1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Box\Spout\Reader\ODS; |
4
|
|
|
|
5
|
|
|
use Box\Spout\Common\Exception\IOException; |
6
|
|
|
use Box\Spout\Reader\Exception\IteratorNotRewindableException; |
7
|
|
|
use Box\Spout\Reader\Exception\XMLProcessingException; |
8
|
|
|
use Box\Spout\Reader\IteratorInterface; |
9
|
|
|
use Box\Spout\Reader\ODS\Helper\CellValueFormatter; |
10
|
|
|
use Box\Spout\Reader\Wrapper\XMLReader; |
11
|
|
|
|
12
|
|
|
/** |
13
|
|
|
* Class RowIterator |
14
|
|
|
* |
15
|
|
|
* @package Box\Spout\Reader\ODS |
16
|
|
|
*/ |
17
|
|
|
class RowIterator implements IteratorInterface |
18
|
|
|
{ |
19
|
|
|
/** Definition of XML nodes names used to parse data */ |
20
|
|
|
const XML_NODE_TABLE = 'table:table'; |
21
|
|
|
const XML_NODE_ROW = 'table:table-row'; |
22
|
|
|
const XML_NODE_CELL = 'table:table-cell'; |
23
|
|
|
const MAX_COLUMNS_EXCEL = 16384; |
24
|
|
|
|
25
|
|
|
/** Definition of XML attribute used to parse data */ |
26
|
|
|
const XML_ATTRIBUTE_NUM_ROWS_REPEATED = 'table:number-rows-repeated'; |
27
|
|
|
const XML_ATTRIBUTE_NUM_COLUMNS_REPEATED = 'table:number-columns-repeated'; |
28
|
|
|
|
29
|
|
|
/** @var \Box\Spout\Reader\Wrapper\XMLReader The XMLReader object that will help read sheet's XML data */ |
30
|
|
|
protected $xmlReader; |
31
|
|
|
|
32
|
|
|
/** @var bool Whether empty rows should be returned or skipped */ |
33
|
|
|
protected $shouldPreserveEmptyRows; |
34
|
|
|
|
35
|
|
|
/** @var Helper\CellValueFormatter Helper to format cell values */ |
36
|
|
|
protected $cellValueFormatter; |
37
|
|
|
|
38
|
|
|
/** @var bool Whether the iterator has already been rewound once */ |
39
|
|
|
protected $hasAlreadyBeenRewound = false; |
40
|
|
|
|
41
|
|
|
/** @var array|null Buffer used to store the row data, while checking if there are more rows to read */ |
42
|
|
|
protected $rowDataBuffer = null; |
43
|
|
|
|
44
|
|
|
/** @var bool Indicates whether all rows have been read */ |
45
|
|
|
protected $hasReachedEndOfFile = false; |
46
|
|
|
|
47
|
|
|
/** @var int Last row index processed (one-based) */ |
48
|
|
|
protected $lastRowIndexProcessed = 0; |
49
|
|
|
|
50
|
|
|
/** @var int Row index to be processed next (one-based) */ |
51
|
|
|
protected $nextRowIndexToBeProcessed = 1; |
52
|
|
|
|
53
|
|
|
/** @var mixed|null Value of the last processed cell (because when reading cell at column N+1, cell N is processed) */ |
54
|
|
|
protected $lastProcessedCellValue = null; |
55
|
|
|
|
56
|
|
|
/** @var int Number of times the last processed row should be repeated */ |
57
|
|
|
protected $numRowsRepeated = 1; |
58
|
|
|
|
59
|
|
|
/** @var int Number of times the last cell value should be copied to the cells on its right */ |
60
|
|
|
protected $numColumnsRepeated = 1; |
61
|
|
|
|
62
|
|
|
/** @var bool Whether at least one cell has been read for the row currently being processed */ |
63
|
|
|
protected $hasAlreadyReadOneCellInCurrentRow = false; |
64
|
|
|
|
65
|
|
|
|
66
|
|
|
/** |
67
|
|
|
* @param XMLReader $xmlReader XML Reader, positioned on the "<table:table>" element |
68
|
|
|
* @param bool $shouldFormatDates Whether date/time values should be returned as PHP objects or be formatted as strings |
