63 | 63 | |
64 | - /** |
|
65 | - * the constructor |
|
66 | - */ |
|
67 | - public function __construct() { |
|
68 | - $this->processRepository = new tx_crawler_domain_process_repository(); |
|
69 | - $this->queueRepository = new tx_crawler_domain_queue_repository(); |
|
70 | - $this->crawlerObj = \TYPO3\CMS\Core\Utility\GeneralUtility::makeInstance('tx_crawler_lib'); |
|
71 | - $this->timeToLive = intval($this->crawlerObj->extensionSettings['processMaxRunTime']); |
|
72 | - $this->countInARun = intval($this->crawlerObj->extensionSettings['countInARun']); |
|
73 | - $this->processLimit = intval($this->crawlerObj->extensionSettings['processLimit']); |
|
74 | - $this->verbose = intval($this->crawlerObj->extensionSettings['processVerbose']); |
|
75 | - } |
|
64 | + /** |
|
65 | + * the constructor |
|
66 | + */ |
|
67 | + public function __construct() { |
|
68 | + $this->processRepository = new tx_crawler_domain_process_repository(); |
|
69 | + $this->queueRepository = new tx_crawler_domain_queue_repository(); |
|
70 | + $this->crawlerObj = \TYPO3\CMS\Core\Utility\GeneralUtility::makeInstance('tx_crawler_lib'); |
|
71 | + $this->timeToLive = intval($this->crawlerObj->extensionSettings['processMaxRunTime']); |
|
72 | + $this->countInARun = intval($this->crawlerObj->extensionSettings['countInARun']); |
|
73 | + $this->processLimit = intval($this->crawlerObj->extensionSettings['processLimit']); |
|
74 | + $this->verbose = intval($this->crawlerObj->extensionSettings['processVerbose']); |
|
75 | + } |
|
76 | 76 | |
77 | - /** |
|
78 | - * starts multiple processes |
|
79 | - * |
|
80 | - * @param integer $timeout |
|
81 | - */ |
|
82 | - public function multiProcess($timeout) { |
|
83 | - $this->verbose = true; // DEBUG |
|
84 | - $timeout = 300; // DEBUG |
|
77 | + /** |
|
78 | + * starts multiple processes |
|
79 | + * |
|
80 | + * @param integer $timeout |
|
81 | + */ |
|
82 | + public function multiProcess($timeout) { |
|
83 | + $this->verbose = true; // DEBUG |
|
84 | + $timeout = 300; // DEBUG |
|
85 | 85 | |
86 | - if ($this->processLimit <= 1) { |
|
87 | - throw new RuntimeException('To run crawler in multi process mode you have to configure the processLimit > 1.' . PHP_EOL); |
|
88 | - } |
|
86 | + if ($this->processLimit <= 1) { |
|
87 | + throw new RuntimeException('To run crawler in multi process mode you have to configure the processLimit > 1.' . PHP_EOL); |
|
88 | + } |
|
89 | 89 | |
90 | - $pendingItemsStart = $this->queueRepository->countAllPendingItems(); |
|
91 | - $itemReportLimit = 20; |
|
92 | - $reportItemCount = $pendingItemsStart - $itemReportLimit; |
|
93 | - if ($this->verbose) { |
|
94 | - $this->reportItemStatus(); |
|
95 | - } |
|
96 | - $this->startRequiredProcesses(); |
|
97 | - $nextTimeOut = time() + $this->timeToLive; |
|
98 | - for ($i = 0; $i < $timeout; $i++) { |
|
99 | - $currentPendingItems = $this->queueRepository->countAllPendingItems(); |
|
100 | - if ($this->startRequiredProcesses($this->verbose)) { |
|
101 | - $nextTimeOut = time() + $this->timeToLive; |
|
102 | - } |
|
103 | - if ($currentPendingItems == 0) { |
|
104 | - if ($this->verbose) { |
|
105 | - echo 'Finished...' . chr(10); |
|
106 | - } |
|
107 | - break; |
|
108 | - } |
|
109 | - if ($currentPendingItems < $reportItemCount) { |
|
110 | - if ($this->verbose) { |
|
111 | - $this->reportItemStatus(); |
|
112 | - } |
|
113 | - $reportItemCount = $currentPendingItems - $itemReportLimit; |
|
114 | - } |
|
115 | - sleep(1); |
|
116 | - if ($nextTimeOut < time()) { |
|
