|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* @author The Midgard Project, http://www.midgard-project.org |
|
4
|
|
|
* @copyright The Midgard Project, http://www.midgard-project.org |
|
5
|
|
|
* @license http://www.gnu.org/licenses/lgpl.html GNU Lesser General Public License |
|
6
|
|
|
*/ |
|
7
|
|
|
|
|
8
|
|
|
namespace midcom\datamanager\indexer; |
|
9
|
|
|
|
|
10
|
|
|
use Symfony\Component\Form\FormView; |
|
11
|
|
|
use midcom\datamanager\datamanager; |
|
12
|
|
|
use midcom_services_indexer_document_midcom; |
|
13
|
|
|
use midcom_error; |
|
14
|
|
|
|
|
15
|
|
|
/** |
|
16
|
|
|
* This class is geared to ease indexing of datamanager driven documents. The |
|
17
|
|
|
* user invoking the indexing must have full read permissions to the object. |
|
18
|
|
|
* |
|
19
|
|
|
* <b>Basic indexing operation</b> |
|
20
|
|
|
* |
|
21
|
|
|
* This class uses a number of conventions, see below, to merge an existing |
|
22
|
|
|
* datamanager driven document into an indexing capable document. It requires |
|
23
|
|
|
* the callee to instantiate the datamanager, as this class would have no |
|
24
|
|
|
* idea where to take the schema database from. |
|
25
|
|
|
* |
|
26
|
|
|
* The RI (the GUID) from the base class is left untouched. |
|
27
|
|
|
* |
|
28
|
|
|
* <b>Indexing field defaults:</b> |
|
29
|
|
|
* |
|
30
|
|
|
* Unless you specify anything else explicitly in the schema, |
|
31
|
|
|
* the class will merge all text based fields together to form the <i>content</i> |
|
32
|
|
|
* field of the index record, to allow for easy searching of the document. |
|
33
|
|
|
* This will *not* include any metadata like keywords or summaries. |
|
34
|
|
|
* |
|
35
|
|
|
* If the schema contains a field <i>abstract</i>, it will also be used as |
|
36
|
|
|
* abstract field for the indexing process. In the same way, fields named |
|
37
|
|
|
* <i>title</i> or <i>author</i> will be used for the index document's title |
|
38
|
|
|
* or author respectively. The contents of abstract, title and author will also |
|
39
|
|
|
* be appended to the content field at the end of the object construction, |
|
40
|
|
|
* easing searching over this fields. |
|
41
|
|
|
* |
|
42
|
|
|
* If no abstract field is present, the first 200 characters of the content |
|
43
|
|
|
* area are used instead. |
|
44
|
|
|
* |
|
45
|
|
|
* Not all types can be indexed, check the various types in question about their |
|
46
|
|
|
* indexing capabilities. In general, if the system should index any non-text |
|
47
|
|
|
* field, it will use the CSV representation for implicit conversion. |
|
48
|
|
|
* |
|
49
|
|
|
* Metadata processing is done by the base class. |
|
50
|
|
|
* |
|
51
|
|
|
* <b>Document title:</b> |
|
52
|
|
|
* |
|
53
|
|
|
* You should either have an auto-indexed title field, or an assortment |
|
54
|
|
|
* of other fields manually assigned to index to the title field. |
|
55
|
|
|
* |
|
56
|
|
|
* <b>Configurability using the Datamanager schema:</b> |
|
57
|
|
|
* |
|
58
|
|
|
* You can decorate datamanager fields with various directives influencing |
|
59
|
|
|