69
|
|
|
* @param bool $shouldPreserveEmptyRows Whether empty rows should be returned or skipped |
70
|
|
|
*/ |
71
|
78 |
|
public function __construct($xmlReader, $shouldFormatDates, $shouldPreserveEmptyRows) |
72
|
|
|
{ |
73
|
78 |
|
$this->xmlReader = $xmlReader; |
74
|
78 |
|
$this->shouldPreserveEmptyRows = $shouldPreserveEmptyRows; |
75
|
78 |
|
$this->cellValueFormatter = new CellValueFormatter($shouldFormatDates); |
76
|
78 |
|
} |
77
|
|
|
|
78
|
|
|
/** |
79
|
|
|
* Rewind the Iterator to the first element. |
80
|
|
|
* NOTE: It can only be done once, as it is not possible to read an XML file backwards. |
81
|
|
|
* @link http://php.net/manual/en/iterator.rewind.php |
82
|
|
|
* |
83
|
|
|
* @return void |
84
|
|
|
* @throws \Box\Spout\Reader\Exception\IteratorNotRewindableException If the iterator is rewound more than once |
85
|
|
|
*/ |
86
|
78 |
|
public function rewind() |
87
|
|
|
{ |
88
|
|
|
// Because sheet and row data is located in the file, we can't rewind both the |
89
|
|
|
// sheet iterator and the row iterator, as XML file cannot be read backwards. |
90
|
|
|
// Therefore, rewinding the row iterator has been disabled. |
91
|
78 |
|
if ($this->hasAlreadyBeenRewound) { |
92
|
3 |
|
throw new IteratorNotRewindableException(); |
93
|
|
|
} |
94
|
|
|
|
95
|
78 |
|
$this->hasAlreadyBeenRewound = true; |
96
|
78 |
|
$this->lastRowIndexProcessed = 0; |
97
|
78 |
|
$this->nextRowIndexToBeProcessed = 1; |
98
|
78 |
|
$this->rowDataBuffer = null; |
99
|
78 |
|
$this->hasReachedEndOfFile = false; |
100
|
|
|
|
101
|
78 |
|
$this->next(); |
102
|
78 |
|
} |
103
|
|
|
|
104
|
|
|
/** |
105
|
|
|
* Checks if current position is valid |
106
|
|
|
* @link http://php.net/manual/en/iterator.valid.php |
107
|
|
|
* |
108
|
|
|
* @return bool |
109
|
|
|
*/ |
110
|
78 |
|
public function valid() |
111
|
|
|
{ |
112
|
78 |
|
return (!$this->hasReachedEndOfFile); |
113
|
|
|
} |
114
|
|
|
|
115
|
|
|
/** |
116
|
|
|
* Move forward to next element. Empty rows will be skipped. |
117
|
|
|
* @link http://php.net/manual/en/iterator.next.php |
118
|
|
|
* |
119
|
|
|
* @return void |
120
|
|
|
* @throws \Box\Spout\Reader\Exception\SharedStringNotFoundException If a shared string was not found |
121
|
|
|
* @throws \Box\Spout\Common\Exception\IOException If unable to read the sheet data XML |
122
|
|
|
*/ |
123
|
78 |
|
public function next() |
124
|
|
|
{ |
125
|
78 |
|
if ($this->doesNeedDataForNextRowToBeProcessed()) { |
126
|
78 |
|
$this->readDataForNextRow($this->xmlReader); |
127
|
78 |
|
} |
128
|
|
|
|
129
|
78 |
|
$this->lastRowIndexProcessed++; |
130
|
78 |
|
} |
131
|
|
|
|
132
|
|
|
/** |
133
|
|
|
* Returns whether we need data for the next row to be processed. |
134
|
|
|
* We DO need to read data if: |
135
|
|
|
* - we have not read any rows yet |
136
|
|
|
* OR |
137
|
|
|
* - the next row to be processed immediately follows the last read row |
138
|
|
|
* |
139
|
|
|
* @return bool Whether we need data for the next row to be processed. |
140
|
|
|
*/ |
141
|
78 |
|
protected function doesNeedDataForNextRowToBeProcessed() |
142
|
|
|
{ |
