1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Box\Spout\Reader\XLSX\Helper; |
4
|
|
|
|
5
|
|
|
use Box\Spout\Reader\XLSX\Manager\SharedStringsManager; |
6
|
|
|
use Box\Spout\Reader\XLSX\Manager\StyleManager; |
7
|
|
|
|
8
|
|
|
/** |
9
|
|
|
* Class CellValueFormatter |
10
|
|
|
* This class provides helper functions to format cell values |
11
|
|
|
*/ |
12
|
|
|
class CellValueFormatter |
13
|
|
|
{ |
14
|
|
|
/** Definition of all possible cell types */ |
15
|
|
|
const CELL_TYPE_INLINE_STRING = 'inlineStr'; |
16
|
|
|
const CELL_TYPE_STR = 'str'; |
17
|
|
|
const CELL_TYPE_SHARED_STRING = 's'; |
18
|
|
|
const CELL_TYPE_BOOLEAN = 'b'; |
19
|
|
|
const CELL_TYPE_NUMERIC = 'n'; |
20
|
|
|
const CELL_TYPE_DATE = 'd'; |
21
|
|
|
const CELL_TYPE_ERROR = 'e'; |
22
|
|
|
|
23
|
|
|
/** Definition of XML nodes names used to parse data */ |
24
|
|
|
const XML_NODE_VALUE = 'v'; |
25
|
|
|
const XML_NODE_INLINE_STRING_VALUE = 't'; |
26
|
|
|
|
27
|
|
|
/** Definition of XML attributes used to parse data */ |
28
|
|
|
const XML_ATTRIBUTE_TYPE = 't'; |
29
|
|
|
const XML_ATTRIBUTE_STYLE_ID = 's'; |
30
|
|
|
|
31
|
|
|
/** Constants used for date formatting */ |
32
|
|
|
const NUM_SECONDS_IN_ONE_DAY = 86400; |
33
|
|
|
const NUM_SECONDS_IN_ONE_HOUR = 3600; |
34
|
|
|
const NUM_SECONDS_IN_ONE_MINUTE = 60; |
35
|
|
|
|
36
|
|
|
/** |
37
|
|
|
* February 29th, 1900 is NOT a leap year but Excel thinks it is... |
38
|
|
|
* @see https://en.wikipedia.org/wiki/Year_1900_problem#Microsoft_Excel |
39
|
|
|
*/ |
40
|
|
|
const ERRONEOUS_EXCEL_LEAP_YEAR_DAY = 60; |
41
|
|
|
|
42
|
|
|
/** @var SharedStringsManager Manages shared strings */ |
43
|
|
|
protected $sharedStringsManager; |
44
|
|
|
|
45
|
|
|
/** @var StyleManager Manages styles */ |
46
|
|
|
protected $styleManager; |
47
|
|
|
|
48
|
|
|
/** @var bool Whether date/time values should be returned as PHP objects or be formatted as strings */ |
49
|
|
|
protected $shouldFormatDates; |
50
|
|
|
|
51
|
|
|
/** @var \Box\Spout\Common\Helper\Escaper\XLSX Used to unescape XML data */ |
52
|
|
|
protected $escaper; |
53
|
|
|
|
54
|
|
|
/** |
55
|
|
|
* @param SharedStringsManager $sharedStringsManager Manages shared strings |
56
|
|
|
* @param StyleManager $styleManager Manages styles |
57
|
|
|
* @param bool $shouldFormatDates Whether date/time values should be returned as PHP objects or be formatted as strings |
58
|
|
|
* @param \Box\Spout\Common\Helper\Escaper\XLSX $escaper Used to unescape XML data |
59
|
|
|
*/ |
60
|
61 |
|
public function __construct($sharedStringsManager, $styleManager, $shouldFormatDates, $escaper) |
61
|
|
|
{ |
62
|
61 |
|
$this->sharedStringsManager = $sharedStringsManager; |
63
|
61 |
|
$this->styleManager = $styleManager; |
64
|
61 |
|
$this->shouldFormatDates = $shouldFormatDates; |
65
|
61 |
|