* the indexing. See the Datamanager's schema documentation for details. |
|
60
|
|
|
* Basically, you can choose from the following indexing methods using the |
|
61
|
|
|
* key 'index_method' for each field: |
|
62
|
|
|
* |
|
63
|
|
|
* - The default <i>auto</i> mode will use the above guidelines to determine |
|
64
|
|
|
* the indexing destination automatically, adding data to the content, abstract, |
|
65
|
|
|
* title and author fields respectively. |
|
66
|
|
|
* - You can specify <i>abstract</i>, <i>content</i>, <i>title</i> or |
|
67
|
|
|
* <i>author</i> to indicate that the field should be used for the indicated |
|
68
|
|
|
* document fields. The content selector may be specified more than once, |
|
69
|
|
|
* indicating that the content of the relevant fields should be merged. |
|
70
|
|
|
* - Any date field can be indexed into its own, range-filterable field using |
|
71
|
|
|
* the <i>date</i> method. In this case, two document fields will be created |
|
72
|
|
|
* actually. One containing the filterable timestamp named directly after |
|
73
|
|
|
* the schema field, and a second one, having the _TS postfix which is set as |
|
74
|
|
|
* noindex containing the plain timestamp. |
|
75
|
|
|
* - Finally, you can explicitly index a field as a separate document field |
|
76
|
|
|
* using one of the five field types <i>keyword</i>, <i>unindexed</i>, |
|
77
|
|
|
* <i>unstored</i> or <i>text</i>. You can further control if the content |
|
78
|
|
|
* of these fields is also added to the main content field. This is useful |
|
79
|
|
|
* if you want to have fields searchable both by explicit field specification |
|
80
|
|
|
* and the default field for simpler searches. This is controlled by setting |
|
81
|
|
|
* the boolean key 'index_merge_with_content' in the field, which defaults |
|
82
|
|
|
* to true. |
|
83
|
|
|
* - <i>noindex</i> will prevent indexing of this field. |
|
84
|
|
|
* |
|
85
|
|
|
* The documents type is "midcom_datamanager". |
|
86
|
|
|
* |
|
87
|
|
|
* @see midcom_services_indexer |
|
88
|
|
|
*/ |
|
89
|
|
|
class document extends midcom_services_indexer_document_midcom |
|
90
|
|
|
{ |
|
91
|
|
|
/** |
|
92
|
|
|
* The datamanager instance of the document we need to index. |
|
93
|
|
|
* |
|
94
|
|
|
* @var datamanager |
|
95
|
|
|
*/ |
|
96
|
|
|
private $datamanager; |
|
97
|
|
|
|
|
98
|
|
|
/** |
|
99
|
|
|
* The constructor initializes the member variables and invokes |
|
100
|
|
|
* _process_datamanager, which will read and process the information |
|
101
|
|
|
* out of that instance. |
|
102
|
|
|
* |
|
103
|
|
|
* The document is ready for indexing after construction. On any |
|
104
|
|
|
* critical error, midcom_error is triggered. |
|
105
|
|
|
* |
|
106
|
|
|
* @param datamanager $datamanager The fully initialized datamanager instance to use |
|
107
|
|
|
*/ |
|
108
|
6 |
|
public function __construct($datamanager) |
|
109
|
|
|
{ |
|
110
|
6 |
|
parent::__construct($datamanager->get_storage()->get_value()); |
|
111
|
|
|
|
|
112
|
6 |
|
$this->_set_type('datamanager'); |
|
113
|
|
|
|
|
114
|
6 |
|
