1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace frictionlessdata\datapackage; |
4
|
|
|
|
5
|
|
|
use frictionlessdata\datapackage\Datapackages\BaseDatapackage; |
6
|
|
|
use frictionlessdata\datapackage\Resources\BaseResource; |
7
|
|
|
use Alchemy\Zippy\Zippy; |
8
|
|
|
|
9
|
|
|
/** |
10
|
|
|
* datapackage and resource have different classes depending on the corresponding profile |
11
|
|
|
* this factory interface allows to validate and create object instances without having to check the profile first. |
12
|
|
|
*/ |
13
|
|
|
class Factory |
14
|
|
|
{ |
15
|
|
|
/** |
16
|
|
|
* how many lines to validate sample when validating data streams. |
17
|
|
|
*/ |
18
|
|
|
const VALIDATE_PEEK_LINES = 10; |
19
|
|
|
|
20
|
|
|
/** |
21
|
|
|
* load, validate and create a datapackage object |
22
|
|
|
* supports loading from the following sources: |
23
|
|
|
* - native PHP object containing the descriptor |
24
|
|
|
* - JSON encoded object |
25
|
|
|
* - URL (must be in either 'http' or 'https' schemes) |
26
|
|
|
* - local filesystem (POSIX) path. |
27
|
|
|
* - local or remote zip file |
28
|
|
|
* |
29
|
|
|
* @param mixed $source |
30
|
|
|
* @param null|string $basePath optional, required only if you want to use relative paths |
31
|
|
|
* |
32
|
|
|
* @return Datapackages\BaseDatapackage |
33
|
|
|
* |
34
|
|
|
* @throws Exceptions\DatapackageInvalidSourceException |
35
|
|
|
* @throws Exceptions\DatapackageValidationFailedException |
36
|
|
|
*/ |
37
|
|
|
public static function datapackage($source, $basePath = null) |
38
|
|
|
{ |
39
|
|
|
$source = static::loadSource($source, $basePath); |
40
|
|
|
$descriptor = $source->descriptor; |
41
|
|
|
$basePath = $source->basePath; |
42
|
|
|
$datapackageClass = static::getDatapackageClass($descriptor); |
43
|
|
|
$datapackage = new $datapackageClass($descriptor, $basePath); |
44
|
|
|
|
45
|
|
|
return $datapackage; |
46
|
|
|
} |
47
|
|
|
|
48
|
|
|
/** |
49
|
|
|
* create a resource object. |
50
|
|
|
* |
51
|
|
|
* @param object $descriptor |
52
|
|
|
* @param null|string $basePath |
53
|
|
|
* @param bool $skipValidations |
54
|
|
|
* |
55
|
|
|
* @return Resources\BaseResource |
56
|
|
|
* |
57
|
|
|
* @throws Exceptions\ResourceValidationFailedException |
58
|
|
|
*/ |
59
|
|
|
public static function resource($descriptor, $basePath = null, $skipValidations = false) |
60
|
|
|
{ |
61
|
|
|
$resourceClass = static::getResourceClass($descriptor); |
62
|
|
|
$resource = new $resourceClass($descriptor, $basePath, $skipValidations); |
63
|
|
|
|
64
|
|
|
return $resource; |
65
|
|
|
} |
66
|
|
|
|
67
|
|
|
/** |
68
|
|
|
* validates a given datapackage descriptor |
69
|
|
|
* will load all resources, and sample 10 lines of data from each data source. |
70
|
|
|
* |
71
|
|
|
* @param mixed $source datapackage source - same as in datapackage function |
72
|
|
|
* @param null|string $basePath same as in datapackage function |
73
|
|
|
* |
74
|
|
|
* @return Validators\DatapackageValidationError[] |
75
|
|
|
*/ |
76
|
|
|
public static function validate($source, $basePath = null) |
77
|
|
|
{ |
78
|
|
|
$curResource = 1; |
79
|
|
|
$curLine = null; |
80
|
|
|
try { |
81
|
|
|
$datapackage = static::datapackage($source, $basePath); |
82
|
|
|
foreach ($datapackage as $resource) { |
83
|
|
|
$curLine = 1; |
84
|
|
|
foreach ($resource as $line) { |
85
|
|
|
if ($curLine == self::VALIDATE_PEEK_LINES) { |
86
|
|
|
