Hunt.Index.Schema

Schema for the ContextIndex.

Every context has a type (e.g. text, int, date, position) and additional schema information. This includes how keys are splitted and normalized when inserted and searched for.

Types

type Schema = Map Context ContextSchema

The global schema assigning schema information to each context.

data ContextSchema

The context schema information. Every context schema has a type and additional to adjust the behavior.

The regular expression splits the text into words which are then transformed by the given normalizations functions (e.g. to lower case).

Constructors

ContextSchema

Fields

cxRegEx :: Maybe RegEx

Optional regex to override the default given by context type.

cxNormalizer :: [CNormalizer]

Normalizers to apply on keys.

cxWeight :: Weight

Context weight to boost results.

cxDefault :: Bool

Whether the context is searched in queries without context-specifier.

cxType :: ContextType

The type of the index (e.g. text, int, date, geo-position).

Instances

Show ContextSchema

ToJSON ContextSchema

FromJSON ContextSchema

Binary ContextSchema

Default ContextSchema

data ContextType

A general context type like text or int.

Constructors

CType

Fields

ctName :: Text

Name used in the (JSON) API.

ctRegEx :: RegEx

Default regex to split words.

ctValidate :: CValidator

Validation function for keys.

ctIxImpl :: IndexImpl

The index implementation used for this type.

Instances

Show ContextType

ToJSON ContextType

FromJSON ContextType

Note: This is only partial (de-)serialization. The other components are environment-dependent and cannot be (de-)serialized. We serialize the name and identify the other compontens of the type later.

Binary ContextType

Default ContextType

type ContextTypes = [ContextType]

Set of context types.

data CValidator

Validation function for single words.

Constructors

CValidator

Fields

validate :: Word -> Bool

Instances

Show CValidator

Default CValidator

data CNormalizer

Normalizer for words/keys of an index.

Constructors

CNormalizer

Fields

cnName :: Text

Name used in (JSON) API.

normalize :: Text -> Text

Normalization function.

Instances

Show CNormalizer

ToJSON CNormalizer

FromJSON CNormalizer

Binary CNormalizer

Default CNormalizer

normalize' :: [CNormalizer] -> Word -> Word

Apply the normalizers to a word.

Default Context Types

ctText :: ContextType

Text context type.

ctTextSimple :: ContextType

Special text context type smaller index but not phrase queries possible (due to not storing the words positions)

ctInt :: ContextType

Int context type.

ctDate :: ContextType

Date context type.

ctPosition :: ContextType

Geographic position context type.

ctPositionRTree :: ContextType

Default Normalizers

cnUpperCase :: CNormalizer

Uppercase normalizer "UpperCase".

cnLowerCase :: CNormalizer

Lowercase normalizer "LowerCase".

cnZeroFill :: CNormalizer

Int normalizer "ZeroFill" to preserve int ordering on strings.