Oihana Php Arango

DocumentsFacetCountsTrait uses trait:short, trait:short

Computes per-value **facet counts** for several dimensions at once, returned alongside (not replacing) the document list.

The dimensions are facet keys from Arango::FACETS; the buckets are counted over the same conjunctive filter as list(). See FacetCountsQueryTrait for the generated AQL.

Constants

FACET_COUNT_ITEM : string = 'item': The unwind loop variable for array-membership facets (kept distinct from {@see FacetCountsQueryTrait::FACET_COUNT_VALUE} to avoid a name collision).
FACET_COUNT_VALUE : string = 'value': The bucket value attribute name in the returned rows (`{ value, count }`).

Properties

$collection : string|null: The default collection name.
$indexes : array<string|int, mixed>|null: The declared indexes of the collection (the `AQL::INDEXES` list of {@see IndexOptions} or raw definitions). Retained at initialization — whether the lazy provisioning ran or not — so the declaration can be compared with the server later ({@see DoctorTrait::diagnose()}).
$arangodb : ArangoDB|null: The ArangoDB database reference.
$type : int: Indicates the type of the collection when is created (document or edge).

Methods

analyzerExists() : bool: Checks if an analyzer exists on the server (built-in analyzers are always reported).
buildFacetCountsQuery() : string: Builds the multi-`LET` facet-counts query, or an empty string when nothing is countable.
collectionCreate() : bool: Creates a new collection if not exist.
collectionDrop() : bool: Drops a collection if exist.
collectionExists() : bool: Check if collection exists
collectionRename() : bool: Renames a collection if exist.
collectionTruncate() : bool: Truncate a collection if exist.
createIndex() : array<string|int, mixed>|null: Creates an index on a collection on the server.
debugQuery() : void: Debug the passed-in query and binds variables.
explain() : ExplainResult: Explains an AQL query — returns the optimizer's execution plan as a typed {@see ExplainResult} (rules applied, collections, estimated cost, indexes actually used) **without executing the query**.
facetCount() : array<string, int>: Flattens one dimension's facet buckets into a `value => count` map.
facetCounts() : array<string, array<int, object|array<string|int, mixed>>>: Returns per-value bucket counts for the requested facet dimensions.
foundRows() : int: For a SELECT with a LIMIT clause, returns the number of rows that would be returned were there no LIMIT clause.
getDatabase() : ArangoDB: Returns the ArangoDB database singleton reference.
getDocuments() : array<string|int, mixed>: Prepare, execute and returns an array of all documents with the passed-in AQL query.
getExtra() : array<string|int, mixed>: Returns the AQL current extra datas.
getFirstResult() : mixed: Prepare, execute and returns the first result of the passed-in AQL query.
getObject() : object|null: Prepare, execute and returns an object with the passed-in AQL query.
getProfile() : ProfileResult: Returns the typed profile of the last profiled query run (per-phase timings, {@see ExecutionStats}, warnings).
getResult() : array<string|int, mixed>|null: Prepare, execute and returns an array with the passed-in AQL query.
getStats() : ExecutionStats: Returns the typed execution statistics of the last query (scanned / filtered / time / memory …). Most meaningful right after a profiled `list()` / `get()` (see {@see Arango::PROFILE}).
initializeCollection() : static: Sets the internal collection reference.
initializeDatabase() : static: Set the internal arangoDB reference.
initializeIndexes() : static: Sets the declared indexes of the collection from the init definition, normalizing a single {@see IndexOptions} value to a one-element list (a raw array always stays the index list) — so every consumer sees a plain `IndexOptions[]`: the {@see initializeCollection()} lazy provisioning and the {@see DoctorTrait} diagnose/repair diffs.
prepareAndExecute() : static: Prepare and execute an ArangoDB AQL query.
registerProperty() : void: Register a specific dynamic property in the binds and values collection to generates a query.
streamDocuments() : Generator<string|int, mixed>: Prepare, execute and returns a generator of documents with the passed-in AQL query.
viewCreate() : bool: Creates an `arangosearch` View if it does not already exist.
viewExists() : bool: Checks if a View exists.
firstRowAsArray() : array<string|int, mixed>: Runs a single-row aggregate query and normalizes its result to an array.
profileOptions() : array<string|int, mixed>: Merges the cursor `profile` option into `$options` when the `$init` array requests profiling via {@see Arango::PROFILE} (`true` → profile level 2, or an explicit integer level). Returns `$options` unchanged otherwise.
buildFacetCountSubquery() : string|null: Builds one dimension's counting sub-query, or null for an unsupported type.
collectOptionalBindNames() : array<int, string>: Collects every {@see AqlBindReference} name declared anywhere in one or more declaration trees — the projections (`$fields` / `$skinFields`) **and** the relation registries (`$edges` / `$joins`), typically inside a `Field::WHERE` / `Field::WHEN` condition. These are the *optional* binds : a skin (or an explicit `?fields`) can drop the carrying field — or the whole relation — from the projection, so the declared bind may not reach the final query text.
facetCountCollect() : array{0: string, 1: string, 2: string}: The shared `COLLECT value = <expr> … SORT count DESC RETURN { value, count }` tail.

