3 namespace Org\Mxchange\CoreFramework\Stacker\Filesystem;
5 // Import framework stuff
6 use Org\Mxchange\CoreFramework\Factory\Filesystem\Stack\FileStackIndexFactory;
7 use Org\Mxchange\CoreFramework\Factory\ObjectFactory;
8 use Org\Mxchange\CoreFramework\Filesystem\File\BaseBinaryFile;
9 use Org\Mxchange\CoreFramework\Generic\UnsupportedOperationException;
10 use Org\Mxchange\CoreFramework\Index\Indexable;
11 use Org\Mxchange\CoreFramework\Iterator\Filesystem\SeekableWritableFileIterator;
12 use Org\Mxchange\CoreFramework\Stacker\BaseStacker;
13 use Org\Mxchange\CoreFramework\Utils\String\StringUtils;
19 * A general file-based stack class
21 * @author Roland Haeder <webmaster@ship-simu.org>
23 * @copyright Copyright (c) 2007, 2008 Roland Haeder, 2009 - 2020 Core Developer Team
24 * @license GNU GPL 3.0 or any newer version
25 * @link http://www.ship-simu.org
27 * This program is free software: you can redistribute it and/or modify
28 * it under the terms of the GNU General Public License as published by
29 * the Free Software Foundation, either version 3 of the License, or
30 * (at your option) any later version.
32 * This program is distributed in the hope that it will be useful,
33 * but WITHOUT ANY WARRANTY; without even the implied warranty of
34 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
35 * GNU General Public License for more details.
37 * You should have received a copy of the GNU General Public License
38 * along with this program. If not, see <http://www.gnu.org/licenses/>.
40 abstract class BaseFileStack extends BaseStacker {
42 * Magic for this stack
44 const STACK_MAGIC = 'STACKv0.1';
47 * Name of array index for gap position
49 const ARRAY_INDEX_GAP_POSITION = 'gap';
52 * Name of array index for hash
54 const ARRAY_INDEX_HASH = 'hash';
57 * Name of array index for length of raw data
59 const ARRAY_INDEX_DATA_LENGTH = 'length';
62 * An instance of an Indexable class
64 private $indexInstance = NULL;
67 * Protected constructor
69 * @param $className Name of the class
72 protected function __construct (string $className) {
73 // Call parent constructor
74 parent::__construct($className);
78 * Setter for Indexable instance
80 * @param $indexInstance An instance of an Indexable class
83 protected final function setIndexInstance (Indexable $indexInstance) {
84 $this->indexInstance = $indexInstance;
88 * Getter for Indexable instance
90 * @return $indexInstance An instance of an Indexable class
92 public final function getIndexInstance () {
93 return $this->indexInstance;
97 * Reads the file header
100 * @todo To hard assertions here, better rewrite them to exceptions
102 public function readFileHeader () {
103 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: CALLED!', __METHOD__, __LINE__));
105 // First rewind to beginning as the header sits at the beginning ...
106 $this->getIteratorInstance()->rewind();
108 // Then read it (see constructor for calculation)
109 $data = $this->getIteratorInstance()->read($this->getIteratorInstance()->getHeaderSize());
110 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: Read %d bytes (%d wanted).', strlen($data), $this->getIteratorInstance()->getHeaderSize()));
112 // Have all requested bytes been read?
