AnalyzerField uses ConstantsTrait

JSON field names exchanged with the ArangoDB analyzer API (`/_api/analyzer`), on both the request side (body of `POST /_api/analyzer`) and the response side (wrapper of `GET /_api/analyzer/{name}` and entries of `GET /_api/analyzer?force=true`).

Tags

see

https://docs.arangodb.com/stable/develop/http-api/analyzers/

author

Marc Alcaraz (ekameleon)

since

1.0.0

Table of Contents

Constants

ACCENT  : string = 'accent'

Whether the analyzer should keep diacritics on the input (`text` / `norm` only).

CASE  : string = 'case'

Case folding strategy applied to the input (`text` / `norm` only). Recognised values: `"lower"`, `"upper"`, `"none"`.

EDGE_NGRAM  : string = 'edgeNgram'

Edge n-gram options nested inside the `properties` of a `text` analyzer — carries `min`, `max`, `preserveOriginal` sub-fields.

END_MARKER  : string = 'endMarker'

String appended to the end of the input before n-gram emission (`ngram` only), so end-of-token n-grams can be distinguished.

FEATURES  : string = 'features'

List of analyzer feature toggles — entries of {@see AnalyzerFeature}.

LOCALE  : string = 'locale'

BCP 47 / ICU locale tag (e.g. `"en"`, `"fr.utf-8"`) driving the language-aware behaviour of the analyzer (`text` / `norm` / `stem`).

MAX  : string = 'max'

Upper bound of the n-gram window (inclusive). Lives under the {@see self::EDGE_NGRAM} wrapper for a `text` analyzer, or at the top level of the `properties` for an `ngram` analyzer.

MIN  : string = 'min'

Lower bound of the n-gram window (inclusive). Lives under the {@see self::EDGE_NGRAM} wrapper for a `text` analyzer, or at the top level of the `properties` for an `ngram` analyzer.

NAME  : string = 'name'

Top-level analyzer name. Must be prefixed with the database name when shared across databases (`mydb::myanalyzer`).

PIPELINE  : string = 'pipeline'

Ordered list of sub-analyzers run as a chain (`pipeline` only).

PRESERVE_ORIGINAL  : string = 'preserveOriginal'

Whether the n-gram emitter should also keep the original (un-trimmed) token in the output stream. Lives under the {@see self::EDGE_NGRAM} wrapper for a `text` analyzer, or at the top level of the `properties` for an `ngram` analyzer.

PROPERTIES  : string = 'properties'

Wrapper field carrying the type-specific options of an analyzer. Always an object — empty (`{}`) for the {@see AnalyzerType::IDENTITY} analyzer.

RESULT  : string = 'result'

Wrapper field carrying the list of analyzers in the response of `GET /_api/analyzer`.

START_MARKER  : string = 'startMarker'

String prepended to the start of the input before n-gram emission (`ngram` only), so start-of-token n-grams can be distinguished. Lives at the top level of the `ngram` `properties`.

STEMMING  : string = 'stemming'

Whether the `text` analyzer should apply Snowball-style stemming on the tokens it emits.

STOPWORDS  : string = 'stopwords'

List of stopwords to drop from the token stream (`text` only).

STOPWORDS_PATH  : string = 'stopwordsPath'

Filesystem path to a newline-separated stopwords file (`text` only). The path is resolved server-side.

STREAM_TYPE  : string = 'streamType'

Input encoding the `ngram` analyzer operates on — `"binary"` (byte-wise, the server default) or `"utf8"` (codepoint-wise).

TYPE  : string = 'type'

Analyzer type discriminator — entries of {@see AnalyzerType}.