143
|
78 |
|
$hasReadAtLeastOneRow = ($this->lastRowIndexProcessed !== 0); |
144
|
|
|
|
145
|
|
|
return ( |
146
|
78 |
|
!$hasReadAtLeastOneRow || |
147
|
69 |
|
$this->lastRowIndexProcessed === $this->nextRowIndexToBeProcessed - 1 |
148
|
78 |
|
); |
149
|
|
|
} |
150
|
|
|
|
151
|
|
|
/** |
152
|
|
|
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XMLReader object |
153
|
|
|
* @return void |
154
|
|
|
* @throws \Box\Spout\Reader\Exception\SharedStringNotFoundException If a shared string was not found |
155
|
|
|
* @throws \Box\Spout\Common\Exception\IOException If unable to read the sheet data XML |
156
|
|
|
*/ |
157
|
78 |
|
protected function readDataForNextRow($xmlReader) |
158
|
|
|
{ |
159
|
78 |
|
$rowData = []; |
160
|
|
|
|
161
|
|
|
try { |
162
|
78 |
|
while ($xmlReader->read()) { |
163
|
78 |
|
if ($xmlReader->isPositionedOnStartingNode(self::XML_NODE_ROW)) { |
164
|
78 |
|
$this->processRowStartingNode($xmlReader); |
165
|
|
|
|
166
|
78 |
|
} else if ($xmlReader->isPositionedOnStartingNode(self::XML_NODE_CELL)) { |
167
|
78 |
|
$rowData = $this->processCellStartingNode($xmlReader, $rowData); |
168
|
|
|
|
169
|
78 |
|
} else if ($xmlReader->isPositionedOnEndingNode(self::XML_NODE_ROW)) { |
170
|
78 |
|
$isEmptyRow = $this->isEmptyRow($rowData, $this->lastProcessedCellValue); |
171
|
|
|
|
172
|
|
|
// if the fetched row is empty and we don't want to preserve it... |
173
|
78 |
|
if (!$this->shouldPreserveEmptyRows && $isEmptyRow) { |
174
|
|
|
// ... skip it |
175
|
15 |
|
continue; |
176
|
|
|
} |
177
|
|
|
|
178
|
75 |
|
$rowData = $this->processRowEndingNode($rowData, $isEmptyRow); |
179
|
|
|
|
180
|
|
|
// at this point, we have all the data we need for the row |
181
|
|
|
// so that we can populate the buffer |
182
|
75 |
|
break; |
183
|
|
|
|
184
|
78 |
|
} else if ($xmlReader->isPositionedOnEndingNode(self::XML_NODE_TABLE)) { |
185
|
72 |
|
$this->processTableEndingNode(); |
186
|
72 |
|
break; |
187
|
|
|
} |
188
|
78 |
|
} |
189
|
|
|
|
190
|
78 |
|
} catch (XMLProcessingException $exception) { |
191
|
|
|
throw new IOException("The sheet's data cannot be read. [{$exception->getMessage()}]"); |
192
|
|
|
} |
193
|
|
|
|
194
|
78 |
|
$this->rowDataBuffer = $rowData; |
195
|
78 |
|
} |
196
|
|
|
|
197
|
|
|
/** |
198
|
|
|
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XMLReader object, positioned on a "<table:table-row>" starting node |
199
|
|
|
* @return void |
200
|
|
|
*/ |
201
|
78 |
|
protected function processRowStartingNode($xmlReader) |
202
|
|
|
{ |
203
|
|
|
// Reset data from current row |
204
|
78 |
|
$this->hasAlreadyReadOneCellInCurrentRow = false; |
205
|
78 |
|
$this->lastProcessedCellValue = null; |
206
|
78 |
|
$this->numColumnsRepeated = 1; |
207
|
78 |
|
$this->numRowsRepeated = $this->getNumRowsRepeatedForCurrentNode($xmlReader); |
208
|
78 |
|
} |
209
|
|
|
|
210
|
|
|
/** |
211
|
|
|
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XMLReader object, positioned on a "<table:table-cell>" starting node |
212
|
|
|
* @param array $rowData Data of all cells read so far |
213
|
|
|
* @return array Original row data + data for the cell that was just read |