FACET_COUNT_ITEM

The unwind loop variable for array-membership facets (kept distinct from {@see FacetCountsQueryTrait::FACET_COUNT_VALUE} to avoid a name collision).


    private
        string
    FACET_COUNT_ITEM
    = 'item'

FACET_COUNT_VALUE

The bucket value attribute name in the returned rows (`{ value, count }`).


    private
        string
    FACET_COUNT_VALUE
    = 'value'

$collection

The default collection name.


        public
            string|null
    $collection

$indexes

The declared indexes of the collection (the `AQL::INDEXES` list of {@see IndexOptions} or raw definitions). Retained at initialization — whether the lazy provisioning ran or not — so the declaration can be compared with the server later ({@see DoctorTrait::diagnose()}).


        public
            array<string|int, mixed>|null
    $indexes
     = null

$arangodb

The ArangoDB database reference.


        protected
            ArangoDB|null
    $arangodb
     = null

$type

Indicates the type of the collection when is created (document or edge).


        protected
            int
    $type

analyzerExists()

Checks if an analyzer exists on the server (built-in analyzers are always reported).


    public
                    analyzerExists(string $name) : bool

Parameters

$name : string: The name of the analyzer.

Return values

bool

buildFacetCountsQuery()

Builds the multi-`LET` facet-counts query, or an empty string when nothing is countable.


    public
                    buildFacetCountsQuery([array<string|int, mixed> $init = [] ][, array<string|int, mixed> &$bindVars = [] ][, string $docRef = AQL::DOC ]) : string

Parameters

$init : array<string|int, mixed> = []: The list query options (Arango::FACET_COUNTS holds the dimensions).
$bindVars : array<string|int, mixed> = []: The bind variables, populated by reference.
$docRef : string = AQL::DOC: The document reference.

Return values

string —

The compiled AQL query, or an empty string.

collectionCreate()

Creates a new collection if not exist.


    public
                    collectionCreate(string $name[, array<string|int, mixed> $options = [] ]) : bool

Parameters

$name : string

The name of the new collection

$options : array<string|int, mixed> = []

an array of options.

Options are:

'type' - 2 -> normal collection, 3 -> edge-collection

'waitForSync' - if set to true, then all removal operations will instantly be synchronised to disk / If this is not specified, then the collection's default sync behavior will be applied.

'isSystem' - false->user collection(default), true->system collection .

'keyOptions' - key options to use.

'distributeShardsLike' - name of prototype collection for identical sharding.

'numberOfShards' - number of shards for the collection.

'replicationFactor' - number of replicas to keep (default: 1).

'writeConcern' - minimum number of replicas to be successful when writing (default: 1).

'shardKeys' - array of shard key attributes.

'shardingStrategy' - sharding strategy to use in cluster.

'smartJoinAttribute' - attribute name for smart joins (if not shard key).

'schema' - collection schema.

