|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace frictionlessdata\datapackage; |
|
4
|
|
|
|
|
5
|
|
|
use frictionlessdata\datapackage\Datapackages\BaseDatapackage; |
|
6
|
|
|
use frictionlessdata\datapackage\Resources\BaseResource; |
|
7
|
|
|
|
|
8
|
|
|
/** |
|
9
|
|
|
* datapackage and resource have different classes depending on the corresponding profile |
|
10
|
|
|
* this factory interface allows to validate and create object instances without having to check the profile first. |
|
11
|
|
|
*/ |
|
12
|
|
|
class Factory |
|
13
|
|
|
{ |
|
14
|
|
|
/** |
|
15
|
|
|
* how many lines to validate sample when validating data streams. |
|
16
|
|
|
*/ |
|
17
|
|
|
const VALIDATE_PEEK_LINES = 10; |
|
18
|
|
|
|
|
19
|
|
|
/** |
|
20
|
|
|
* load, validate and create a datapackage object |
|
21
|
|
|
* supports loading from the following sources: |
|
22
|
|
|
* - native PHP object containing the descriptor |
|
23
|
|
|
* - JSON encoded object |
|
24
|
|
|
* - URL (must be in either 'http' or 'https' schemes) |
|
25
|
|
|
* - local filesystem (POSIX) path. |
|
26
|
|
|
* |
|
27
|
|
|
* @param mixed $source |
|
28
|
|
|
* @param null|string $basePath optional, required only if you want to use relative paths |
|
29
|
|
|
* |
|
30
|
|
|
* @return Datapackages\BaseDatapackage |
|
31
|
|
|
* |
|
32
|
|
|
* @throws Exceptions\DatapackageInvalidSourceException |
|
33
|
|
|
* @throws Exceptions\DatapackageValidationFailedException |
|
34
|
|
|
*/ |
|
35
|
|
|
public static function datapackage($source, $basePath = null) |
|
36
|
|
|
{ |
|
37
|
|
|
$source = static::loadSource($source, $basePath); |
|
38
|
|
|
$descriptor = $source->descriptor; |
|
39
|
|
|
$basePath = $source->basePath; |
|
40
|
|
|
$datapackageClass = static::getDatapackageClass($descriptor); |
|
41
|
|
|
$datapackage = new $datapackageClass($descriptor, $basePath); |
|
42
|
|
|
|
|
43
|
|
|
return $datapackage; |
|
44
|
|
|
} |
|
45
|
|
|
|
|
46
|
|
|
/** |
|
47
|
|
|
* create a resource object. |
|
48
|
|
|
* |
|
49
|
|
|
* @param object $descriptor |
|
50
|
|
|
* @param null|string $basePath |
|
51
|
|
|
* @param bool $skipValidations |
|
52
|
|
|
* |
|
53
|
|
|
* @return Resources\BaseResource |
|
54
|
|
|
* |
|
55
|
|
|
* @throws Exceptions\ResourceValidationFailedException |
|
56
|
|
|
*/ |
|
57
|
|
|
public static function resource($descriptor, $basePath = null, $skipValidations = false) |
|
58
|
|
|
{ |
|
59
|
|
|
$resourceClass = static::getResourceClass($descriptor); |
|
60
|
|
|
$resource = new $resourceClass($descriptor, $basePath, $skipValidations); |
|
61
|
|
|
|
|
62
|
|
|
return $resource; |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
/** |
|
66
|
|
|
* validates a given datapackage descriptor |
|
67
|
|
|
* will load all resources, and sample 10 lines of data from each data source. |
|
68
|
|
|
* |
|
69
|
|
|
* @param mixed $source datapackage source - same as in datapackage function |
|
70
|
|
|
* @param null|string $basePath same as in datapackage function |
|
71
|
|
|
* |
|
72
|
|
|
* @return Validators\DatapackageValidationError[] |
|
73
|
|
|
*/ |
|
74
|
|
|
public static function validate($source, $basePath = null) |