214
|
|
|
*/ |
215
|
78 |
|
protected function processCellStartingNode($xmlReader, $rowData) |
216
|
|
|
{ |
217
|
78 |
|
$currentNumColumnsRepeated = $this->getNumColumnsRepeatedForCurrentNode($xmlReader); |
218
|
|
|
|
219
|
78 |
|
$node = $xmlReader->expand(); |
220
|
78 |
|
$currentCellValue = $this->getCellValue($node); |
221
|
|
|
|
222
|
|
|
// process cell N only after having read cell N+1 (see below why) |
223
|
78 |
|
if ($this->hasAlreadyReadOneCellInCurrentRow) { |
224
|
63 |
|
for ($i = 0; $i < $this->numColumnsRepeated; $i++) { |
225
|
63 |
|
$rowData[] = $this->lastProcessedCellValue; |
226
|
63 |
|
} |
227
|
63 |
|
} |
228
|
|
|
|
229
|
78 |
|
$this->hasAlreadyReadOneCellInCurrentRow = true; |
230
|
78 |
|
$this->lastProcessedCellValue = $currentCellValue; |
231
|
78 |
|
$this->numColumnsRepeated = $currentNumColumnsRepeated; |
232
|
|
|
|
233
|
78 |
|
return $rowData; |
234
|
|
|
} |
235
|
|
|
|
236
|
|
|
/** |
237
|
|
|
* @param array $rowData Data of all cells read so far |
238
|
|
|
* @param bool $isEmptyRow Whether the given row is empty |
239
|
|
|
* @return array |
240
|
|
|
*/ |
241
|
75 |
|
protected function processRowEndingNode($rowData, $isEmptyRow) |
242
|
|
|
{ |
243
|
|
|
// if the row is empty, we don't want to return more than one cell |
244
|
75 |
|
$actualNumColumnsRepeated = (!$isEmptyRow) ? $this->numColumnsRepeated : 1; |
245
|
|
|
|
246
|
|
|
// Only add the value if the last read cell is not a trailing empty cell repeater in Excel. |
247
|
|
|
// The current count of read columns is determined by counting the values in $rowData. |
248
|
|
|
// This is to avoid creating a lot of empty cells, as Excel adds a last empty "<table:table-cell>" |
249
|
|
|
// with a number-columns-repeated value equals to the number of (supported columns - used columns). |
250
|
|
|
// In Excel, the number of supported columns is 16384, but we don't want to returns rows with |
251
|
|
|
// always 16384 cells. |
252
|
75 |
|
if ((count($rowData) + $actualNumColumnsRepeated) !== self::MAX_COLUMNS_EXCEL) { |
253
|
69 |
|
for ($i = 0; $i < $actualNumColumnsRepeated; $i++) { |
254
|
69 |
|
$rowData[] = $this->lastProcessedCellValue; |
255
|
69 |
|
} |
256
|
69 |
|
} |
257
|
|
|
|
258
|
|
|
// If we are processing row N and the row is repeated M times, |
259
|
|
|
// then the next row to be processed will be row (N+M). |
260
|
75 |
|
$this->nextRowIndexToBeProcessed += $this->numRowsRepeated; |
261
|
|
|
|
262
|
75 |
|
return $rowData; |
263
|
|
|
} |
264
|
|
|
|
265
|
|
|
/** |
266
|
|
|
* @return void |
267
|
|
|
*/ |
268
|
72 |
|
protected function processTableEndingNode() |
269
|
|
|
{ |
270
|
|
|
// The closing "</table:table>" marks the end of the file |
271
|
72 |
|
$this->hasReachedEndOfFile = true; |
272
|
72 |
|
} |
273
|
|
|
|
274
|
|
|
/** |
275
|
|
|
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XMLReader object, positioned on a "<table:table-row>" starting node |
276
|
|
|
* @return int The value of "table:number-rows-repeated" attribute of the current node, or 1 if attribute missing |
277
|
|
|
*/ |
278
|
78 |
|
protected function getNumRowsRepeatedForCurrentNode($xmlReader) |