Return values

bool —

Returns true if the new collection is created.

collectionDrop()

Drops a collection if exist.


    public
                    collectionDrop(string $name) : bool

Parameters

$name : string: The name of the new collection

Return values

bool —

Returns true if the new collection is dropped.

collectionExists()

Check if collection exists


    public
                    collectionExists(string $name) : bool

Parameters

$name : string: The name of the collection

Return values

bool

collectionRename()

Renames a collection if exist.


    public
                    collectionRename(string $oldName, string $name) : bool

Parameters

$oldName : string: The old name of the collection
$name : string: The new name of the collection

Return values

bool —

Returns true if the collection is renamed.

collectionTruncate()

Truncate a collection if exist.


    public
                    collectionTruncate(string $name) : bool

Parameters

$name : string: The name of the collection to truncate.

Return values

bool

createIndex()

Creates an index on a collection on the server.


    public
                    createIndex(string|Collection $collection, array<string|int, mixed>|IndexOptions $indexOptions) : array<string|int, mixed>|null

Parameters

$collection : string|Collection: Collection name or Collection client handle.
$indexOptions : array<string|int, mixed>|IndexOptions: An IndexOptions definition or an associative array of options for the index like array('type' => 'persistent', 'fields' => ['id','additionalType'], 'sparse' => false)

Return values

array<string|int, mixed>|null —

The server response of the created index or null

debugQuery()

Debug the passed-in query and binds variables.


    public
                    debugQuery(string $method, string $query, array<string|int, mixed>|null $binds) : void

Parameters

$method : string
$query : string
$binds : array<string|int, mixed>|null

explain()

Explains an AQL query — returns the optimizer's execution plan as a typed {@see ExplainResult} (rules applied, collections, estimated cost, indexes actually used) **without executing the query**.


    public
                    explain(AqlQuery|string $query[, array<string, mixed> $bindVars = [] ][, array<string, mixed> $options = [] ]) : ExplainResult

Parameters

$query : AqlQuery|string: The AQL query to explain.
$bindVars : array<string, mixed> = []: Bind variables (omit when $query is an AqlQuery).
$options : array<string, mixed> = []: Explain options (allPlans, optimizer.rules, …).

Return values

ExplainResult

facetCount()

Flattens one dimension's facet buckets into a `value => count` map.


    public
                    facetCount(string $dimension[, array<string|int, mixed> $init = [] ]) : array<string, int>

A thin projection over DocumentsFacetCountsTrait::facetCounts() for the common "count by <dimension>" need: instead of the multi-dimension [ { value, count }, ... ] bucket lists, it returns a single flat map for one dimension, preserving the buckets' count-descending order. Malformed rows (missing value) are skipped and each count is cast to int.

The dimension must be a declared facet (Arango::FACETS); an unknown key produces an empty map (nothing is countable). The sum of the returned counts equals the number of matching documents for a scalar facet; for a multi-valued (array-membership) facet a document contributes to several buckets, so the sum over-counts — pair DocumentsFacetCountsTrait::facetCounts() with a dedicated count() when an exact total is needed in that case.

Parameters

$dimension : string: The facet key to count.
$init : array<string|int, mixed> = []: The filtered-set options (same keys as DocumentsFacetCountsTrait::facetCounts()).

Return values

array<string, int> —

A value => count map (empty when nothing countable).

facetCounts()

Returns per-value bucket counts for the requested facet dimensions.


    public
                    facetCounts([array<string|int, mixed> $init = [] ]) : array<string, array<int, object|array<string|int, mixed>>>

Parameters

$init : array<string|int, mixed> = []: The list options. Arango::FACET_COUNTS holds the facet keys to count; the other keys (active, facets, filter, search, conditions) define the filtered set.