|
75
|
|
|
{ |
|
76
|
|
|
$curResource = 1; |
|
77
|
|
|
$curLine = null; |
|
78
|
|
|
try { |
|
79
|
|
|
$datapackage = static::datapackage($source, $basePath); |
|
80
|
|
|
foreach ($datapackage as $resource) { |
|
81
|
|
|
$curLine = 1; |
|
82
|
|
|
foreach ($resource as $line) { |
|
83
|
|
|
if ($curLine == self::VALIDATE_PEEK_LINES) { |
|
84
|
|
|
break; |
|
85
|
|
|
} |
|
86
|
|
|
++$curLine; |
|
87
|
|
|
} |
|
88
|
|
|
++$curResource; |
|
89
|
|
|
} |
|
90
|
|
|
// no validation errors |
|
91
|
|
|
return []; |
|
92
|
|
|
} catch (Exceptions\DatapackageInvalidSourceException $e) { |
|
93
|
|
|
// failed to load the datapackage descriptor |
|
94
|
|
|
// return a list containing a single LOAD_FAILED validation error |
|
95
|
|
|
return [ |
|
96
|
|
|
new Validators\DatapackageValidationError( |
|
97
|
|
|
Validators\DatapackageValidationError::LOAD_FAILED, $e->getMessage() |
|
98
|
|
|
), |
|
99
|
|
|
]; |
|
100
|
|
|
} catch (Exceptions\DatapackageValidationFailedException $e) { |
|
101
|
|
|
// datapackage descriptor failed validation - return the validation errors |
|
102
|
|
|
return $e->validationErrors; |
|
103
|
|
|
} catch (Exceptions\ResourceValidationFailedException $e) { |
|
104
|
|
|
// resource descriptor failed validation - return the validation errors |
|
105
|
|
|
return [ |
|
106
|
|
|
new Validators\DatapackageValidationError( |
|
107
|
|
|
Validators\DatapackageValidationError::RESOURCE_FAILED_VALIDATION, |
|
108
|
|
|
[ |
|
109
|
|
|
'resource' => $curResource, |
|
110
|
|
|
'validationErrors' => $e->validationErrors, |
|
111
|
|
|
] |
|
112
|
|
|
), |
|
113
|
|
|
]; |
|
114
|
|
|
} catch (Exceptions\DataStreamOpenException $e) { |
|
115
|
|
|
// failed to open data stream |
|
116
|
|
|
return [ |
|
117
|
|
|
new Validators\DatapackageValidationError( |
|
118
|
|
|
Validators\DatapackageValidationError::DATA_STREAM_FAILURE, |
|
119
|
|
|
[ |
|
120
|
|
|
'resource' => $curResource, |
|
121
|
|
|
'line' => 0, |
|
122
|
|
|
'error' => $e->getMessage(), |
|
123
|
|
|
] |
|
124
|
|
|
), |
|
125
|
|
|
]; |
|
126
|
|
|
} catch (Exceptions\DataStreamValidationException $e) { |
|
127
|
|
|
// failed to validate the data stream |
|
128
|
|
|
return [ |
|
129
|
|
|
new Validators\DatapackageValidationError( |
|
130
|
|
|
Validators\DatapackageValidationError::DATA_STREAM_FAILURE, |
|
131
|
|
|
[ |
|
132
|
|
|
'resource' => $curResource, |
|
133
|
|
|
'line' => $curLine, |
|
134
|
|
|
'error' => $e->getMessage(), |
|
135
|
|
|
] |
|
136
|
|
|
), |
|
137
|
|
|
]; |
|
138
|
|
|
} |
|
139
|
|
|
} |
|
140
|
|
|
|
|
141
|
|
|
public static function registerDatapackageClass($datapackageClass) |
|
142
|
|
|
{ |
|
143
|
|
|
static::$registeredDatapackageClasses[] = $datapackageClass; |
|
144
|
|
|
} |
|
145
|
|
|
|
|
146
|
|
|
public static function clearRegisteredDatapackageClasses() |
|
147
|
|
|
{ |
|
148
|
|
|
static::$registeredDatapackageClasses = []; |
|
149
|
|
|
} |
|
150
|
|
|
|
|
151
|
|
|
/** |
|
152
|
|
|
* @param $descriptor |
|
153
|
|
|
* |
|
154
|
|
|
* @return BaseDatapackage::class |
|
|
|
|
|
|
155
|
|
|
*/ |
|
156
|
|
View Code Duplication |
