SentenceSplitter
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:20
Parse text with a preference for complete sentences.
Extends
Section titled “Extends”SentenceSplitter<
Options
>(nodes
,options?
):TextNode
<Metadata
>[]
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:20
Parse text with a preference for complete sentences.
Type Parameters
Section titled “Type Parameters”Options
Section titled “Options”Options
extends Record
<string
, unknown
>
Parameters
Section titled “Parameters”options?
Section titled “options?”Options
Returns
Section titled “Returns”Constructors
Section titled “Constructors”Constructor
Section titled “Constructor”new SentenceSplitter(
params?
):SentenceSplitter
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:53
Parameters
Section titled “Parameters”params?
Section titled “params?”PartialWithUndefined
<{ chunkSize
: number
; chunkOverlap
: number
; separator
: string
; paragraphSeparator
: string
; secondaryChunkingRegex
: string
; extraAbbreviations
: string
[]; }> & SplitterParams
& object
Returns
Section titled “Returns”SentenceSplitter
Overrides
Section titled “Overrides”MetadataAwareTextSplitter
.constructor
Properties
Section titled “Properties”includeMetadata
Section titled “includeMetadata”includeMetadata:
boolean
=true
Defined in: .build/typescript/packages/core/src/node-parser/base.ts:17
Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.includeMetadata
includePrevNextRel
Section titled “includePrevNextRel”includePrevNextRel:
boolean
=true
Defined in: .build/typescript/packages/core/src/node-parser/base.ts:18
Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.includePrevNextRel
chunkSize
Section titled “chunkSize”chunkSize:
number
=1024
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:24
The token chunk size for each chunk.
chunkOverlap
Section titled “chunkOverlap”chunkOverlap:
number
=200
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:28
The token overlap of each chunk when splitting.
separator
Section titled “separator”separator:
string
=" "
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:32
Default separator for splitting into words
paragraphSeparator
Section titled “paragraphSeparator”paragraphSeparator:
string
="\n\n\n"
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:36
Separator between paragraphs.
secondaryChunkingRegex
Section titled “secondaryChunkingRegex”secondaryChunkingRegex:
string
="[^,.;。?!]+[,.;。?!]?"
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:40
Backup regex for splitting into sentences.
extraAbbreviations
Section titled “extraAbbreviations”extraAbbreviations:
undefined
|string
[] =[]
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:45
Extra abbreviations to consider while splitting into sentences. For example, for contracts, you may want to consider “LLC.” as an important abbreviation
id:
string
Defined in: .build/typescript/packages/core/src/schema/type.ts:22
Inherited from
Section titled “Inherited from”Methods
Section titled “Methods”postProcessParsedNodes()
Section titled “postProcessParsedNodes()”
protected
postProcessParsedNodes(nodes
,parentDocMap
):TextNode
<Metadata
>[]
Defined in: .build/typescript/packages/core/src/node-parser/base.ts:27
Parameters
Section titled “Parameters”parentDocMap
Section titled “parentDocMap”Map
<string
, TextNode
<Metadata
>>
Returns
Section titled “Returns”Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.postProcessParsedNodes
getNodesFromDocuments()
Section titled “getNodesFromDocuments()”Defined in: .build/typescript/packages/core/src/node-parser/base.ts:84
Parameters
Section titled “Parameters”documents
Section titled “documents”Returns
Section titled “Returns”Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.getNodesFromDocuments
splitTexts()
Section titled “splitTexts()”splitTexts(
texts
):string
[]
Defined in: .build/typescript/packages/core/src/node-parser/base.ts:126
Parameters
Section titled “Parameters”string
[]
Returns
Section titled “Returns”string
[]
Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.splitTexts
splitTextsMetadataAware()
Section titled “splitTextsMetadataAware()”splitTextsMetadataAware(
texts
,metadata
):string
[]
Defined in: .build/typescript/packages/core/src/node-parser/base.ts:142
Parameters
Section titled “Parameters”string
[]
metadata
Section titled “metadata”string
[]
Returns
Section titled “Returns”string
[]
Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.splitTextsMetadataAware
getMetadataString()
Section titled “getMetadataString()”
protected
getMetadataString(node
):string
Defined in: .build/typescript/packages/core/src/node-parser/base.ts:151
Parameters
Section titled “Parameters”Returns
Section titled “Returns”string
Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.getMetadataString
parseNodes()
Section titled “parseNodes()”Defined in: .build/typescript/packages/core/src/node-parser/base.ts:161
Parameters
Section titled “Parameters”Returns
Section titled “Returns”Inherited from
Section titled “Inherited from”MetadataAwareTextSplitter
.parseNodes
splitTextMetadataAware()
Section titled “splitTextMetadataAware()”splitTextMetadataAware(
text
,metadata
):string
[]
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:80
Parameters
Section titled “Parameters”string
metadata
Section titled “metadata”string
Returns
Section titled “Returns”string
[]
Overrides
Section titled “Overrides”MetadataAwareTextSplitter
.splitTextMetadataAware
splitText()
Section titled “splitText()”splitText(
text
):string
[]
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:95
Parameters
Section titled “Parameters”string
Returns
Section titled “Returns”string
[]
Overrides
Section titled “Overrides”MetadataAwareTextSplitter
.splitText
_splitText()
Section titled “_splitText()”_splitText(
text
,chunkSize
):string
[]
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:99
Parameters
Section titled “Parameters”string
chunkSize
Section titled “chunkSize”number
Returns
Section titled “Returns”string
[]
tokenSize()
Section titled “tokenSize()”tokenSize(
text
):number
Defined in: .build/typescript/packages/core/src/node-parser/sentence-splitter.ts:228
Parameters
Section titled “Parameters”string
Returns
Section titled “Returns”number