Return values

array<string, array<int, object|array<string|int, mixed>>> —

A dimension => [ { value, count }, ... ] map (empty when nothing countable).

foundRows()

For a SELECT with a LIMIT clause, returns the number of rows that would be returned were there no LIMIT clause.


    public
                    foundRows() : int

Return values

int

getDatabase()

Returns the ArangoDB database singleton reference.


    public
                    getDatabase() : ArangoDB

Return values

ArangoDB

getDocuments()

Prepare, execute and returns an array of all documents with the passed-in AQL query.


    public
                    getDocuments(string $query[, array<string|int, mixed> $bindVars = [] ][, array<string|int, mixed> $options = [] ][, bool $raw = false ][, null|SchemaResolver|Closure|string $schema = null ][, array<string|int, mixed> $context = [] ]) : array<string|int, mixed>

Parameters

$query : string: The AQL query string to execute
$bindVars : array<string|int, mixed> = []: Optional bind variables for the query
$options : array<string|int, mixed> = []: Optional execution options
$raw : bool = false: If true, returns the object raw (no schema or alter applied)
$schema : null|SchemaResolver|Closure|string = null: The optional class name to map the document.
$context : array<string|int, mixed> = []: Optional opaque context forwarded to AlterDocumentTrait::alter() and thus to the Alter::MAP callbacks (e.g. the originating $init, a skin, a locale…). Default [].

Return values

array<string|int, mixed>

getExtra()

Returns the AQL current extra datas.


    public
                    getExtra() : array<string|int, mixed>

Return values

array<string|int, mixed>

getFirstResult()

Prepare, execute and returns the first result of the passed-in AQL query.


    public
                    getFirstResult(string $query[, array<string|int, mixed> $bindVars = [] ][, array<string|int, mixed> $options = [] ][, bool $raw = false ][, null|SchemaResolver|Closure|string $schema = null ][, array<string|int, mixed> $context = [] ]) : mixed

Parameters

$query : string: The AQL query string to execute
$bindVars : array<string|int, mixed> = []: Optional bind variables for the query
$options : array<string|int, mixed> = []: Optional execution options
$raw : bool = false: If true, returns the object raw (no schema or alter applied)
$schema : null|SchemaResolver|Closure|string = null: The optional class name to map the document.
$context : array<string|int, mixed> = []: Optional opaque context forwarded to AlterDocumentTrait::alter() and thus to the Alter::MAP callbacks (e.g. the originating $init, a skin, a locale…). Default [].

getObject()

Prepare, execute and returns an object with the passed-in AQL query.


    public
                    getObject(string $query[, array<string|int, mixed> $bindVars = [] ][, array<string|int, mixed> $options = [] ][, bool $raw = false ][, null|SchemaResolver|Closure|string $schema = null ][, array<string|int, mixed> $context = [] ]) : object|null

Parameters

$query : string: The AQL query string to execute
$bindVars : array<string|int, mixed> = []: Optional bind variables for the query
$options : array<string|int, mixed> = []: Optional execution options
$raw : bool = false: If true, returns the object raw (no schema or alter applied)
$schema : null|SchemaResolver|Closure|string = null: The optional class name to map the document.
$context : array<string|int, mixed> = []: Optional opaque context forwarded to AlterDocumentTrait::alter() and thus to the Alter::MAP callbacks (e.g. the originating $init, a skin, a locale…). Default [].

Return values

object|null

getProfile()

Returns the typed profile of the last profiled query run (per-phase timings, {@see ExecutionStats}, warnings).


    public
                    getProfile() : ProfileResult

Return values

ProfileResult

getResult()

Prepare, execute and returns an array with the passed-in AQL query.


    public
                    getResult(string $query[, array<string|int, mixed> $bindVars = [] ][, array<string|int, mixed> $options = [] ][, bool $raw = false ][, null|SchemaResolver|Closure|string $schema = null ][, array<string|int, mixed> $context = [] ]) : array<string|int, mixed>|null

Parameters

$query : string: The AQL query string to execute
$bindVars : array<string|int, mixed> = []: Optional bind variables for the query
$options : array<string|int, mixed> = []: Optional execution options
$raw : bool = false: If true, returns the object raw (no schema or alter applied)
$schema : null|SchemaResolver|Closure|string = null: The optional class name to map the document.
$context : array<string|int, mixed> = []: Optional opaque context forwarded to AlterDocumentTrait::alter() and thus to the Alter::MAP callbacks (e.g. the originating $init, a skin, a locale…). Default [].

