1
|
|
|
<?php |
2
|
|
|
namespace JsonTable\Analyse; |
3
|
|
|
|
4
|
|
|
/** |
5
|
|
|
* Perform lexical analysis. |
6
|
|
|
* |
7
|
|
|
* @package JsonTable |
8
|
|
|
*/ |
9
|
|
|
class Lexical extends Analyse implements AnalyseInterface |
10
|
|
|
{ |
11
|
|
|
/** |
12
|
|
|
* @var string The description for fields with invalid formats. |
13
|
|
|
*/ |
14
|
|
|
const ERROR_INVALID_PATTERN = 'There are <strong>%d</strong> fields that don\'t have the correct pattern:'; |
15
|
|
|
|
16
|
|
|
/** |
17
|
|
|
* @var string The description for fields with invalid formats. |
18
|
|
|
*/ |
19
|
|
|
const ERROR_INVALID_FORMAT = 'There are <strong>%d</strong> fields that don\'t have the correct format:'; |
20
|
|
|
|
21
|
|
|
/** |
22
|
|
|
* @var array The current CSV row being analysed. |
23
|
|
|
*/ |
24
|
|
|
private $currentCsvRow; |
25
|
|
|
|
26
|
|
|
/** |
27
|
|
|
* @var int The position of the CSV column currently being analysed. |
28
|
|
|
*/ |
29
|
|
|
private $csvColumnPosition; |
30
|
|
|
|
31
|
|
|
/** |
32
|
|
|
* @var int The position of the current CSV row row in the CSV file. |
33
|
|
|
*/ |
34
|
|
|
private $rowNumber; |
35
|
|
|
|
36
|
|
|
/** |
37
|
|
|
* @var object The schema definition for the column currently being analysed. |
38
|
|
|
*/ |
39
|
|
|
private $schemaColumn; |
40
|
|
|
|
41
|
|
|
/** |
42
|
|
|
* @var int The number of columns in the currently analysed row. |
43
|
|
|
*/ |
44
|
|
|
private $columnCount; |
45
|
|
|
|
46
|
|
|
/** |
47
|
|
|
* @var int The number of columns expected in each row. |
48
|
|
|
* This is taken from the CSV header row. |
49
|
|
|
*/ |
50
|
|
|
private $expectedColumnCount; |
51
|
|
|
|
52
|
|
|
/** |
53
|
|
|
* @var string The pattern to validate the current field against. |
54
|
|
|
*/ |
55
|
|
|
private $pattern; |
56
|
|
|
|
57
|
|
|
/** |
58
|
|
|
* @var string The format to validate the current field against. |
59
|
|
|
*/ |
60
|
|
|
private $format; |
61
|
|
|
|
62
|
|
|
/** |
63
|
|
|
* @var bool Whether the file is valid. |
64
|
|
|
*/ |
65
|
|
|
private $valid; |
66
|
|
|
|
67
|
|
|
|
68
|
|
|
/** |
69
|
|
|
* Validate that all fields are of the correct type, format and pattern. |
70
|
|
|
* This also checks that each CSV row has the expected number of columns. |
71
|
|
|
* |
72
|
|
|
* @access public |
73
|
|
|
* |
74
|
|
|
* @return boolean Is all data lexically valid. |
75
|
|
|
*/ |
76
|
28 |
|
public function validate() |
77
|
|
|
{ |
78
|
28 |
|
$this->valid = true; |
79
|
28 |
|
$this->rowNumber = 1; |
80
|
|
|
|
81
|
28 |
|
parent::rewindFilePointerToFirstData(); |
|
|
|
|
82
|
|
|
|
83
|
28 |
|
while ($this->currentCsvRow = parent::loopThroughFileRows()) { |
|
|
|
|
84
|
28 |
|
if (!$this->checkRowHasExpectedColumnCount()) { |
85
|
|
|
$this->handleUnexpectedColumnCount(); |
86
|
|
|
} |
87
|
|
|
|
88
|
28 |
|
for ($this->csvColumnPosition = 0; $this->csvColumnPosition < $this->columnCount; $this->csvColumnPosition++) { |
89
|
28 |
|
$this->schemaColumn = $this->getSchemaColumnFromCsvColumnPosition($this->csvColumnPosition); |
90
|
|
|
|
91
|
28 |
|
if (!$this->checkMandatoryColumnHasData()) { |
92
|
2 |
|
$this->handleInvalidMandatoryColumn(); |
93
|
|
|
|
94
|
2 |
|
if ($this->stopIfInvalid) { |
95
|
|
|
return false; |
96
|
|
|
} |
97
|
2 |
|
} |
98
|
|
|
|
99
|
28 |
|
if (!$this->validateSpecificFormat()) { |
100
|
8 |
|
$this->handleInvalidFormat(); |
101
|
|
|
|
102
|
8 |
|
if ($this->stopIfInvalid) { |
103
|
|
|
return false; |
104
|
|
|
} |
105
|
8 |
|
} |
106
|
|
|
|
107
|
28 |
|