$this->escaper = $escaper; |
66
|
61 |
|
} |
67
|
|
|
|
68
|
|
|
/** |
69
|
|
|
* Returns the (unescaped) correctly marshalled, cell value associated to the given XML node. |
70
|
|
|
* |
71
|
|
|
* @param \DOMNode $node |
72
|
|
|
* @return string|int|float|bool|\DateTime|null The value associated with the cell (null when the cell has an error) |
73
|
|
|
*/ |
74
|
41 |
|
public function extractAndFormatNodeValue($node) |
75
|
|
|
{ |
76
|
|
|
// Default cell type is "n" |
77
|
41 |
|
$cellType = $node->getAttribute(self::XML_ATTRIBUTE_TYPE) ?: self::CELL_TYPE_NUMERIC; |
78
|
41 |
|
$cellStyleId = (int) $node->getAttribute(self::XML_ATTRIBUTE_STYLE_ID); |
79
|
41 |
|
$vNodeValue = $this->getVNodeValue($node); |
80
|
|
|
|
81
|
41 |
|
if (($vNodeValue === '') && ($cellType !== self::CELL_TYPE_INLINE_STRING)) { |
82
|
1 |
|
return $vNodeValue; |
83
|
|
|
} |
84
|
|
|
|
85
|
|
|
switch ($cellType) { |
86
|
41 |
|
case self::CELL_TYPE_INLINE_STRING: |
87
|
7 |
|
return $this->formatInlineStringCellValue($node); |
88
|
35 |
|
case self::CELL_TYPE_SHARED_STRING: |
89
|
18 |
|
return $this->formatSharedStringCellValue($vNodeValue); |
90
|
21 |
|
case self::CELL_TYPE_STR: |
91
|
1 |
|
return $this->formatStrCellValue($vNodeValue); |
92
|
21 |
|
case self::CELL_TYPE_BOOLEAN: |
93
|
1 |
|
return $this->formatBooleanCellValue($vNodeValue); |
94
|
21 |
|
case self::CELL_TYPE_NUMERIC: |
95
|
21 |
|
return $this->formatNumericCellValue($vNodeValue, $cellStyleId); |
96
|
2 |
|
case self::CELL_TYPE_DATE: |
97
|
2 |
|
return $this->formatDateCellValue($vNodeValue); |
98
|
|
|
default: |
99
|
1 |
|
return null; |
100
|
|
|
} |
101
|
|
|
} |
102
|
|
|
|
103
|
|
|
/** |
104
|
|
|
* Returns the cell's string value from a node's nested value node |
105
|
|
|
* |
106
|
|
|
* @param \DOMNode $node |
107
|
|
|
* @return string The value associated with the cell |
108
|
|
|
*/ |
109
|
41 |
|
protected function getVNodeValue($node) |
110
|
|
|
{ |
111
|
|
|
// for cell types having a "v" tag containing the value. |
112
|
|
|
// if not, the returned value should be empty string. |
113
|
41 |
|
$vNode = $node->getElementsByTagName(self::XML_NODE_VALUE)->item(0); |
114
|
|
|
|
115
|
41 |
|
return ($vNode !== null) ? $vNode->nodeValue : ''; |
116
|
|
|
} |
117
|
|
|
|
118
|
|
|
/** |
119
|
|
|
* Returns the cell String value where string is inline. |
120
|
|
|
* |
121
|
|
|
* @param \DOMNode $node |
122
|
|
|
* @return string The value associated with the cell (null when the cell has an error) |
123
|
|
|
*/ |
124
|
11 |
|
protected function formatInlineStringCellValue($node) |
125
|
|
|
{ |
126
|
|
|
// inline strings are formatted this way: |
127
|
|
|
// <c r="A1" t="inlineStr"><is><t>[INLINE_STRING]</t></is></c> |
128
|
11 |
|
$tNode = $node->getElementsByTagName(self::XML_NODE_INLINE_STRING_VALUE)->item(0); |