Return values

array<string|int, mixed>|null

getStats()

Returns the typed execution statistics of the last query (scanned / filtered / time / memory …). Most meaningful right after a profiled `list()` / `get()` (see {@see Arango::PROFILE}).


    public
                    getStats() : ExecutionStats

Return values

ExecutionStats

initializeCollection()

Sets the internal collection reference.


    public
                    initializeCollection([array<string|int, mixed> $init = [] ][, int $type = CollectionType::DOCUMENT ]) : static

Parameters

$init : array<string|int, mixed> = []

The options to lazy creates the collection (document or edge) or not.

collection (string) Indicates if the name of the collection.
indexes (array) The optional list of indexes to creates (if not exist and lazy).
lazy (bool) Indicates if the collection is created if not exist — resolved through LazyTrait::isLazy(), so a lazy entry defined in the DI container always wins (orchestration kill-switch), then this init key, then the property default.
options (array) The options are:
- 'waitForSync' : if set to true, then all removal operations will instantly be synchronised to disk / If this is not specified, then the collection's default sync behavior will be applied.
- 'isSystem' : false->user collection(default), true->system collection .
- 'keyOptions' : key options to use.
- 'distributeShardsLike' : name of prototype collection for identical sharding.
- 'numberOfShards' : number of shards for the collection.
- 'replicationFactor' : number of replicas to keep (default: 1).
- 'writeConcern' : minimum number of replicas to be successful when writing (default: 1).
- 'shardKeys' : array of shard key attributes.
- 'shardingStrategy' : sharding strategy to use in cluster.
- 'smartJoinAttribute' : attribute name for smart joins (if not shard key).
- 'schema' : collection schema.

$type : int = CollectionType::DOCUMENT

The default type of the collection (Default -> 'document' [2] )

Return values

static

initializeDatabase()

Set the internal arangoDB reference.


    public
                    initializeDatabase([array<string|int, mixed> $init = [] ][, ContainerInterface|null $container = null ]) : static

Parameters

$init : array<string|int, mixed> = []
$container : ContainerInterface|null = null

Return values

static

initializeIndexes()

Sets the declared indexes of the collection from the init definition, normalizing a single {@see IndexOptions} value to a one-element list (a raw array always stays the index list) — so every consumer sees a plain `IndexOptions[]`: the {@see initializeCollection()} lazy provisioning and the {@see DoctorTrait} diagnose/repair diffs.


    public
                    initializeIndexes([array<string|int, mixed> $init = [] ]) : static

Parameters

$init : array<string|int, mixed> = []: The init definition (reads the Arango::INDEXES key).

Return values

static

prepareAndExecute()

Prepare and execute an ArangoDB AQL query.


    public
                    prepareAndExecute(string $query[, array<string|int, mixed> $bindVars = [] ][, array<string|int, mixed> $options = [] ][, array<string|int, mixed>|null $optionalBinds = null ]) : static

A bind variable may be optional : supplied by the caller (through Arango::BINDS) yet not guaranteed to appear in the final query text. This happens with a conditional projection — e.g. an aqlBindRef() inside a Field::WHERE / Field::WHEN carried by a field that the active skin (or an explicit ?fields) does not project. ArangoDB rejects a query that declares a bind it never references ("bind parameter 'x' was not declared in the query"), so such a leftover bind must be dropped before execution.

$optionalBinds names the binds that are allowed to be absent :

it is opt-in and bounded — a bind whose name is not listed is never touched, so a query that uses no conditional bind behaves exactly as before ;
a listed bind is kept only when the query actually references it (@name, word-boundary matched to avoid a prefix collision such as @offers matching inside @offersScope) ;
passing [] disables the pruning explicitly.

When $optionalBinds is null (the default), the list is derived from the model's own declarations — every AqlBindReference found anywhere in $this->fields / $this->skinFields and in the relation registries $this->edges / $this->joins (see collectOptionalBindNames()). The registries matter as much as the projections: a relation definition is a declaration tree of its own, it can carry a conditional bind (in its own AQL::FIELDS sub-projection, or in a definition-level predicate), and a skin dropping the relation leaves that bind unreferenced exactly the same way. The source of truth is the same aqlBindRef() that declared the conditional bind, so no host wiring is required, and over-listing a bind that turns out to be always present is inert (a referenced bind is always kept). Because this runs at the single execution chokepoint, it protects every query method (get(), list(), count(), exist(), edges…) uniformly.