if (!$this->validatePattern()) { |
108
|
|
|
$this->handleInvalidPattern(); |
109
|
|
|
|
110
|
|
|
if ($this->stopIfInvalid) { |
111
|
|
|
return false; |
112
|
|
|
} |
113
|
|
|
} |
114
|
28 |
|
} |
115
|
|
|
|
116
|
28 |
|
$this->rowNumber++; |
117
|
28 |
|
} |
118
|
|
|
|
119
|
28 |
|
$this->setRowsAnalysedStatistic(); |
120
|
|
|
|
121
|
28 |
|
return $this->valid; |
122
|
|
|
} |
123
|
|
|
|
124
|
|
|
|
125
|
|
|
/** |
126
|
|
|
* Check that the specified row has the expected number of columns. |
127
|
|
|
* The expected number of columns is the number of columns in the CSV header row. |
128
|
|
|
* |
129
|
|
|
* @return boolean Whether the current row has the expected number of columns. |
130
|
|
|
*/ |
131
|
28 |
|
private function checkRowHasExpectedColumnCount() |
132
|
|
|
{ |
133
|
28 |
|
$this->columnCount = count($this->currentCsvRow); |
134
|
28 |
|
$this->expectedColumnCount = count(parent::$headerColumns); |
135
|
|
|
|
136
|
28 |
|
return ($this->expectedColumnCount === $this->columnCount); |
137
|
|
|
} |
138
|
|
|
|
139
|
|
|
|
140
|
|
|
/** |
141
|
|
|
* Set an error and update the application as the current row has an unexpected number of columns. |
142
|
|
|
* |
143
|
|
|
* @access private |
144
|
|
|
* |
145
|
|
|
* @return void |
146
|
|
|
*/ |
147
|
|
|
private function handleUnexpectedColumnCount() |
148
|
|
|
{ |
149
|
|
|
$errorMessage = "Row $this->rowNumber has $this->columnCount columns but should have $this->expectedColumnCount."; |
150
|
|
|
$this->setError(Analyse::ERROR_INCORRECT_COLUMN_COUNT, $errorMessage); |
151
|
|
|
$this->setErrorRowStatistic($this->rowNumber); |
152
|
|
|
} |
153
|
|
|
|
154
|
|
|
|
155
|
|
|
/** |
156
|
|
|
* Check whether the current column is mandatory and if so, whether it has data in it. |
157
|
|
|
* |
158
|
|
|
* @access private |
159
|
|
|
* |
160
|
|
|
* @return boolean Whether the column has data in it. |
161
|
|
|
*/ |
162
|
28 |
|
private function checkMandatoryColumnHasData() |
163
|
|
|
{ |
164
|
28 |
|
if ($this->isColumnMandatory($this->schemaColumn)) { |
165
|
28 |
|
return ('' !== $this->currentCsvRow[$this->csvColumnPosition]); |
166
|
|
|
} |
167
|
|
|
|
168
|
25 |
|
return true; |
169
|
|
|
} |
170
|
|
|
|
171
|
|
|
|
172
|
|
|
/** |
173
|
|
|
* Set an error and update the application as the current column is mandatory and has no data in it. |
174
|
|
|
* |
175
|
|
|
* @access private |
176
|
|
|
* |
177
|
|
|
* @return void |
178
|
|
|
*/ |
179
|
2 |
View Code Duplication |
private function handleInvalidMandatoryColumn() |
|
|
|
|
180
|
|
|
{ |
181
|
2 |
|
$errorMessage = $this->schemaColumn->name . " on row $this->rowNumber is missing."; |
182
|
2 |
|
$this->setError(Analyse::ERROR_REQUIRED_FIELD_MISSING_DATA, $errorMessage); |
183
|
2 |
|
$this->setErrorRowStatistic($this->rowNumber); |
184
|
2 |
|
$this->valid = false; |
185
|
2 |
|
} |
186
|
|
|
|
187
|
|
|
|
188
|
|
|
/** |
189
|
|
|
* Check that the data in the current field is of a valid format as specified in the schema for this column. |
190
|
|
|
* This instantiates and passed the data to the format validator for this field type. |
191
|
|
|
* |
192
|
|
|
* @access private |
193
|
|
|
* |
194
|
|
|
* @return boolean Whether the current field is of a valid format. |
195
|
|
|
*/ |
196
|
28 |
|
private function validateSpecificFormat() |
197
|
|
|
{ |
198
|
28 |
|
$type = $this->getColumnType($this->schemaColumn); |
199
|
28 |
|
$this->format = $this->getColumnFormat($this->schemaColumn); |
200
|
28 |
|
$validator = $this->instantiateValidator(Analyse::VALIDATION_TYPE_FORMAT, $type); |