129
|
11 |
|
$cellValue = $this->escaper->unescape($tNode->nodeValue); |
130
|
|
|
|
131
|
11 |
|
return $cellValue; |
132
|
|
|
} |
133
|
|
|
|
134
|
|
|
/** |
135
|
|
|
* Returns the cell String value from shared-strings file using nodeValue index. |
136
|
|
|
* |
137
|
|
|
* @param string $nodeValue |
138
|
|
|
* @return string The value associated with the cell (null when the cell has an error) |
139
|
|
|
*/ |
140
|
18 |
|
protected function formatSharedStringCellValue($nodeValue) |
141
|
|
|
{ |
142
|
|
|
// shared strings are formatted this way: |
143
|
|
|
// <c r="A1" t="s"><v>[SHARED_STRING_INDEX]</v></c> |
144
|
18 |
|
$sharedStringIndex = (int) $nodeValue; |
145
|
18 |
|
$escapedCellValue = $this->sharedStringsManager->getStringAtIndex($sharedStringIndex); |
146
|
18 |
|
$cellValue = $this->escaper->unescape($escapedCellValue); |
147
|
|
|
|
148
|
18 |
|
return $cellValue; |
149
|
|
|
} |
150
|
|
|
|
151
|
|
|
/** |
152
|
|
|
* Returns the cell String value, where string is stored in value node. |
153
|
|
|
* |
154
|
|
|
* @param string $nodeValue |
155
|
|
|
* @return string The value associated with the cell (null when the cell has an error) |
156
|
|
|
*/ |
157
|
1 |
|
protected function formatStrCellValue($nodeValue) |
158
|
|
|
{ |
159
|
1 |
|
$escapedCellValue = trim($nodeValue); |
160
|
1 |
|
$cellValue = $this->escaper->unescape($escapedCellValue); |
161
|
|
|
|
162
|
1 |
|
return $cellValue; |
163
|
|
|
} |
164
|
|
|
|
165
|
|
|
/** |
166
|
|
|
* Returns the cell Numeric value from string of nodeValue. |
167
|
|
|
* The value can also represent a timestamp and a DateTime will be returned. |
168
|
|
|
* |
169
|
|
|
* @param string $nodeValue |
170
|
|
|
* @param int $cellStyleId 0 being the default style |
171
|
|
|
* @return int|float|\DateTime|null The value associated with the cell |
172
|
|
|
*/ |
173
|
34 |
|
protected function formatNumericCellValue($nodeValue, $cellStyleId) |
174
|
|
|
{ |
175
|
|
|
// Numeric values can represent numbers as well as timestamps. |
176
|
|
|
// We need to look at the style of the cell to determine whether it is one or the other. |
177
|
34 |
|
$shouldFormatAsDate = $this->styleManager->shouldFormatNumericValueAsDate($cellStyleId); |
178
|
|
|
|
179
|
34 |
|
if ($shouldFormatAsDate) { |
180
|
16 |
|
$cellValue = $this->formatExcelTimestampValue((float) $nodeValue, $cellStyleId); |
181
|
|
|
} else { |
182
|
19 |
|
$nodeIntValue = (int) $nodeValue; |
183
|
19 |
|
$nodeFloatValue = (float) $nodeValue; |
184
|
19 |
|
$cellValue = ((float) $nodeIntValue === $nodeFloatValue) ? $nodeIntValue : $nodeFloatValue; |
185
|
|
|
} |
186
|
|
|
|
187
|
34 |
|
return $cellValue; |
188
|
|
|
} |
189
|
|
|
|
190
|
|
|
/** |
191
|
|
|
* Returns a cell's PHP Date value, associated to the given timestamp. |
192
|
|
|