Parameters

$query : string: The AQL query string to execute.
$bindVars : array<string|int, mixed> = []: Optional bind variables for the query.
$options : array<string|int, mixed> = []: Optional execution options.
$optionalBinds : array<string|int, mixed>|null = null: Names of binds allowed to be absent from the query. null derives the list from the model field definitions and relation registries ; [] disables the pruning.

Return values

static

registerProperty()


    public
                    registerProperty(string $name, mixed $value, array<string|int, mixed> &$binds, array<string|int, mixed> &$values[, string $prefix = Char::EMPTY ][, string $separator = ': ' ]) : void

Parameters

$name : string
$value : mixed
$binds : array<string|int, mixed>
$values : array<string|int, mixed>
$prefix : string = Char::EMPTY
$separator : string = ': '

streamDocuments()

Prepare, execute and returns a generator of documents with the passed-in AQL query.


    public
                    streamDocuments(string $query[, array<string|int, mixed> $bindVars = [] ][, array<string|int, mixed> $options = [] ][, bool $raw = false ][, null|SchemaResolver|Closure|string $schema = null ][, array<string|int, mixed> $context = [] ]) : Generator<string|int, mixed>

Documents are yielded one by one, allowing efficient memory usage for large result sets.

Parameters

$query : string: The AQL query string to execute
$bindVars : array<string|int, mixed> = []: Optional bind variables for the query
$options : array<string|int, mixed> = []: Optional execution options
$raw : bool = false: If true, returns the object raw (no schema or alter applied)
$schema : null|SchemaResolver|Closure|string = null: The optional class name to map the document.
$context : array<string|int, mixed> = []: Optional opaque context forwarded to AlterDocumentTrait::alter() and thus to the Alter::MAP callbacks (e.g. the originating $init, a skin, a locale…). Default [].

Return values

Generator<string|int, mixed> —

Generator yielding documents one by one

viewCreate()

Creates an `arangosearch` View if it does not already exist.


    public
                    viewCreate(string $name[, array<string|int, mixed> $links = [] ][, array<string|int, mixed> $options = [] ]) : bool

Parameters

$name : string: The name of the new View.
$links : array<string|int, mixed> = []: Per-collection link map (collection name → link definition).
$options : array<string|int, mixed> = []: Extra arangosearch options forwarded verbatim.

Return values

bool —

Returns true if the new View has been created.

viewExists()

Checks if a View exists.


    public
                    viewExists(string $name) : bool

Parameters

$name : string: The name of the View.

Return values

bool

firstRowAsArray()

Runs a single-row aggregate query and normalizes its result to an array.


    protected
                    firstRowAsArray(string $query[, array<string|int, mixed> $bindVars = [] ]) : array<string|int, mixed>

The aggregate queries — DocumentsBoundsTrait::bounds(), DocumentsFacetCountsTrait::facetCounts() — all return one row holding one entry per requested dimension, and all face the same two shapes: their builder yields an empty string when nothing is computable, and the row itself comes back as an object or as an array depending on how the driver decoded it. Both are handled here rather than at each call site.

The result is read raw — no schema, no alter() — because the row is a map of computed values, not a document.

Parameters

$query : string: The compiled aggregate query, or an empty string when there is nothing to compute.
$bindVars : array<string|int, mixed> = []: The bind variables of that query.

Return values

array<string|int, mixed> —

The single row as an associative array; an empty array when the query is empty or the row is neither an object nor an array.

profileOptions()

Merges the cursor `profile` option into `$options` when the `$init` array requests profiling via {@see Arango::PROFILE} (`true` → profile level 2, or an explicit integer level). Returns `$options` unchanged otherwise.


    protected
                    profileOptions(array<string|int, mixed> $init[, array<string|int, mixed> $options = [] ]) : array<string|int, mixed>

Parameters

$init : array<string|int, mixed>: The model input array (list() / get()).
$options : array<string|int, mixed> = []: The cursor options to augment.