117 | - $timedOutProcesses = $this->processRepository->findAll('', 'DESC', NULL, 0, 'ttl >' . $nextTimeOut); |
|
118 | - $nextTimeOut = time() + $this->timeToLive; |
|
119 | - if ($this->verbose) { |
|
120 | - echo 'Cleanup' . implode(',', $timedOutProcesses->getProcessIds()) . chr(10); |
|
121 | - } |
|
122 | - $this->crawlerObj->CLI_releaseProcesses($timedOutProcesses->getProcessIds(), true); |
|
123 | - } |
|
124 | - } |
|
125 | - if ($currentPendingItems > 0 && $this->verbose) { |
|
126 | - echo 'Stop with timeout' . chr(10); |
|
127 | - } |
|
128 | - } |
|
90 | + $pendingItemsStart = $this->queueRepository->countAllPendingItems(); |
|
91 | + $itemReportLimit = 20; |
|
92 | + $reportItemCount = $pendingItemsStart - $itemReportLimit; |
|
93 | + if ($this->verbose) { |
|
94 | + $this->reportItemStatus(); |
|
95 | + } |
|
96 | + $this->startRequiredProcesses(); |
|
97 | + $nextTimeOut = time() + $this->timeToLive; |
|
98 | + for ($i = 0; $i < $timeout; $i++) { |
|
99 | + $currentPendingItems = $this->queueRepository->countAllPendingItems(); |
|
100 | + if ($this->startRequiredProcesses($this->verbose)) { |
|
101 | + $nextTimeOut = time() + $this->timeToLive; |
|
102 | + } |
|
103 | + if ($currentPendingItems == 0) { |
|
104 | + if ($this->verbose) { |
|
105 | + echo 'Finished...' . chr(10); |
|
106 | + } |
|
107 | + break; |
|
108 | + } |
|
109 | + if ($currentPendingItems < $reportItemCount) { |
|
110 | + if ($this->verbose) { |
|
111 | + $this->reportItemStatus(); |
|
112 | + } |
|
113 | + $reportItemCount = $currentPendingItems - $itemReportLimit; |
|
114 | + } |
|
115 | + sleep(1); |
|
116 | + if ($nextTimeOut < time()) { |
|
117 | + $timedOutProcesses = $this->processRepository->findAll('', 'DESC', NULL, 0, 'ttl >' . $nextTimeOut); |
|
118 | + $nextTimeOut = time() + $this->timeToLive; |
|
119 | + if ($this->verbose) { |
|
120 | + echo 'Cleanup' . implode(',', $timedOutProcesses->getProcessIds()) . chr(10); |
|
121 | + } |
|
122 | + $this->crawlerObj->CLI_releaseProcesses($timedOutProcesses->getProcessIds(), true); |
|
123 | + } |
|
124 | + } |
|
125 | + if ($currentPendingItems > 0 && $this->verbose) { |
|
126 | + echo 'Stop with timeout' . chr(10); |
|
127 | + } |
|
128 | + } |
|
129 | 129 | |
130 | - /** |
|
131 | - * Reports curent Status of queue |
|
132 | - */ |
|
133 | - protected function reportItemStatus() { |
|
134 | - echo 'Pending:'.$this->queueRepository->countAllPendingItems().' / Assigned:'.$this->queueRepository->countAllAssignedPendingItems().chr(10); |
|
135 | - } |
|
130 | + /** |
|
131 | + * Reports curent Status of queue |
|
132 | + */ |
|
133 | + protected function reportItemStatus() { |
|
134 | + echo 'Pending:'.$this->queueRepository->countAllPendingItems().' / Assigned:'.$this->queueRepository->countAllAssignedPendingItems().chr(10); |
|
135 | + } |
|
136 | 136 | |
137 | - /** |
|
138 | - * according to the given count of pending items and the countInARun Setting this method |
|
139 | - * starts more crawling processes |
|
140 | - * @return boolean if processes are started |
|
141 | - */ |
|
142 | - private function startRequiredProcesses() { |
|
143 | - $ret = FALSE; |
|
144 | - $currentProcesses= $this->processRepository->countActive(); |
|
145 | - $availableProcessesCount = $this->processLimit-$currentProcesses; |
|
146 | - $requiredProcessesCount = ceil($this->queueRepository->countAllUnassignedPendingItems() / $this->countInARun); |
|
147 | - $startProcessCount = min(array($availableProcessesCount,$requiredProcessesCount)); |
|
148 | - if ($startProcessCount <= 0) { |
|
149 | - return $ret; |
|