break; |
87
|
|
|
} |
88
|
|
|
++$curLine; |
89
|
|
|
} |
90
|
|
|
++$curResource; |
91
|
|
|
} |
92
|
|
|
// no validation errors |
93
|
|
|
return []; |
94
|
|
|
} catch (Exceptions\DatapackageInvalidSourceException $e) { |
95
|
|
|
// failed to load the datapackage descriptor |
96
|
|
|
// return a list containing a single LOAD_FAILED validation error |
97
|
|
|
return [ |
98
|
|
|
new Validators\DatapackageValidationError( |
99
|
|
|
Validators\DatapackageValidationError::LOAD_FAILED, $e->getMessage() |
100
|
|
|
), |
101
|
|
|
]; |
102
|
|
|
} catch (Exceptions\DatapackageValidationFailedException $e) { |
103
|
|
|
// datapackage descriptor failed validation - return the validation errors |
104
|
|
|
return $e->validationErrors; |
105
|
|
|
} catch (Exceptions\ResourceValidationFailedException $e) { |
106
|
|
|
// resource descriptor failed validation - return the validation errors |
107
|
|
|
return [ |
108
|
|
|
new Validators\DatapackageValidationError( |
109
|
|
|
Validators\DatapackageValidationError::RESOURCE_FAILED_VALIDATION, |
110
|
|
|
[ |
111
|
|
|
'resource' => $curResource, |
112
|
|
|
'validationErrors' => $e->validationErrors, |
113
|
|
|
] |
114
|
|
|
), |
115
|
|
|
]; |
116
|
|
|
} catch (Exceptions\DataStreamOpenException $e) { |
117
|
|
|
// failed to open data stream |
118
|
|
|
return [ |
119
|
|
|
new Validators\DatapackageValidationError( |
120
|
|
|
Validators\DatapackageValidationError::DATA_STREAM_FAILURE, |
121
|
|
|
[ |
122
|
|
|
'resource' => $curResource, |
123
|
|
|
'line' => 0, |
124
|
|
|
'error' => $e->getMessage(), |
125
|
|
|
] |
126
|
|
|
), |
127
|
|
|
]; |
128
|
|
|
} catch (Exceptions\DataStreamValidationException $e) { |
129
|
|
|
// failed to validate the data stream |
130
|
|
|
return [ |
131
|
|
|
new Validators\DatapackageValidationError( |
132
|
|
|
Validators\DatapackageValidationError::DATA_STREAM_FAILURE, |
133
|
|
|
[ |
134
|
|
|
'resource' => $curResource, |
135
|
|
|
'line' => $curLine, |
136
|
|
|
'error' => $e->getMessage(), |
137
|
|
|
] |
138
|
|
|
), |
139
|
|
|
]; |
140
|
|
|
} |
141
|
|
|
} |
142
|
|
|
|
143
|
|
|
public static function registerDatapackageClass($datapackageClass) |
144
|
|
|
{ |
145
|
|
|
static::$registeredDatapackageClasses[] = $datapackageClass; |
146
|
|
|
} |
147
|
|
|
|
148
|
|
|
public static function clearRegisteredDatapackageClasses() |
149
|
|
|
{ |
150
|
|
|
static::$registeredDatapackageClasses = []; |
151
|
|
|
} |
152
|
|
|
|
153
|
|
|
/** |
154
|
|
|
* @param $descriptor |
155
|
|
|
* |
156
|
|
|
* @return BaseDatapackage::class |
|
|
|
|
157
|
|
|
*/ |
158
|
|
View Code Duplication |
public static function getDatapackageClass($descriptor) |
|
|
|
|
159
|
|
|
{ |
160
|
|
|
$datapackageClasses = array_merge( |
161
|
|
|
// custom classes |
162
|
|
|
static::$registeredDatapackageClasses, |
163
|
|
|
// core classes |
164
|
|
|
[ |
165
|
|
|
"frictionlessdata\\datapackage\\Datapackages\TabularDatapackage", |
166
|
|
|
"frictionlessdata\\datapackage\\Datapackages\DefaultDatapackage", |
167
|
|
|
] |
168
|
|
|
); |
169
|
|
|
$res = null; |
170
|
|
|
foreach ($datapackageClasses as $datapackageClass) { |
171
|
|
|
if (call_user_func([$datapackageClass, 'handlesDescriptor'], $descriptor)) { |
172
|
|
|
$res = $datapackageClass; |
173
|
|
|
break; |
174
|
|
|
} |
175
|
|
|
} |
176
|
|
|
if (!$res) { |
177
|
|
|
// not matched by any known classes |
178
|
|
|