279
|
|
|
{ |
280
|
78 |
|
$numRowsRepeated = $xmlReader->getAttribute(self::XML_ATTRIBUTE_NUM_ROWS_REPEATED); |
281
|
78 |
|
return ($numRowsRepeated !== null) ? intval($numRowsRepeated) : 1; |
282
|
|
|
} |
283
|
|
|
|
284
|
|
|
/** |
285
|
|
|
* @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReader XMLReader object, positioned on a "<table:table-cell>" starting node |
286
|
|
|
* @return int The value of "table:number-columns-repeated" attribute of the current node, or 1 if attribute missing |
287
|
|
|
*/ |
288
|
78 |
|
protected function getNumColumnsRepeatedForCurrentNode($xmlReader) |
289
|
|
|
{ |
290
|
78 |
|
$numColumnsRepeated = $xmlReader->getAttribute(self::XML_ATTRIBUTE_NUM_COLUMNS_REPEATED); |
291
|
78 |
|
return ($numColumnsRepeated !== null) ? intval($numColumnsRepeated) : 1; |
292
|
|
|
} |
293
|
|
|
|
294
|
|
|
/** |
295
|
|
|
* Returns the (unescaped) correctly marshalled, cell value associated to the given XML node. |
296
|
|
|
* |
297
|
|
|
* @param \DOMNode $node |
298
|
|
|
* @return string|int|float|bool|\DateTime|\DateInterval|null The value associated with the cell, empty string if cell's type is void/undefined, null on error |
299
|
|
|
*/ |
300
|
78 |
|
protected function getCellValue($node) |
301
|
|
|
{ |
302
|
78 |
|
return $this->cellValueFormatter->extractAndFormatNodeValue($node); |
303
|
|
|
} |
304
|
|
|
|
305
|
|
|
/** |
306
|
|
|
* After finishing processing each cell, a row is considered empty if it contains |
307
|
|
|
* no cells or if the value of the last read cell is an empty string. |
308
|
|
|
* After finishing processing each cell, the last read cell is not part of the |
309
|
|
|
* row data yet (as we still need to apply the "num-columns-repeated" attribute). |
310
|
|
|
* |
311
|
|
|
* @param array $rowData |
312
|
|
|
* @param string|int|float|bool|\DateTime|\DateInterval|null The value of the last read cell |
313
|
|
|
* @return bool Whether the row is empty |
314
|
|
|
*/ |
315
|
78 |
|
protected function isEmptyRow($rowData, $lastReadCellValue) |
316
|
|
|
{ |
317
|
|
|
return ( |
318
|
78 |
|
count($rowData) === 0 && |
319
|
33 |
|
(!isset($lastReadCellValue) || trim($lastReadCellValue) === '') |
320
|
78 |
|
); |
321
|
|
|
} |
322
|
|
|
|
323
|
|
|
/** |
324
|
|
|
* Return the current element, from the buffer. |
325
|
|
|
* @link http://php.net/manual/en/iterator.current.php |
326
|
|
|
* |
327
|
|
|
* @return array|null |
328
|
|
|
*/ |
329
|
75 |
|
public function current() |
330
|
|
|
{ |
331
|
75 |
|
return $this->rowDataBuffer; |
332
|
|
|
} |
333
|
|
|
|
334
|
|
|
/** |
335
|
|
|
* Return the key of the current element |
336
|
|
|
* @link http://php.net/manual/en/iterator.key.php |
337
|
|
|
* |
338
|
|
|
* @return int |
339
|
|
|
*/ |
340
|
69 |
|
public function key() |
341
|
|
|
{ |
342
|
69 |
|
return $this->lastRowIndexProcessed; |
343
|
|
|
} |
344
|
|
|
|
345
|
|
|
|
346
|
|
|
/** |
347
|
|
|
* Cleans up what was created to iterate over the object. |
348
|
|
|
* |
349
|
|
|
* @return void |
350
|
|
|
*/ |
351
|
|
|
public function end() |
352
|
|
|
{ |
353
|
|
|
$this->xmlReader->close(); |
354
|
|
|
} |
355
|
|
|
} |
356
|
|
|
|