150 | - } |
|
151 | - if ($startProcessCount && $this->verbose) { |
|
152 | - echo 'Start '.$startProcessCount.' new processes (Running:'.$currentProcesses.')'; |
|
153 | - } |
|
154 | - for($i=0;$i<$startProcessCount;$i++) { |
|
155 | - usleep(100); |
|
156 | - if ($this->startProcess()) { |
|
157 | - if ($this->verbose) { |
|
158 | - echo '.'; |
|
159 | - $ret = TRUE; |
|
160 | - } |
|
161 | - } |
|
162 | - } |
|
163 | - if ($this->verbose) { |
|
164 | - echo chr(10); |
|
165 | - } |
|
166 | - return $ret; |
|
167 | - } |
|
137 | + /** |
|
138 | + * according to the given count of pending items and the countInARun Setting this method |
|
139 | + * starts more crawling processes |
|
140 | + * @return boolean if processes are started |
|
141 | + */ |
|
142 | + private function startRequiredProcesses() { |
|
143 | + $ret = FALSE; |
|
144 | + $currentProcesses= $this->processRepository->countActive(); |
|
145 | + $availableProcessesCount = $this->processLimit-$currentProcesses; |
|
146 | + $requiredProcessesCount = ceil($this->queueRepository->countAllUnassignedPendingItems() / $this->countInARun); |
|
147 | + $startProcessCount = min(array($availableProcessesCount,$requiredProcessesCount)); |
|
148 | + if ($startProcessCount <= 0) { |
|
149 | + return $ret; |
|
150 | + } |
|
151 | + if ($startProcessCount && $this->verbose) { |
|
152 | + echo 'Start '.$startProcessCount.' new processes (Running:'.$currentProcesses.')'; |
|
153 | + } |
|
154 | + for($i=0;$i<$startProcessCount;$i++) { |
|
155 | + usleep(100); |
|
156 | + if ($this->startProcess()) { |
|
157 | + if ($this->verbose) { |
|
158 | + echo '.'; |
|
159 | + $ret = TRUE; |
|
160 | + } |
|
161 | + } |
|
162 | + } |
|
163 | + if ($this->verbose) { |
|
164 | + echo chr(10); |
|
165 | + } |
|
166 | + return $ret; |
|
167 | + } |
|
168 | 168 | |
169 | - /** |
|
170 | - * starts new process |
|
171 | - * @throws Exception if no crawlerprocess was started |
|
172 | - */ |
|
173 | - public function startProcess() { |
|
174 | - $ttl = (time() + $this->timeToLive -1); |
|
175 | - $current = $this->processRepository->countNotTimeouted($ttl); |
|
176 | - $completePath = '(' .escapeshellcmd($this->getCrawlerCliPath()) . ' &) > /dev/null'; |
|
177 | - if (system($completePath) === FALSE) { |
|
178 | - throw new Exception('could not start process!'); |
|
179 | - } |
|
180 | - else { |
|
181 | - for ($i=0;$i<10;$i++) { |
|
182 | - if ($this->processRepository->countNotTimeouted($ttl) > $current) { |
|
183 | - return true; |
|
184 | - } |
|
185 | - sleep(1); |
|
186 | - } |
|
187 | - throw new Exception('Something went wrong: process did not appear within 10 seconds.'); |
|
188 | - } |
|
189 | - } |
|
169 | + /** |
|
170 | + * starts new process |
|
171 | + * @throws Exception if no crawlerprocess was started |
|
172 | + */ |
|
173 | + public function startProcess() { |
|
174 | + $ttl = (time() + $this->timeToLive -1); |
|
175 | + $current = $this->processRepository->countNotTimeouted($ttl); |
|
176 | + $completePath = '(' .escapeshellcmd($this->getCrawlerCliPath()) . ' &) > /dev/null'; |
|
177 | + if (system($completePath) === FALSE) { |
|
178 | + throw new Exception('could not start process!'); |
|
179 | + } |
|
180 | + else { |
|
181 | + for ($i=0;$i<10;$i++) { |
|
182 | + if ($this->processRepository->countNotTimeouted($ttl) > $current) { |
|
183 | + return true; |
|
184 | + } |
|
185 | + sleep(1); |
|
186 | + } |
|
187 | + throw new Exception('Something went wrong: process did not appear within 10 seconds.'); |
|
188 | + } |
|
189 | + } |
|
190 | 190 | |
