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