* NOTE: The timestamp is a float representing the number of days since January 1st, 1900. |
193
|
|
|
* NOTE: The timestamp can also represent a time, if it is a value between 0 and 1. |
194
|
|
|
* |
195
|
|
|
* @param float $nodeValue |
196
|
|
|
* @param int $cellStyleId 0 being the default style |
197
|
|
|
* @return \DateTime|null The value associated with the cell or NULL if invalid date value |
198
|
|
|
*/ |
199
|
16 |
|
protected function formatExcelTimestampValue($nodeValue, $cellStyleId) |
200
|
|
|
{ |
201
|
|
|
// Fix for the erroneous leap year in Excel |
202
|
16 |
|
if (ceil($nodeValue) > self::ERRONEOUS_EXCEL_LEAP_YEAR_DAY) { |
203
|
7 |
|
--$nodeValue; |
204
|
|
|
} |
205
|
|
|
|
206
|
16 |
|
if ($nodeValue >= 1) { |
207
|
|
|
// Values greater than 1 represent "dates". The value 1.0 representing the "base" date: 1900-01-01. |
208
|
9 |
|
$cellValue = $this->formatExcelTimestampValueAsDateValue($nodeValue, $cellStyleId); |
209
|
9 |
|
} elseif ($nodeValue >= 0) { |
210
|
|
|
// Values between 0 and 1 represent "times". |
211
|
8 |
|
$cellValue = $this->formatExcelTimestampValueAsTimeValue($nodeValue, $cellStyleId); |
212
|
|
|
} else { |
213
|
|
|
// invalid date |
214
|
1 |
|
$cellValue = null; |
215
|
|
|
} |
216
|
|
|
|
217
|
16 |
|
return $cellValue; |
218
|
|
|
} |
219
|
|
|
|
220
|
|
|
/** |
221
|
|
|
* Returns a cell's PHP DateTime value, associated to the given timestamp. |
222
|
|
|
* Only the time value matters. The date part is set to Jan 1st, 1900 (base Excel date). |
223
|
|
|
* |
224
|
|
|
* @param float $nodeValue |
225
|
|
|
* @param int $cellStyleId 0 being the default style |
226
|
|
|
* @return \DateTime|string The value associated with the cell |
227
|
|
|
*/ |
228
|
8 |
|
protected function formatExcelTimestampValueAsTimeValue($nodeValue, $cellStyleId) |
229
|
|
|
{ |
230
|
8 |
|
$time = round($nodeValue * self::NUM_SECONDS_IN_ONE_DAY); |
231
|
8 |
|
$hours = floor($time / self::NUM_SECONDS_IN_ONE_HOUR); |
232
|
8 |
|
$minutes = floor($time / self::NUM_SECONDS_IN_ONE_MINUTE) - ($hours * self::NUM_SECONDS_IN_ONE_MINUTE); |
233
|
8 |
|
$seconds = $time - ($hours * self::NUM_SECONDS_IN_ONE_HOUR) - ($minutes * self::NUM_SECONDS_IN_ONE_MINUTE); |
234
|
|
|
|
235
|
|
|
// using the base Excel date (Jan 1st, 1900) - not relevant here |
236
|
8 |
|
$dateObj = new \DateTime('1900-01-01'); |
237
|
8 |
|
$dateObj->setTime($hours, $minutes, $seconds); |
238
|
|
|
|
239
|
8 |
|
if ($this->shouldFormatDates) { |
240
|
2 |
|
$styleNumberFormatCode = $this->styleManager->getNumberFormatCode($cellStyleId); |
241
|
2 |
|
$phpDateFormat = DateFormatHelper::toPHPDateFormat($styleNumberFormatCode); |
242
|
2 |
|
$cellValue = $dateObj->format($phpDateFormat); |
243
|
|
|
} else { |
244
|
6 |
|
$cellValue = $dateObj; |
245
|
|
|
} |
246
|
|
|
|
247
|
8 |
|
return $cellValue; |