113 assert(strlen($data) == $this->getIteratorInstance()->getHeaderSize());
114 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: Passed assert().', __METHOD__, __LINE__));
116 // Last character must be the separator
117 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: data(-1)=%s', dechex(ord(substr($data, -1, 1)))));
118 assert(substr($data, -1, 1) == chr(BaseBinaryFile::SEPARATOR_HEADER_ENTRIES));
119 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: Passed assert().', __METHOD__, __LINE__));
121 // Okay, then remove it
122 $data = substr($data, 0, -1);
124 // And update seek position
125 $this->getIteratorInstance()->updateSeekPosition();
132 * 2 => current seek position
134 $header = explode(chr(BaseBinaryFile::SEPARATOR_HEADER_DATA), $data);
137 $this->getIteratorInstance()->setHeader($header);
139 // Check if the array has only 3 elements
140 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: header(%d)=%s', count($header), print_r($header, true)));
141 assert(count($header) == 3);
142 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: Passed assert().', __METHOD__, __LINE__));
145 assert($header[0] == self::STACK_MAGIC);
146 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: Passed assert().', __METHOD__, __LINE__));
148 // Check length of count and seek position
149 assert(strlen($header[1]) == BaseBinaryFile::LENGTH_COUNT);
150 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: Passed assert().', __METHOD__, __LINE__));
151 assert(strlen($header[2]) == BaseBinaryFile::LENGTH_POSITION);
152 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: Passed assert().', __METHOD__, __LINE__));
154 // Decode count and seek position
155 $header[1] = hex2bin($header[1]);
156 $header[2] = hex2bin($header[2]);
158 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: EXIT!', __METHOD__, __LINE__));
162 * Flushes the file header
166 public function flushFileHeader () {
167 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: CALLED!', __METHOD__, __LINE__));
169 // Put all informations together
170 $header = sprintf('%s%s%s%s%s%s',
174 // Separator magic<->count
175 chr(BaseBinaryFile::SEPARATOR_HEADER_DATA),
177 // Total entries (will be zero) and pad it to 20 chars
178 str_pad(StringUtils::dec2hex($this->getIteratorInstance()->getCounter()), BaseBinaryFile::LENGTH_COUNT, '0', STR_PAD_LEFT),
180 // Separator count<->seek position
181 chr(BaseBinaryFile::SEPARATOR_HEADER_DATA),
183 // Position (will be zero)
184 str_pad(StringUtils::dec2hex($this->getIteratorInstance()->getSeekPosition(), 2), BaseBinaryFile::LENGTH_POSITION, '0', STR_PAD_LEFT),
186 // Separator position<->entries
187 chr(BaseBinaryFile::SEPARATOR_HEADER_ENTRIES)
190 // Write it to disk (header is always at seek position 0)
191 $this->getIteratorInstance()->writeData(0, $header, false);
193 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: EXIT!', __METHOD__, __LINE__));
197 * Initializes this file-based stack.
199 * @param $fileInfoInstance An instance of a SplFileInfo class
200 * @param $type Type of this stack (e.g. url_source for URL sources)
202 * @todo Currently the stack file is not cached, please implement a memory-handling class and if enough RAM is found, cache the whole stack file.
204 protected function initFileStack (SplFileInfo $fileInfoInstance, string $type) {
205 // Get a stack file instance
206 $fileInstance = ObjectFactory::createObjectByConfiguredName('stack_file_class', array($fileInfoInstance, $this));
208 // Get iterator instance
209 $iteratorInstance = ObjectFactory::createObjectByConfiguredName('file_iterator_class', array($fileInstance));
211 // Is the instance implementing the right interface?
212 assert($iteratorInstance instanceof SeekableWritableFileIterator);
215 $this->setIteratorInstance($iteratorInstance);
217 // Calculate header size
218 $this->getIteratorInstance()->setHeaderSize(
219 strlen(self::STACK_MAGIC) +
220 strlen(chr(BaseBinaryFile::SEPARATOR_HEADER_DATA)) +
221 BaseBinaryFile::LENGTH_COUNT +
222 strlen(chr(BaseBinaryFile::SEPARATOR_HEADER_DATA)) +
223 BaseBinaryFile::LENGTH_POSITION +
224 strlen(chr(BaseBinaryFile::SEPARATOR_HEADER_ENTRIES))
227 // Init counters and gaps array
228 $this->getIteratorInstance()->initCountersGapsArray();
230 // Is the file's header initialized?
231 if (!$this->getIteratorInstance()->isFileHeaderInitialized()) {
232 // No, then create it (which may pre-allocate the stack)
233 $this->getIteratorInstance()->createFileHeader();
235 // And pre-allocate a bit
236 $this->getIteratorInstance()->preAllocateFile('file_stack');
239 // Load the file header
240 $this->readFileHeader();
242 // Count all entries in file
243 $this->getIteratorInstance()->analyzeFile();
246 * Get stack index instance. This can be used for faster
247 * "defragmentation" and startup.
249 $indexInstance = FileStackIndexFactory::createFileStackIndexInstance($fileInfoInstance, $type);
252 $this->setIndexInstance($indexInstance);
256 * Adds a value to given stack
258 * @param $stackerName Name of the stack
259 * @param $value Value to add to this stacker
261 * @throws FullStackerException If the stack is full
263 protected function addValue (string $stackerName, $value) {
265 if ($this->isStackFull($stackerName)) {
267 throw new FullStackerException(array($this, $stackerName, $value), self::EXCEPTION_STACKER_IS_FULL);
270 // No objects/resources are allowed as their serialization takes to long
271 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput('[' . __METHOD__ . ':' . __LINE__ . '] stackerName=' . $stackerName . ',value[' . gettype($value) . ']=' . print_r($value, true));
272 assert(!is_object($value));
273 assert(!is_resource($value));
276 * Now add the value to the file stack which returns gap position, a
277 * hash and length of the raw data.