public static function getDatapackageClass($descriptor) |
|
|
|
|
|
|
157
|
|
|
{ |
|
158
|
|
|
$datapackageClasses = array_merge( |
|
159
|
|
|
// custom classes |
|
160
|
|
|
static::$registeredDatapackageClasses, |
|
161
|
|
|
// core classes |
|
162
|
|
|
[ |
|
163
|
|
|
"frictionlessdata\\datapackage\\Datapackages\TabularDatapackage", |
|
164
|
|
|
"frictionlessdata\\datapackage\\Datapackages\DefaultDatapackage", |
|
165
|
|
|
] |
|
166
|
|
|
); |
|
167
|
|
|
$res = null; |
|
168
|
|
|
foreach ($datapackageClasses as $datapackageClass) { |
|
169
|
|
|
if (call_user_func([$datapackageClass, 'handlesDescriptor'], $descriptor)) { |
|
170
|
|
|
$res = $datapackageClass; |
|
171
|
|
|
break; |
|
172
|
|
|
} |
|
173
|
|
|
} |
|
174
|
|
|
if (!$res) { |
|
175
|
|
|
// not matched by any known classes |
|
176
|
|
|
$res = "frictionlessdata\\datapackage\\Datapackages\CustomDatapackage"; |
|
177
|
|
|
} |
|
178
|
|
|
|
|
179
|
|
|
return $res; |
|
180
|
|
|
} |
|
181
|
|
|
|
|
182
|
|
|
public static function registerResourceClass($resourceClass) |
|
183
|
|
|
{ |
|
184
|
|
|
static::$registeredResourceClasses[] = $resourceClass; |
|
185
|
|
|
} |
|
186
|
|
|
|
|
187
|
|
|
public static function clearRegisteredResourceClasses() |
|
188
|
|
|
{ |
|
189
|
|
|
static::$registeredResourceClasses = []; |
|
190
|
|
|
} |
|
191
|
|
|
|
|
192
|
|
|
/** |
|
193
|
|
|
* @param $descriptor |
|
194
|
|
|
* |
|
195
|
|
|
* @return BaseResource::class |
|
|
|
|
|
|
196
|
|
|
*/ |
|
197
|
|
View Code Duplication |
public static function getResourceClass($descriptor) |
|
|
|
|
|
|
198
|
|
|
{ |
|
199
|
|
|
$descriptor = Utils::objectify($descriptor); |
|
200
|
|
|
$resourceClasses = array_merge( |
|
201
|
|
|
// custom classes |
|
202
|
|
|
static::$registeredResourceClasses, |
|
203
|
|
|
// core classes |
|
204
|
|
|
[ |
|
205
|
|
|
'frictionlessdata\\datapackage\\Resources\\TabularResource', |
|
206
|
|
|
'frictionlessdata\\datapackage\\Resources\\DefaultResource', |
|
207
|
|
|
] |
|
208
|
|
|
); |
|
209
|
|
|
$res = null; |
|
210
|
|
|
foreach ($resourceClasses as $resourceClass) { |
|
211
|
|
|
if (call_user_func([$resourceClass, 'handlesDescriptor'], $descriptor)) { |
|
212
|
|
|
$res = $resourceClass; |
|
213
|
|
|
break; |
|
214
|
|
|
} |
|
215
|
|
|
} |
|
216
|
|
|
if (!$res) { |
|
217
|
|
|
// not matched by any known classes |
|
218
|
|
|
$res = 'frictionlessdata\\datapackage\\Resources\\CustomResource'; |
|
219
|
|
|
} |
|
220
|
|
|
|
|
221
|
|
|
return $res; |
|
222
|
|
|
} |
|
223
|
|
|
|
|
224
|
|
|
protected static $registeredDatapackageClasses = []; |
|
225
|
|
|
protected static $registeredResourceClasses = []; |
|
226
|
|
|
|
|
227
|
|
|
/** |
|
228
|
|
|
* allows extending classes to add custom sources |
|
229
|
|
|
* used by unit tests to add a mock http source. |
|
230
|
|
|
*/ |
|
231
|
|
|
protected static function normalizeHttpSource($source) |
|
232
|
|
|
{ |
|
233
|
|
|
return $source; |
|
234
|
|
|
} |
|
235
|
|
|
|
|
236
|
|
|
/** |
|
237
|
|
|
* allows extending classes to add custom sources |
|
238
|
|
|
* used by unit tests to add a mock http source. |
|
239
|
|
|
*/ |
|
240
|
|
|
protected static function isHttpSource($source) |