248
|
|
|
} |
249
|
|
|
|
250
|
|
|
/** |
251
|
|
|
* Returns a cell's PHP Date value, associated to the given timestamp. |
252
|
|
|
* NOTE: The timestamp is a float representing the number of days since January 1st, 1900. |
253
|
|
|
* |
254
|
|
|
* @param float $nodeValue |
255
|
|
|
* @param int $cellStyleId 0 being the default style |
256
|
|
|
* @return \DateTime|string|null The value associated with the cell or NULL if invalid date value |
257
|
|
|
*/ |
258
|
9 |
|
protected function formatExcelTimestampValueAsDateValue($nodeValue, $cellStyleId) |
259
|
|
|
{ |
260
|
|
|
// Do not use any unix timestamps for calculation to prevent |
261
|
|
|
// issues with numbers exceeding 2^31. |
262
|
9 |
|
$secondsRemainder = fmod($nodeValue, 1) * self::NUM_SECONDS_IN_ONE_DAY; |
263
|
9 |
|
$secondsRemainder = round($secondsRemainder, 0); |
264
|
|
|
|
265
|
|
|
try { |
266
|
9 |
|
$dateObj = \DateTime::createFromFormat('|Y-m-d', '1899-12-31'); |
267
|
9 |
|
$dateObj->modify('+' . (int) $nodeValue . 'days'); |
268
|
9 |
|
$dateObj->modify('+' . $secondsRemainder . 'seconds'); |
269
|
|
|
|
270
|
9 |
|
if ($this->shouldFormatDates) { |
271
|
2 |
|
$styleNumberFormatCode = $this->styleManager->getNumberFormatCode($cellStyleId); |
272
|
2 |
|
$phpDateFormat = DateFormatHelper::toPHPDateFormat($styleNumberFormatCode); |
273
|
2 |
|
$cellValue = $dateObj->format($phpDateFormat); |
274
|
|
|
} else { |
275
|
9 |
|
$cellValue = $dateObj; |
276
|
|
|
} |
277
|
|
|
} catch (\Exception $e) { |
278
|
|
|
$cellValue = null; |
279
|
|
|
} |
280
|
|
|
|
281
|
9 |
|
return $cellValue; |
282
|
|
|
} |
283
|
|
|
|
284
|
|
|
/** |
285
|
|
|
* Returns the cell Boolean value from a specific node's Value. |
286
|
|
|
* |
287
|
|
|
* @param string $nodeValue |
288
|
|
|
* @return bool The value associated with the cell |
289
|
|
|
*/ |
290
|
1 |
|
protected function formatBooleanCellValue($nodeValue) |
291
|
|
|
{ |
292
|
1 |
|
return (bool) $nodeValue; |
293
|
|
|
} |
294
|
|
|
|
295
|
|
|
/** |
296
|
|
|
* Returns a cell's PHP Date value, associated to the given stored nodeValue. |
297
|
|
|
* @see ECMA-376 Part 1 - §18.17.4 |
298
|
|
|
* |
299
|
|
|
* @param string $nodeValue ISO 8601 Date string |
300
|
|
|
* @return \DateTime|string|null The value associated with the cell or NULL if invalid date value |
301
|
|
|
*/ |
302
|
2 |
|
protected function formatDateCellValue($nodeValue) |
303
|
|
|
{ |
304
|
|
|
// Mitigate thrown Exception on invalid date-time format (http://php.net/manual/en/datetime.construct.php) |
305
|
|
|
try { |
306
|
2 |
|
$cellValue = ($this->shouldFormatDates) ? $nodeValue : new \DateTime($nodeValue); |
307
|
1 |
|
} catch (\Exception $e) { |
308
|
1 |
|
$cellValue = null; |
309
|
|
|
} |
310
|
|
|
|
311
|
2 |
|
return $cellValue; |
312
|
|
|
} |
313
|
|
|
} |
314
|
|
|
|