$res = "frictionlessdata\\datapackage\\Datapackages\CustomDatapackage"; |
179
|
|
|
} |
180
|
|
|
|
181
|
|
|
return $res; |
182
|
|
|
} |
183
|
|
|
|
184
|
|
|
public static function registerResourceClass($resourceClass) |
185
|
|
|
{ |
186
|
|
|
static::$registeredResourceClasses[] = $resourceClass; |
187
|
|
|
} |
188
|
|
|
|
189
|
|
|
public static function clearRegisteredResourceClasses() |
190
|
|
|
{ |
191
|
|
|
static::$registeredResourceClasses = []; |
192
|
|
|
} |
193
|
|
|
|
194
|
|
|
/** |
195
|
|
|
* @param $descriptor |
196
|
|
|
* |
197
|
|
|
* @return BaseResource::class |
|
|
|
|
198
|
|
|
*/ |
199
|
|
View Code Duplication |
public static function getResourceClass($descriptor) |
|
|
|
|
200
|
|
|
{ |
201
|
|
|
$descriptor = Utils::objectify($descriptor); |
202
|
|
|
$resourceClasses = array_merge( |
203
|
|
|
// custom classes |
204
|
|
|
static::$registeredResourceClasses, |
205
|
|
|
// core classes |
206
|
|
|
[ |
207
|
|
|
'frictionlessdata\\datapackage\\Resources\\TabularResource', |
208
|
|
|
'frictionlessdata\\datapackage\\Resources\\DefaultResource', |
209
|
|
|
] |
210
|
|
|
); |
211
|
|
|
$res = null; |
212
|
|
|
foreach ($resourceClasses as $resourceClass) { |
213
|
|
|
if (call_user_func([$resourceClass, 'handlesDescriptor'], $descriptor)) { |
214
|
|
|
$res = $resourceClass; |
215
|
|
|
break; |
216
|
|
|
} |
217
|
|
|
} |
218
|
|
|
if (!$res) { |
219
|
|
|
// not matched by any known classes |
220
|
|
|
$res = 'frictionlessdata\\datapackage\\Resources\\CustomResource'; |
221
|
|
|
} |
222
|
|
|
|
223
|
|
|
return $res; |
224
|
|
|
} |
225
|
|
|
|
226
|
|
|
protected static $registeredDatapackageClasses = []; |
227
|
|
|
protected static $registeredResourceClasses = []; |
228
|
|
|
|
229
|
|
|
/** |
230
|
|
|
* allows extending classes to add custom sources |
231
|
|
|
* used by unit tests to add a mock http source. |
232
|
|
|
*/ |
233
|
|
|
protected static function normalizeHttpSource($source) |
234
|
|
|
{ |
235
|
|
|
return $source; |
236
|
|
|
} |
237
|
|
|
|
238
|
|
|
/** |
239
|
|
|
* allows extending classes to add custom sources |
240
|
|
|
* used by unit tests to add a mock http source. |
241
|
|
|
*/ |
242
|
|
|
protected static function isHttpSource($source) |
243
|
|
|
{ |
244
|
|
|
return Utils::isHttpSource($source); |
245
|
|
|
} |
246
|
|
|
|
247
|
|
|
/** |
248
|
|
|
* loads the datapackage descriptor from different sources |
249
|
|
|
* returns an object containing: |
250
|
|
|
* - the datapackage descriptor as native php object |
251
|
|
|
* - normalized basePath. |
252
|
|
|
* |
253
|
|
|
* @param $source |
254
|
|
|
* @param $basePath |
255
|
|
|
* |
256
|
|
|
* @return object |
257
|
|
|
* |
258
|
|
|
* @throws Exceptions\DatapackageInvalidSourceException |
259
|
|
|
*/ |
260
|
|
|
protected static function loadSource($source, $basePath) |
261
|
|
|
{ |
262
|
|
|
if (is_object($source)) { |
263
|
|
|
$descriptor = $source; |
264
|
|
|
} elseif (is_string($source)) { |
265
|
|
|
if (Utils::isJsonString($source)) { |
266
|
|
|
try { |
267
|
|
|
$descriptor = json_decode($source); |
268
|
|
|
} catch (\Exception $e) { |
269
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
270
|
|
|
'Failed to load source: '.json_encode($source).': '.$e->getMessage() |
271
|
|
|
); |
272
|
|
|
} |
273
|
|
|
} elseif (static::isHttpSource($source)) { |
274
|
|
|
if (static::isHttpZipSource($source)) { |
275
|
|
|
return static::loadHttpZipSource($source); |
276