|
241
|
|
|
{ |
|
242
|
|
|
return Utils::isHttpSource($source); |
|
243
|
|
|
} |
|
244
|
|
|
|
|
245
|
|
|
/** |
|
246
|
|
|
* loads the datapackage descriptor from different sources |
|
247
|
|
|
* returns an object containing: |
|
248
|
|
|
* - the datapackage descriptor as native php object |
|
249
|
|
|
* - normalized basePath. |
|
250
|
|
|
* |
|
251
|
|
|
* @param $source |
|
252
|
|
|
* @param $basePath |
|
253
|
|
|
* |
|
254
|
|
|
* @return object |
|
255
|
|
|
* |
|
256
|
|
|
* @throws Exceptions\DatapackageInvalidSourceException |
|
257
|
|
|
*/ |
|
258
|
|
|
protected static function loadSource($source, $basePath) |
|
259
|
|
|
{ |
|
260
|
|
|
if (is_object($source)) { |
|
261
|
|
|
$descriptor = $source; |
|
262
|
|
|
} elseif (is_string($source)) { |
|
263
|
|
|
if (Utils::isJsonString($source)) { |
|
264
|
|
|
try { |
|
265
|
|
|
$descriptor = json_decode($source); |
|
266
|
|
|
} catch (\Exception $e) { |
|
267
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
|
268
|
|
|
'Failed to load source: '.json_encode($source).': '.$e->getMessage() |
|
269
|
|
|
); |
|
270
|
|
|
} |
|
271
|
|
|
} elseif (static::isHttpSource($source)) { |
|
272
|
|
|
try { |
|
273
|
|
|
$descriptor = json_decode(file_get_contents(static::normalizeHttpSource($source))); |
|
274
|
|
|
} catch (\Exception $e) { |
|
275
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
|
276
|
|
|
'Failed to load source: '.json_encode($source).': '.$e->getMessage() |
|
277
|
|
|
); |
|
278
|
|
|
} |
|
279
|
|
|
// http sources don't allow relative paths, hence basePath should remain null |
|
280
|
|
|
$basePath = null; |
|
281
|
|
|
} else { |
|
282
|
|
|
// not a json string and not a url - assume it's a file path |
|
283
|
|
|
if (empty($basePath)) { |
|
284
|
|
|
// no basePath |
|
285
|
|
|
// - assume source is the absolute path of the file |
|
286
|
|
|
// - set it's directory as the basePath |
|
287
|
|
|
$basePath = dirname($source); |
|
288
|
|
|
} else { |
|
289
|
|
|
// got a basePath |
|
290
|
|
|
// - try to prepend it to the source and see if such a file exists |
|
291
|
|
|
// - if not - assume it's an absolute path |
|
292
|
|
|
$absPath = $basePath.DIRECTORY_SEPARATOR.$source; |
|
293
|
|
|
if (file_exists($absPath)) { |
|
294
|
|
|
$source = $absPath; |
|
295
|
|
|
} |
|
296
|
|
|
} |
|
297
|
|
|
try { |
|
298
|
|
|
$descriptor = json_decode(file_get_contents($source)); |
|
299
|
|
|
} catch (\Exception $e) { |
|
300
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
|
301
|
|
|
'Failed to load source: '.json_encode($source).': '.$e->getMessage() |
|
302
|
|
|
); |
|
303
|
|
|
} |
|
304
|
|
|
} |
|
305
|
|
|
} else { |
|
306
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
|
307
|
|
|
'Invalid source: '.json_encode($source) |
|
308
|
|
|
); |
|
309
|
|
|
} |
|
310
|
|
|
|
|
311
|
|
|
return (object) ['descriptor' => $descriptor, 'basePath' => $basePath]; |
|
312
|
|
|
} |
|
313
|
|
|
} |
|
314
|
|
|
|
This check marks PHPDoc comments that could not be parsed by our parser. To see which comment annotations we can parse, please refer to our documentation on supported doc-types.