279 $data = $this->getIteratorInstance()->writeValueToFile($stackerName, $value);
281 // Add the hash and gap position to the index
282 $this->getIndexInstance()->addHashToIndex($stackerName, $data);
286 * Get last value from named stacker
288 * @param $stackerName Name of the stack
289 * @return $value Value of last added value
290 * @throws EmptyStackerException If the stack is empty
292 protected function getLastValue (string $stackerName) {
293 // Is the stack not yet initialized or full?
294 if ($this->isStackEmpty($stackerName)) {
295 // Throw an exception
296 throw new EmptyStackerException(array($this, $stackerName), self::EXCEPTION_STACKER_IS_EMPTY);
299 // Now get the last value
300 /* NOISY-DEBUG: */ $this->partialStub('[' . __METHOD__ . ':' . __LINE__ . '] stackerName=' . $stackerName);
308 * Get first value from named stacker
310 * @param $stackerName Name of the stack
311 * @return $value Value of last added value
312 * @throws EmptyStackerException If the stack is empty
314 protected function getFirstValue (string $stackerName) {
315 // Is the stack not yet initialized or full?
316 if ($this->isStackEmpty($stackerName)) {
317 // Throw an exception
318 throw new EmptyStackerException(array($this, $stackerName), self::EXCEPTION_STACKER_IS_EMPTY);
321 // Now get the first value
322 /* NOISY-DEBUG: */ $this->partialStub('[' . __METHOD__ . ':' . __LINE__ . '] stackerName=' . $stackerName);
330 * "Pops" last entry from stack
332 * @param $stackerName Name of the stack
333 * @return $value Value "poped" from array
334 * @throws EmptyStackerException If the stack is empty
336 protected function popLast (string $stackerName) {
337 // Is the stack not yet initialized or full?
338 if ($this->isStackEmpty($stackerName)) {
339 // Throw an exception
340 throw new EmptyStackerException(array($this, $stackerName), self::EXCEPTION_STACKER_IS_EMPTY);
343 // Now, remove the last entry, we don't care about the return value here, see elseif() block above
344 /* NOISY-DEBUG: */ $this->partialStub('[' . __METHOD__ . ':' . __LINE__ . '] stackerName=' . $stackerName);
349 * "Pops" first entry from stack
351 * @param $stackerName Name of the stack
352 * @return $value Value "shifted" from array
353 * @throws EmptyStackerException If the named stacker is empty
355 protected function popFirst (string $stackerName) {
356 // Is the stack not yet initialized or full?
357 if ($this->isStackEmpty($stackerName)) {
358 // Throw an exception
359 throw new EmptyStackerException(array($this, $stackerName), self::EXCEPTION_STACKER_IS_EMPTY);
362 // Now, remove the last entry, we don't care about the return value here, see elseif() block above
363 /* NOISY-DEBUG: */ $this->partialStub('[' . __METHOD__ . ':' . __LINE__ . '] stackerName=' . $stackerName);
368 * Checks whether the given stack is full
370 * @param $stackerName Name of the stack
371 * @return $isFull Whether the stack is full
373 protected function isStackFull (string $stackerName) {
374 // File-based stacks will only run full if the disk space is low.
375 // @TODO Please implement this, returning false
383 * Checks whether the given stack is empty
385 * @param $stackerName Name of the stack
386 * @return $isEmpty Whether the stack is empty
387 * @throws NoStackerException If given stack is missing
389 public function isStackEmpty (string $stackerName) {
390 // So, is the stack empty?
391 $isEmpty = (($this->getStackCount($stackerName)) == 0);
398 * Initializes given stacker
400 * @param $stackerName Name of the stack
401 * @param $forceReInit Force re-initialization
403 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
405 public function initStack (string $stackerName, bool $forceReInit = false) {
406 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
410 * Initializes all stacks
413 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
415 public function initStacks (array $stacks, bool $forceReInit = false) {
416 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
420 * Checks whether the given stack is initialized (set in array $stackers)
422 * @param $stackerName Name of the stack
423 * @return $isInitialized Whether the stack is initialized
424 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
426 public function isStackInitialized (string $stackerName) {
427 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
431 * Determines whether the EOF has been reached
433 * @return $isEndOfFileReached Whether the EOF has been reached
434 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
436 public function isEndOfFileReached () {
437 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
441 * Getter for size of given stack (array count)
443 * @param $stackerName Name of the stack
444 * @return $count Size of stack (array count)
446 public function getStackCount (string $stackerName) {
447 // Now, simply return the found count value, this must be up-to-date then!