$this->datamanager = $datamanager; |
|
115
|
|
|
|
|
116
|
6 |
|
$this->process_datamanager(); |
|
117
|
6 |
|
$this->complete_fields(); |
|
118
|
6 |
|
} |
|
119
|
|
|
|
|
120
|
|
|
/** |
|
121
|
|
|
* Completes all fields which are not yet complete: |
|
122
|
|
|
* |
|
123
|
|
|
* content is completed with author, title and, if necessary, abstract. |
|
124
|
|
|
* |
|
125
|
|
|
* The title is set to the documents' URL in case that no title is set yet. The title |
|
126
|
|
|
* is not added to the content field in that case. |
|
127
|
|
|
*/ |
|
128
|
6 |
|
private function complete_fields() |
|
129
|
|
|
{ |
|
130
|
6 |
|
$this->content .= "{$this->author}\n{$this->title}\n"; |
|
131
|
|
|
|
|
132
|
|
|
// Add the abstract only if we haven't done so already. |
|
133
|
6 |
|
if (strstr($this->abstract, $this->content) === false) { |
|
134
|
6 |
|
$this->content .= "{$this->abstract}\n"; |
|
135
|
|
|
} |
|
136
|
|
|
|
|
137
|
6 |
|
if (! $this->title) { |
|
138
|
1 |
|
$this->title = $this->document_url; |
|
139
|
|
|
} |
|
140
|
6 |
|
} |
|
141
|
|
|
|
|
142
|
|
|
/** |
|
143
|
|
|
* Processes the information contained in the datamanager instance. |
|
144
|
|
|
* |
|
145
|
|
|
* The function iterates over the fields in the schema, and processes them |
|
146
|
|
|
* according to the rules given in the introduction. |
|
147
|
|
|
*/ |
|
148
|
6 |
|
private function process_datamanager() |
|
149
|
|
|
{ |
|
150
|
6 |
|
$renderer = $this->datamanager->get_renderer('view'); |
|
151
|
6 |
|
foreach ($renderer->get_view() as $name => $field) { |
|
152
|
6 |
|
$method = $field->vars['index_method']; |
|
153
|
6 |
|
if ($method == 'auto') { |
|
154
|
6 |
|
$method = $this->resolve_auto_method($field->vars['name']); |
|
155
|
|
|
} |
|
156
|
|
|
|
|
157
|
|
|
switch ($method) { |
|
158
|
6 |
|
case 'abstract': |
|
159
|
6 |
|
case 'title': |
|
160
|
6 |
|
case 'author': |
|
161
|
6 |
|
$this->{$method} = $renderer->widget($field); |
|
162
|
6 |
|
break; |
|
163
|
|
|
|
|
164
|
6 |
|
case 'content': |
|
165
|
6 |
|
$this->content .= $renderer->widget($field) . "\n"; |
|
166
|
6 |
|
break; |
|
167
|
|
|
|
|
168
|
6 |
|
case 'date': |
|
169
|
|
|
$this->add_as_date_field($field); |
|
170
|
|
|
break; |
|
171
|
|
|
|
|
172
|
6 |
|
case 'attachment': |
|
173
|
|
|
if (!empty($field->vars['value'])) { |
|
174
|
|
|
//only index the first attachment for now |
|
175
|
|
|
$attachment = array_shift($field->vars['value']); |
|
176
|
|
|
if ( !$attachment instanceof \midcom_db_attachment |
|
177
|
|
|
&& !empty($attachment['object'])) { |
|
178
|
|
|
//This is the form edit case |
|
179
|
|
|
//@todo: In create case, nothing is found currently |
|
180
|
|
|
$attachment = $attachment['object']; |
|
181
|
|
|
} |
|
182
|
|
|
if ($attachment instanceof \midcom_db_attachment) { |
|
183
|
|
|
$att_doc = new \midcom_services_indexer_document_attachment($attachment); |
|
184
|
|
|
$this->content .= $att_doc->content; |
|
185
|
|
|
$this->abstract .= $att_doc->abstract; |
|
186
|
|
|
} |
|
187
|
|
|
} |
|
188
|
|
|
|
|
189
|
|
|
break; |
|
190
|
|
|
|
|
191
|
6 |
|
case 'unstored': |
|
192
|
