282 lines
6.9 KiB
PHP
282 lines
6.9 KiB
PHP
<?php
|
|
|
|
/**
|
|
* Logic around Solr field schema information.
|
|
*/
|
|
class SearchApiSolrField {
|
|
|
|
/**
|
|
* @var array
|
|
* Human-readable labels for Solr schema properties.
|
|
*/
|
|
public static $schemaLabels = array(
|
|
'I' => 'Indexed',
|
|
'T' => 'Tokenized',
|
|
'S' => 'Stored',
|
|
'M' => 'Multivalued',
|
|
'V' => 'TermVector Stored',
|
|
'o' => 'Store Offset With TermVector',
|
|
'p' => 'Store Position With TermVector',
|
|
'O' => 'Omit Norms',
|
|
'L' => 'Lazy',
|
|
'B' => 'Binary',
|
|
'C' => 'Compressed',
|
|
'f' => 'Sort Missing First',
|
|
'l' => 'Sort Missing Last',
|
|
);
|
|
|
|
/**
|
|
* @var stdclass
|
|
* The original field object.
|
|
*/
|
|
protected $field;
|
|
|
|
/**
|
|
* @var array
|
|
* An array of schema properties for this field. This will be a subset of
|
|
* the SearchApiSolrField::schemaLabels array.
|
|
*/
|
|
protected $schema;
|
|
|
|
/**
|
|
* Constructs a field information object.
|
|
*
|
|
* @param stdClass $field
|
|
* A field object from Solr's "Luke" servlet.
|
|
*/
|
|
public function __construct($field) {
|
|
$this->field = $field;
|
|
}
|
|
|
|
/**
|
|
* Gets the raw information of the field.
|
|
*
|
|
* @return object
|
|
* A field metadata object.
|
|
*/
|
|
public function getRaw() {
|
|
return $this->field;
|
|
}
|
|
|
|
/**
|
|
* Gets the type of the Solr field, according to the Solr schema.
|
|
*
|
|
* Note that field types like "text", "boolean", and "date" are conventions,
|
|
* but their presence and behavior are entirely determined by the particular
|
|
* schema.xml file used by a Solr core.
|
|
*
|
|
* @return string
|
|
* The type of the Solr field.
|
|
*/
|
|
public function getType() {
|
|
return $this->field->type;
|
|
}
|
|
|
|
/**
|
|
* Gets an array of field properties.
|
|
*
|
|
* @return array
|
|
* An array of properties describing the solr schema. The array keys are
|
|
* single-character codes, and the values are human-readable labels. This
|
|
* will be a subset of the SearchApiSolrField::schemaLabels array.
|
|
*/
|
|
public function getSchema() {
|
|
if (!isset($this->schema)) {
|
|
foreach (str_split(str_replace('-', '', $this->field->schema)) as $key) {
|
|
$this->schema[$key] = isset(self::$schemaLabels[$key]) ? self::$schemaLabels[$key] : $key;
|
|
}
|
|
}
|
|
return $this->schema;
|
|
}
|
|
|
|
/**
|
|
* Gets the "dynamic base" of this field.
|
|
*
|
|
* This typically looks like 'ss_*, and is used to aggregate fields based on
|
|
* "hungarian" naming conventions.
|
|
*
|
|
* @return string
|
|
* The mask describing the solr aggregate field, if there is one.
|
|
*/
|
|
public function getDynamicBase() {
|
|
return isset($this->field->dynamicBase) ? $this->field->dynamicBase : NULL;
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field may be suitable for use as a key field.
|
|
*
|
|
* Unfortunately, it seems like the best way to find an actual uniqueKey field
|
|
* according to Solr is to examine the Solr core's schema.xml.
|
|
*
|
|
* @return bool
|
|
* Whether the field is suitable for use as a key.
|
|
*/
|
|
public function isPossibleKey() {
|
|
return !$this->getDynamicBase()
|
|
&& !in_array($this->getType(), array('boolean', 'date', 'text'))
|
|
&& $this->isStored()
|
|
&& !$this->isMultivalued();
|
|
}
|
|
|
|
/**
|
|
* Determines whether a field is suitable for sorting.
|
|
*
|
|
* In order for a field to yield useful sorted results in Solr, it must be
|
|
* indexed, not multivalued, and not tokenized. It's ok if a field is
|
|
* tokenized and yields only one token, but there's no general way to check
|
|
* for that.
|
|
*
|
|
* @return bool
|
|
* Whether the field is suitable for sorting.
|
|
*/
|
|
public function isSortable() {
|
|
return $this->isIndexed()
|
|
&& !$this->isMultivalued()
|
|
&& !$this->isTokenized();
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field is indexed.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field is indexed, FALSE otherwise.
|
|
*/
|
|
public function isIndexed() {
|
|
$this->getSchema();
|
|
return isset($this->schema['I']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field is tokenized.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field is tokenized, FALSE otherwise.
|
|
*/
|
|
public function isTokenized() {
|
|
$this->getSchema();
|
|
return isset($this->schema['T']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field is stored.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field is stored, FALSE otherwise.
|
|
*/
|
|
public function isStored() {
|
|
$this->getSchema();
|
|
return isset($this->schema['S']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field is multi-valued.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field is multi-valued, FALSE otherwise.
|
|
*/
|
|
public function isMultivalued() {
|
|
$this->getSchema();
|
|
return isset($this->schema['M']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field has stored term vectors.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field has stored term vectors, FALSE otherwise.
|
|
*/
|
|
public function isTermVectorStored() {
|
|
$this->getSchema();
|
|
return isset($this->schema['V']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field has the "termOffsets" option set.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field has the "termOffsets" option set, FALSE otherwise.
|
|
*/
|
|
public function isStoreOffsetWithTermVector() {
|
|
$this->getSchema();
|
|
return isset($this->schema['o']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field has the "termPositions" option set.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field has the "termPositions" option set, FALSE otherwise.
|
|
*/
|
|
public function isStorePositionWithTermVector() {
|
|
$this->getSchema();
|
|
return isset($this->schema['p']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field omits norms when indexing.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field omits norms, FALSE otherwise.
|
|
*/
|
|
public function isOmitNorms() {
|
|
$this->getSchema();
|
|
return isset($this->schema['O']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field is lazy-loaded.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field is lazy-loaded, FALSE otherwise.
|
|
*/
|
|
public function isLazy() {
|
|
$this->getSchema();
|
|
return isset($this->schema['L']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field is binary.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field is binary, FALSE otherwise.
|
|
*/
|
|
public function isBinary() {
|
|
$this->getSchema();
|
|
return isset($this->schema['B']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field is compressed.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field is compressed, FALSE otherwise.
|
|
*/
|
|
public function isCompressed() {
|
|
$this->getSchema();
|
|
return isset($this->schema['C']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field sorts missing entries first.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field sorts missing entries first, FALSE otherwise.
|
|
*/
|
|
public function isSortMissingFirst() {
|
|
$this->getSchema();
|
|
return isset($this->schema['f']);
|
|
}
|
|
|
|
/**
|
|
* Determines whether this field sorts missing entries last.
|
|
*
|
|
* @return bool
|
|
* TRUE if the field sorts missing entries last, FALSE otherwise.
|
|
*/
|
|
public function isSortMissingLast() {
|
|
$this->getSchema();
|
|
return isset($this->schema['l']);
|
|
}
|
|
|
|
}
|