kmemo-core/dev.kmemo.guard

Package-level declarations

Types

AnswerAnchorGuard

class AnswerAnchorGuard(stopwords: Set<String> = Vocabulary.STOPWORDS, maxSubstitutions: Int = 2) : ResponseAwareGuard

Rejects a candidate whose stored answer repeats the very word that the two prompts disagree on.

AntonymGuard

class AntonymGuard(antonyms: Set<Pair<String, String>> = Vocabulary.ANTONYMS) : MatchGuard

Rejects matches where one prompt asks for the opposite of the other.

DirectionGuard

class DirectionGuard(cues: Set<String> = Vocabulary.DIRECTIONAL_CUES, stopwords: Set<String> = Vocabulary.STOPWORDS) : MatchGuard

Rejects matches where the same words appear in an order that reverses the question.

EntityGuard

class EntityGuard(stopwords: Set<String> = Vocabulary.STOPWORDS, sentenceOpeners: Set<String> = Vocabulary.SENTENCE_OPENERS, nonEntityCapitals: Set<String> = Vocabulary.NON_ENTITY_CAPITALS) : MatchGuard

Rejects matches that talk about different named things.

GuardVerdict

sealed interface GuardVerdict

A guard's decision about one candidate match.

GuardVocabulary

data class GuardVocabulary(val stopwords: Set<String>, val sentenceOpeners: Set<String>, val nonEntityCapitals: Set<String>, val negationMarkers: Set<String>, val antonyms: Set<Pair<String, String>>, val temporalMarkers: Set<String>, val scopeMarkers: Set<String>, val directionalCues: Set<String>, val units: Map<String, MeasurementUnit>, val qualifierOpeners: Set<String> = emptySet())

The complete set of word lists the standard guards read from, bundled so a whole language can be swapped in one object.

LengthRatioGuard

class LengthRatioGuard(maxRatio: Double = DEFAULT_MAX_RATIO) : MatchGuard

Rejects matches where one prompt is far longer than the other.

LexicalDivergenceGuard

class LexicalDivergenceGuard(minOverlap: Double = DEFAULT_MIN_OVERLAP, minTokens: Int = DEFAULT_MIN_TOKENS, stopwords: Set<String> = Vocabulary.STOPWORDS) : MatchGuard

Rejects matches whose prompts share too few meaningful words. The backstop under the specialised guards: it catches the swaps nobody wrote a rule for.

MatchGuard

interface MatchGuard

A veto on a candidate match that already cleared the similarity threshold.

MatchGuards

object MatchGuards

Ready-made guard sets for dev.kmemo.SemanticCache.

MeasurementUnit

data class MeasurementUnit(val canonical: String, val dimension: String)

A unit of measure: its canonical name, and the kind of quantity it measures.

NegationGuard

class NegationGuard(markers: Set<String> = Vocabulary.NEGATION_MARKERS, stopwords: Set<String> = Vocabulary.STOPWORDS) : MatchGuard

Rejects matches where one prompt is negated and the other is not.

NumericGuard

class NumericGuard : MatchGuard

Rejects matches whose prompts do not contain the same numbers.

ResponseAwareGuard

interface ResponseAwareGuard : MatchGuard

A MatchGuard that also reads the candidate's stored response, not just the two prompts.

ScopeGuard

class ScopeGuard(markers: Set<String> = Vocabulary.SCOPE_MARKERS) : MatchGuard

Rejects matches that ask for the same subject in a different shape.

SubSpanGuard

class SubSpanGuard(stopwords: Set<String> = Vocabulary.STOPWORDS, qualifierOpeners: Set<String> = Vocabulary.QUALIFIER_OPENERS) : MatchGuard

Rejects a match where one prompt is the other plus a clause that narrows the question.

SubstitutionGuard

class SubstitutionGuard : MatchGuard

Rejects matches whose prompts are identical except for one word.

TemporalGuard

class TemporalGuard(markers: Set<String> = Vocabulary.TEMPORAL_MARKERS, stopwords: Set<String> = Vocabulary.STOPWORDS) : MatchGuard

Rejects matches that pin the question to different moments in time.

UnitGuard

class UnitGuard(units: Map<String, MeasurementUnit> = Vocabulary.UNITS) : MatchGuard

Rejects matches whose prompts mention different units or currencies.

Vocabularies

object Vocabularies

Curated GuardVocabulary packs for the highest-traffic languages, and a forLocale lookup.

Vocabulary

object Vocabulary

WordOrderGuard

class WordOrderGuard(stopwords: Set<String> = Vocabulary.STOPWORDS, sentenceOpeners: Set<String> = Vocabulary.SENTENCE_OPENERS, nonEntityCapitals: Set<String> = Vocabulary.NON_ENTITY_CAPITALS, minAnchors: Int = DEFAULT_MIN_ANCHORS) : MatchGuard

Rejects matches whose prompts name the same things in a different order.

Functions

forLocale

fun Vocabularies.forLocale(locale: Locale): GuardVocabulary

The pack for locale's language.

standard

fun MatchGuards.standard(locale: Locale): List<MatchGuard>

java.util.Locale conveniences for the guard vocabularies.