448 return $this->getIteratorInstance()->getCounter();
452 * Calculates minimum length for one entry/block
454 * @return $length Minimum length for one entry/block
456 public function calculateMinimumBlockLength () {
459 // Length of entry group
460 BaseBinaryFile::LENGTH_GROUP + strlen(chr(BaseBinaryFile::SEPARATOR_GROUP_HASH)) +
462 self::getHashLength() + strlen(chr(BaseBinaryFile::SEPARATOR_HASH_VALUE)) + 1 +
464 strlen(chr(BaseBinaryFile::SEPARATOR_ENTRIES));
471 * Initializes counter for valid entries, arrays for damaged entries and
472 * an array for gap seek positions. If you call this method on your own,
473 * please re-analyze the file structure. So you are better to call
474 * analyzeFile() instead of this method.
477 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
479 public function initCountersGapsArray () {
480 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
484 * Getter for header size
486 * @return $totalEntries Size of file header
487 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
489 public final function getHeaderSize () {
490 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
494 * Setter for header size
496 * @param $headerSize Size of file header
498 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
500 public final function setHeaderSize (int $headerSize) {
501 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
505 * Getter for header array
507 * @return $totalEntries Size of file header
508 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
510 public final function getHeader () {
511 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
517 * @param $header Array for a file header
519 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
521 public final function setHeader (array $header) {
522 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
526 * Updates seekPosition attribute from file to avoid to much access on file.
529 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
531 public function updateSeekPosition () {
532 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
536 * Getter for total entries
538 * @return $totalEntries Total entries in this file
539 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
541 public final function getCounter () {
542 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
546 * Writes data at given position
548 * @param $seekPosition Seek position
549 * @param $data Data to be written
550 * @param $flushHeader Whether to flush the header (default: flush)
552 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
554 public function writeData ($seekPosition, $data, bool $flushHeader = true) {
555 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: seekPosition=%s,data[]=%s,flushHeader=%d', $seekPosition, gettype($data), intval($flushHeader)));
556 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
560 * Writes given value to the file and returns a hash and gap position for it
562 * @param $groupId Group identifier
563 * @param $value Value to be added to the stack
564 * @return $data Hash and gap position
565 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
567 public function writeValueToFile ($groupId, $value) {
568 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: groupId=%s,value[%s]=%s', $groupId, gettype($value), print_r($value, true)));
569 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
573 * Searches for next suitable gap the given length of data can fit in
574 * including padding bytes.
576 * @param $length Length of raw data
577 * @return $seekPosition Found next gap's seek position
578 * @throws UnsupportedOperationException This method is not (and maybe never will be) supported
580 public function searchNextGap (int $length) {
581 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: length=%s', $length));
582 throw new UnsupportedOperationException(array($this, __FUNCTION__, $this->getIteratorInstance()->getPointerInstance()), self::EXCEPTION_UNSPPORTED_OPERATION);
586 * "Getter" for file size
588 * @return $fileSize Size of currently loaded file
590 public function getFileSize () {
591 // Call iterator's method
592 return $this->getIteratorInstance()->getFileSize();
596 * Writes given raw data to the file and returns a gap position and length
598 * @param $groupId Group identifier
599 * @param $hash Hash from encoded value
600 * @param $encoded Encoded value to be written to the file
601 * @return $data Gap position and length of the raw data
603 public function writeDataToFreeGap ($groupId, string $hash, $encoded) {
604 // Raw data been written to the file
605 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: groupId=%s,hash=%s,encoded()=%d - CALLED!', $groupId, $hash, strlen($encoded)));
606 $rawData = sprintf('%s%s%s%s%s',
608 BaseBinaryFile::SEPARATOR_GROUP_HASH,
610 BaseBinaryFile::SEPARATOR_HASH_VALUE,
614 // Search for next free gap
615 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: groupId=%s,hash=%s,rawData()=%d', $groupId, $hash, strlen($rawData)));
616 $gapPosition = $this->getIteratorInstance()->searchNextGap(strlen($rawData));
618 // Gap position cannot be smaller than header length + 1
619 assert($gapPosition > $this->getIteratorInstance()->getHeaderSize());
621 // Then write the data at that gap
622 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: groupId=%s,hash=%s,gapPosition=%s', $groupId, $hash, $gapPosition));
623 $this->getIteratorInstance()->writeData($gapPosition, $rawData);
625 // Return gap position, hash and length of raw data
626 /* NOISY-DEBUG: */ self::createDebugInstance(__CLASS__, __LINE__)->debugOutput(sprintf('BASE-FILE-STACK: groupId=%s,hash=%s,rawData()=%d - EXIT!', $groupId, $hash, strlen($rawData)));
628 self::ARRAY_INDEX_GAP_POSITION => $gapPosition,
629 self::ARRAY_INDEX_HASH => $hash,
630 self::ARRAY_INDEX_DATA_LENGTH => strlen($rawData),