6 |
|
case 'unindexed': |
|
193
|
6 |
|
case 'text': |
|
194
|
6 |
|
case 'keyword': |
|
195
|
|
|
$data = $renderer->widget($field); |
|
196
|
|
|
$function = 'add_' . $method; |
|
197
|
|
|
$this->$function($name, $data); |
|
198
|
|
|
if ($field->vars['index_merge_with_content']) { |
|
199
|
|
|
$this->content .= $data . "\n"; |
|
200
|
|
|
} |
|
201
|
|
|
break; |
|
202
|
|
|
|
|
203
|
6 |
|
case 'noindex': |
|
204
|
6 |
|
break; |
|
205
|
|
|
|
|
206
|
|
|
default: |
|
207
|
6 |
|
throw new midcom_error(" Unknown indexing method {$method} for field {$name} discovered, aborting."); |
|
208
|
|
|
} |
|
209
|
|
|
} |
|
210
|
|
|
|
|
211
|
6 |
|
if ($this->abstract == '') { |
|
212
|
6 |
|
$this->abstract = $this->html2text($this->content); |
|
213
|
6 |
|
if (mb_strlen($this->abstract) > 200) { |
|
214
|
|
|
$this->abstract = mb_substr($this->abstract, 0, 200) . ' ...'; |
|
215
|
|
|
} |
|
216
|
|
|
} |
|
217
|
6 |
|
} |
|
218
|
|
|
|
|
219
|
|
|
/** |
|
220
|
|
|
* This function tries to convert the $field into a date |
|
221
|
|
|
* representation. Unixdate fields are used directly (localtime is used, |
|
222
|
|
|
* not GMT), other fields will be parsed with strtodate. |
|
223
|
|
|
* |
|
224
|
|
|
* Invalid strings which are not parseable using strtotime will be |
|
225
|
|
|
* stored as a "0" timestamp. |
|
226
|
|
|
* |
|
227
|
|
|
* Be aware, that this will work only for current dates in range of an |
|
228
|
|
|
* UNIX timestamp. For all other cases you should use an ISO 8601 representation, |
|
229
|
|
|
* which should work as well with Lucene range queries. |
|
230
|
|
|
* |
|
231
|
|
|
* @param FormView $field The field that should be stored |
|
232
|
|
|
*/ |
|
233
|
|
|
private function add_as_date_field(FormView $field) |
|
234
|
|
|
{ |
|
235
|
|
|
if (is_array($field->vars['value']) && array_key_exists('date', $field->vars['value'])) { |
|
236
|
|
|
$timestamp = 0; |
|
237
|
|
|
if (!empty($field->vars['value']['date'])) { |
|
238
|
|
|
$timestamp = $field->vars['value']['date']->format('U'); |
|
239
|
|
|
} |
|
240
|
|
|
$this->add_date_pair($field->vars['name'], $timestamp); |
|
241
|
|
|
} else { |
|
242
|
|
|
$string = (string) $field->vars['value']; |
|
243
|
|
|
$timestamp = strtotime($string); |
|
244
|
|
|
if ($timestamp === -1) { |
|
245
|
|
|
debug_add("The string representation of the field {$field->vars['name']} could not be parsed into a timestamp; treating as 0.", MIDCOM_LOG_INFO); |
|
246
|
|
|
debug_print_r('String representation was:', $string); |
|
247
|
|
|
$timestamp = 0; |
|
248
|
|
|
} |
|
249
|
|
|
$this->add_date_pair($field->vars['name'], $timestamp); |
|
250
|
|
|
} |
|
251
|
|
|
} |
|
252
|
|
|
|
|
253
|
|
|
/** |
|
254
|
|
|
* @param string $name The field name |
|
255
|
|
|
* @return string index method |
|
256
|
|
|
*/ |
|
257
|
6 |
|
private function resolve_auto_method($name) |
|
258
|
|
|
{ |
|
259
|
6 |
|
if (in_array($name, ['abstract', 'title', 'author'])) { |
|
260
|
6 |
|
return $name; |
|
261
|
|
|
} |
|
262
|
6 |
|
return 'content'; |
|
263
|
|
|
} |
|
264
|
|
|
} |
|
265
|
|
|
|
flack /
openpsa