191 | - /** |
|
192 | - * Returns the path to start the crawler from the command line |
|
193 | - * |
|
194 | - * @return string |
|
195 | - */ |
|
196 | - public function getCrawlerCliPath(){ |
|
197 | - $phpPath = $this->crawlerObj->extensionSettings['phpPath'] . ' '; |
|
198 | - $pathToTypo3 = rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_DOCUMENT_ROOT'), '/'); |
|
199 | - $pathToTypo3 .= rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_SITE_PATH'), '/'); |
|
200 | - $cliPart = '/typo3/cli_dispatch.phpsh crawler'; |
|
201 | - return $phpPath.$pathToTypo3.$cliPart; |
|
202 | - } |
|
191 | + /** |
|
192 | + * Returns the path to start the crawler from the command line |
|
193 | + * |
|
194 | + * @return string |
|
195 | + */ |
|
196 | + public function getCrawlerCliPath(){ |
|
197 | + $phpPath = $this->crawlerObj->extensionSettings['phpPath'] . ' '; |
|
198 | + $pathToTypo3 = rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_DOCUMENT_ROOT'), '/'); |
|
199 | + $pathToTypo3 .= rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_SITE_PATH'), '/'); |
|
200 | + $cliPart = '/typo3/cli_dispatch.phpsh crawler'; |
|
201 | + return $phpPath.$pathToTypo3.$cliPart; |
|
202 | + } |
|
203 | 203 | |
204 | 204 | } |
@@ -26,7 +26,7 @@ discard block |
||
26 | 26 | * Manages cralwer processes and can be used to start a new process or multiple processes |
27 | 27 | * |
28 | 28 | */ |
29 | -class tx_crawler_domain_process_manager { |
|
29 | +class tx_crawler_domain_process_manager { |
|
30 | 30 | /** |
31 | 31 | * @var $timeToLive integer |
32 | 32 | */ |
@@ -65,8 +65,8 @@ discard block |
||
65 | 65 | * the constructor |
66 | 66 | */ |
67 | 67 | public function __construct() { |
68 | - $this->processRepository = new tx_crawler_domain_process_repository(); |
|
69 | - $this->queueRepository = new tx_crawler_domain_queue_repository(); |
|
68 | + $this->processRepository = new tx_crawler_domain_process_repository(); |
|
69 | + $this->queueRepository = new tx_crawler_domain_queue_repository(); |
|
70 | 70 | $this->crawlerObj = \TYPO3\CMS\Core\Utility\GeneralUtility::makeInstance('tx_crawler_lib'); |
71 | 71 | $this->timeToLive = intval($this->crawlerObj->extensionSettings['processMaxRunTime']); |
72 | 72 | $this->countInARun = intval($this->crawlerObj->extensionSettings['countInARun']); |
@@ -84,7 +84,7 @@ discard block |
||
84 | 84 | $timeout = 300; // DEBUG |
85 | 85 | |
86 | 86 | if ($this->processLimit <= 1) { |
87 | - throw new RuntimeException('To run crawler in multi process mode you have to configure the processLimit > 1.' . PHP_EOL); |
|
87 | + throw new RuntimeException('To run crawler in multi process mode you have to configure the processLimit > 1.'.PHP_EOL); |
|
88 | 88 | } |
89 | 89 | |
90 | 90 | $pendingItemsStart = $this->queueRepository->countAllPendingItems(); |
@@ -102,7 +102,7 @@ discard block |
||
102 | 102 | } |
103 | 103 | if ($currentPendingItems == 0) { |
104 | 104 | if ($this->verbose) { |
105 | - echo 'Finished...' . chr(10); |
|
105 | + echo 'Finished...'.chr(10); |
|
106 | 106 | } |
107 | 107 | break; |
108 | 108 | } |
@@ -114,16 +114,16 @@ discard block |
||
114 | 114 | } |
115 | 115 | sleep(1); |
116 | 116 | if ($nextTimeOut < time()) { |
117 | - $timedOutProcesses = $this->processRepository->findAll('', 'DESC', NULL, 0, 'ttl >' . $nextTimeOut); |
|
117 | + $timedOutProcesses = $this->processRepository->findAll('', 'DESC', NULL, 0, 'ttl >'.$nextTimeOut); |
|
118 | 118 | $nextTimeOut = time() + $this->timeToLive; |
119 | 119 | if ($this->verbose) { |