|
|
|
} else { |
277
|
|
|
try { |
278
|
|
|
$descriptor = json_decode(file_get_contents(static::normalizeHttpSource($source))); |
279
|
|
|
} catch (\Exception $e) { |
280
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
281
|
|
|
'Failed to load source: '.json_encode($source).': '.$e->getMessage() |
282
|
|
|
); |
283
|
|
|
} |
284
|
|
|
// http sources don't allow relative paths, hence basePath should remain null |
285
|
|
|
$basePath = null; |
286
|
|
|
} |
287
|
|
|
} else { |
288
|
|
|
// not a json string and not a url - assume it's a file path |
289
|
|
|
if (static::isFileZipSource($source)) { |
290
|
|
|
return static::loadFileZipSource($source); |
291
|
|
|
} else { |
292
|
|
|
if (empty($basePath)) { |
293
|
|
|
// no basePath |
294
|
|
|
// - assume source is the absolute path of the file |
295
|
|
|
// - set it's directory as the basePath |
296
|
|
|
$basePath = dirname($source); |
297
|
|
|
} else { |
298
|
|
|
// got a basePath |
299
|
|
|
// - try to prepend it to the source and see if such a file exists |
300
|
|
|
// - if not - assume it's an absolute path |
301
|
|
|
$absPath = $basePath.DIRECTORY_SEPARATOR.$source; |
302
|
|
|
if (file_exists($absPath)) { |
303
|
|
|
$source = $absPath; |
304
|
|
|
} |
305
|
|
|
} |
306
|
|
|
try { |
307
|
|
|
$descriptor = json_decode(file_get_contents($source)); |
308
|
|
|
} catch (\Exception $e) { |
309
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
310
|
|
|
'Failed to load source: '.json_encode($source).': '.$e->getMessage() |
311
|
|
|
); |
312
|
|
|
} |
313
|
|
|
} |
314
|
|
|
} |
315
|
|
|
} else { |
316
|
|
|
throw new Exceptions\DatapackageInvalidSourceException( |
317
|
|
|
'Invalid source: '.json_encode($source) |
318
|
|
|
); |
319
|
|
|
} |
320
|
|
|
|
321
|
|
|
return (object) ['descriptor' => $descriptor, 'basePath' => $basePath]; |
322
|
|
|
} |
323
|
|
|
|
324
|
|
|
protected static function isHttpZipSource($source) |
325
|
|
|
{ |
326
|
|
|
return (strtolower(substr($source, -4)) == '.zip'); |
327
|
|
|
} |
328
|
|
|
|
329
|
|
|
protected static function isFileZipSource($source) |
330
|
|
|
{ |
331
|
|
|
return (strtolower(substr($source, -4)) == '.zip'); |
332
|
|
|
} |
333
|
|
|
|
334
|
|
|
protected static function loadHttpZipSource($source) |
335
|
|
|
{ |
336
|
|
|
$tempfile = tempnam(sys_get_temp_dir(), 'datapackage-php'); |
337
|
|
|
unlink($tempfile); |
338
|
|
|
$tempfile.='.zip'; |
339
|
|
|
stream_copy_to_stream(fopen($source, 'r'), fopen($tempfile, 'w')); |
340
|
|
|
register_shutdown_function(function() use ($tempfile) {unlink($tempfile);}); |
341
|
|
|
return self::loadFileZipSource($tempfile); |
342
|
|
|
} |
343
|
|
|
|
344
|
|
|
protected static function loadFileZipSource($source) |
345
|
|
|
{ |
346
|
|
|
$zippy = Zippy::load(); |
347
|
|
|
$tempdir = tempnam(sys_get_temp_dir(), 'datapackage-php'); |
348
|
|
|
unlink($tempdir); |
349
|
|
|
mkdir($tempdir); |
350
|
|
|
register_shutdown_function(function() use ($tempdir) {Utils::removeDir($tempdir);}); |
351
|
|
|
$zippy->open($source)->extract($tempdir); |
352
|
|
|
if (!file_exists($tempdir."/datapackage.json")) { |
353
|
|
|
throw new Exceptions\DatapackageInvalidSourceException("zip file must contain a datappackage.json file"); |
354
|
|
|
} |
355
|
|
|
return static::loadSource($tempdir."/datapackage.json", $tempdir); |
356
|
|
|
} |
357
|
|
|
} |
358
|
|
|
|
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.