201
|
28 |
|
$validator->setInput($this->currentCsvRow[$this->csvColumnPosition]); |
202
|
|
|
|
203
|
28 |
|
return $validator->validateFormat($this->format); |
204
|
|
|
} |
205
|
|
|
|
206
|
|
|
|
207
|
|
|
/** |
208
|
|
|
* Set an error and update the application as the current data didn't match the specified format. |
209
|
|
|
* |
210
|
|
|
* @access private |
211
|
|
|
* |
212
|
|
|
* @return void |
213
|
|
|
*/ |
214
|
8 |
View Code Duplication |
private function handleInvalidFormat() |
|
|
|
|
215
|
|
|
{ |
216
|
8 |
|
$errorMessage = "The data in column " . $this->schemaColumn->name ." on row $this->rowNumber doesn't "; |
217
|
8 |
|
$errorMessage .= "match the required format of $this->format."; |
218
|
8 |
|
$this->setError(self::ERROR_INVALID_FORMAT, $errorMessage); |
219
|
8 |
|
$this->setErrorRowStatistic($this->rowNumber); |
220
|
8 |
|
$this->valid = false; |
221
|
8 |
|
} |
222
|
|
|
|
223
|
|
|
|
224
|
|
|
/** |
225
|
|
|
* Get the pattern of the specified column. |
226
|
|
|
* |
227
|
|
|
* @access private |
228
|
|
|
* |
229
|
|
|
* @return string The pattern or null if no pattern is specified. |
230
|
|
|
*/ |
231
|
28 |
|
private function getColumnPattern() |
232
|
|
|
{ |
233
|
28 |
|
$propertyExists = property_exists($this->schemaColumn, 'constraints') && |
234
|
28 |
|
property_exists($this->schemaColumn->constraints, 'pattern'); |
235
|
|
|
|
236
|
28 |
|
return $propertyExists ? $this->schemaColumn->constraints->pattern : null; |
237
|
|
|
} |
238
|
|
|
|
239
|
|
|
|
240
|
|
|
/** |
241
|
|
|
* Check that the input matches the specified pattern. |
242
|
|
|
* |
243
|
|
|
* @access private |
244
|
|
|
* |
245
|
|
|
* @return boolean Is the data valid. |
246
|
|
|
*/ |
247
|
28 |
|
private function validatePattern() |
248
|
|
|
{ |
249
|
28 |
|
$this->pattern = $this->getColumnPattern(); |
250
|
28 |
|
$input = $this->currentCsvRow[$this->csvColumnPosition]; |
251
|
|
|
|
252
|
28 |
|
if (is_null($this->pattern) || '' === $input) { |
253
|
28 |
|
return true; |
254
|
|
|
} |
255
|
|
|
|
256
|
2 |
|
return (false !== filter_var($input, FILTER_VALIDATE_REGEXP, ['options' => ['regexp' => $this->pattern]])); |
257
|
|
|
} |
258
|
|
|
|
259
|
|
|
|
260
|
|
|
/** |
261
|
|
|
* Set an error and update the application as the current data didn't match the specified pattern. |
262
|
|
|
* |
263
|
|
|
* @access private |
264
|
|
|
* |
265
|
|
|
* @return void |
266
|
|
|
*/ |
267
|
|
View Code Duplication |
private function handleInvalidPattern() |
|
|
|
|
268
|
|
|
{ |
269
|
|
|
$errorMessage = "The data in column " . $this->schemaColumn->name . " on row $this->rowNumber doesn't "; |
270
|
|
|
$errorMessage .= "match the required pattern of $this->pattern."; |
271
|
|
|
$this->setError(self::ERROR_INVALID_PATTERN, $errorMessage); |
272
|
|
|
$this->setErrorRowStatistic($this->rowNumber); |
273
|
|
|
$this->valid = false; |
274
|
|
|
} |
275
|
|
|
|
276
|
|
|
|
277
|
|
|
/** |
278
|
|
|
* Add the number of rows analysed to the statistics. |
279
|
|
|
* |
280
|
|
|
* @access private |
281
|
|
|
* |
282
|
|
|
* @return void |
283
|
|
|
*/ |
284
|
28 |
|
private function setRowsAnalysedStatistic() |
285
|
|
|
{ |
286
|
28 |
|
parent::$statistics['rows_analysed'] = ($this->rowNumber - 1); |
287
|
|
|
} |
288
|
|
|
} |
This check looks for a call to a parent method whose name is different than the method from which it is called.
Consider the following code:
The
getFirstName()
method in theSon
calls the wrong method in the parent class.