120 | - echo 'Cleanup' . implode(',', $timedOutProcesses->getProcessIds()) . chr(10); |
|
120 | + echo 'Cleanup'.implode(',', $timedOutProcesses->getProcessIds()).chr(10); |
|
121 | 121 | } |
122 | 122 | $this->crawlerObj->CLI_releaseProcesses($timedOutProcesses->getProcessIds(), true); |
123 | 123 | } |
124 | 124 | } |
125 | 125 | if ($currentPendingItems > 0 && $this->verbose) { |
126 | - echo 'Stop with timeout' . chr(10); |
|
126 | + echo 'Stop with timeout'.chr(10); |
|
127 | 127 | } |
128 | 128 | } |
129 | 129 | |
@@ -141,17 +141,17 @@ discard block |
||
141 | 141 | */ |
142 | 142 | private function startRequiredProcesses() { |
143 | 143 | $ret = FALSE; |
144 | - $currentProcesses= $this->processRepository->countActive(); |
|
145 | - $availableProcessesCount = $this->processLimit-$currentProcesses; |
|
144 | + $currentProcesses = $this->processRepository->countActive(); |
|
145 | + $availableProcessesCount = $this->processLimit - $currentProcesses; |
|
146 | 146 | $requiredProcessesCount = ceil($this->queueRepository->countAllUnassignedPendingItems() / $this->countInARun); |
147 | - $startProcessCount = min(array($availableProcessesCount,$requiredProcessesCount)); |
|
147 | + $startProcessCount = min(array($availableProcessesCount, $requiredProcessesCount)); |
|
148 | 148 | if ($startProcessCount <= 0) { |
149 | 149 | return $ret; |
150 | 150 | } |
151 | 151 | if ($startProcessCount && $this->verbose) { |
152 | 152 | echo 'Start '.$startProcessCount.' new processes (Running:'.$currentProcesses.')'; |
153 | 153 | } |
154 | - for($i=0;$i<$startProcessCount;$i++) { |
|
154 | + for ($i = 0; $i < $startProcessCount; $i++) { |
|
155 | 155 | usleep(100); |
156 | 156 | if ($this->startProcess()) { |
157 | 157 | if ($this->verbose) { |
@@ -171,14 +171,14 @@ discard block |
||
171 | 171 | * @throws Exception if no crawlerprocess was started |
172 | 172 | */ |
173 | 173 | public function startProcess() { |
174 | - $ttl = (time() + $this->timeToLive -1); |
|
174 | + $ttl = (time() + $this->timeToLive - 1); |
|
175 | 175 | $current = $this->processRepository->countNotTimeouted($ttl); |
176 | - $completePath = '(' .escapeshellcmd($this->getCrawlerCliPath()) . ' &) > /dev/null'; |
|
176 | + $completePath = '('.escapeshellcmd($this->getCrawlerCliPath()).' &) > /dev/null'; |
|
177 | 177 | if (system($completePath) === FALSE) { |
178 | 178 | throw new Exception('could not start process!'); |
179 | 179 | } |
180 | 180 | else { |
181 | - for ($i=0;$i<10;$i++) { |
|
181 | + for ($i = 0; $i < 10; $i++) { |
|
182 | 182 | if ($this->processRepository->countNotTimeouted($ttl) > $current) { |
183 | 183 | return true; |
184 | 184 | } |
@@ -193,10 +193,10 @@ discard block |
||
193 | 193 | * |
194 | 194 | * @return string |
195 | 195 | */ |
196 | - public function getCrawlerCliPath(){ |
|
197 | - $phpPath = $this->crawlerObj->extensionSettings['phpPath'] . ' '; |
|
198 | - $pathToTypo3 = rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_DOCUMENT_ROOT'), '/'); |
|
199 | - $pathToTypo3 .= rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_SITE_PATH'), '/'); |
|
196 | + public function getCrawlerCliPath() { |
|
197 | + $phpPath = $this->crawlerObj->extensionSettings['phpPath'].' '; |
|
198 | + $pathToTypo3 = rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_DOCUMENT_ROOT'), '/'); |
|
199 | + $pathToTypo3 .= rtrim(\TYPO3\CMS\Core\Utility\GeneralUtility::getIndpEnv('TYPO3_SITE_PATH'), '/'); |
|
200 | 200 | $cliPart = '/typo3/cli_dispatch.phpsh crawler'; |
201 | 201 | return $phpPath.$pathToTypo3.$cliPart; |
202 | 202 | } |