Return values

array<string|int, mixed>

buildFacetCountSubquery()

Builds one dimension's counting sub-query, or null for an unsupported type.


    private
                    buildFacetCountSubquery(array<string|int, mixed> $facet, string $key, string $for, string|null $filter, string $docRef) : string|null

Parameters

$facet : array<string|int, mixed>: The facet definition (Facet::TYPE, Facet::PROPERTY).
$key : string: The facet key (default property).
$for : string: The pre-built FOR segment shared by every dimension — the bound collection, or the bound View with its SEARCH segment when the View search is active.
$filter : string|null: The shared FILTER clause.
$docRef : string: The document reference.

Return values

string|null

collectOptionalBindNames()

Collects every {@see AqlBindReference} name declared anywhere in one or more declaration trees — the projections (`$fields` / `$skinFields`) **and** the relation registries (`$edges` / `$joins`), typically inside a `Field::WHERE` / `Field::WHEN` condition. These are the *optional* binds : a skin (or an explicit `?fields`) can drop the carrying field — or the whole relation — from the projection, so the declared bind may not reach the final query text.


    private
                    collectOptionalBindNames(array<string|int, mixed> ...$trees) : array<int, string>

The walk is recursive and depth-agnostic : it descends into Field::FIELDS, edges and joins sub-definitions, and picks up a bind reference wherever it sits. Non-array leaves are visited but ignored, so a registry holding a resolved model instance or a callable is harmless. Collecting the raw (un-skinned) definitions is intentional — the superset of all possibly-optional binds is exactly what the execution layer must be allowed to prune.

Parameters

$trees : array<string|int, mixed>: One or more declaration arrays to scan.

Return values

array<int, string> —

The de-duplicated list of bind names.

facetCountCollect()

The shared `COLLECT value = <expr> … SORT count DESC RETURN { value, count }` tail.


    private
                    facetCountCollect(string $expression, string $sort[, string|null $distinctKey = null ]) : array{0: string, 1: string, 2: string}

By default the bucket count is the number of unwound rows (WITH COUNT INTO count). When $distinctKey is provided (opt-in Facet::DISTINCT => true on an unwinding facet), the count becomes the number of DISTINCT root documents in the bucket (AGGREGATE count = COUNT_DISTINCT( <distinctKey> )), so a document whose array repeats the same sub-field value is counted once — matching the ?filter= existence semantics. The aggregate is deliberately named Group::COUNT_NAME (count) so the derived RETURN { value, count } and the SORT count DESC clause stay identical in both modes.

Parameters

$expression : string: The value expression to group on.
$sort : string: The pre-built SORT count DESC clause.
$distinctKey : string|null = null: The root document key expression to count distinctly (e.g. doc._key), or null for the default per-element count.

Return values

array{0: string, 1: string, 2: string} —

[ COLLECT, SORT, RETURN ] fragments.

DocumentsFacetCountsTrait uses trait:short, trait:short

Table of Contents

Constants

Properties

Methods

Constants

FACET_COUNT_ITEM

FACET_COUNT_VALUE

Properties

$collection

$indexes

$arangodb

$type

Methods

analyzerExists()

Parameters

Return values

buildFacetCountsQuery()

Parameters

Tags

Return values

collectionCreate()

Parameters

Return values

collectionDrop()

Parameters

Return values

collectionExists()

Parameters

Return values

collectionRename()

Parameters

Tags

Return values

collectionTruncate()

Parameters

Return values

createIndex()

Parameters

Tags

Return values

debugQuery()

Parameters

explain()

Parameters

Tags

Return values

facetCount()

Parameters

Tags

Return values

facetCounts()

Parameters

Tags

Return values

foundRows()

Return values

getDatabase()

Return values

getDocuments()

Parameters

Tags

Return values

getExtra()

Return values

getFirstResult()

Parameters

Tags

getObject()

Parameters

Tags

Return values

getProfile()

Return values

getResult()

Parameters

Tags

Return values

getStats()

Return values