Class: SimpleNodeParser
SimpleNodeParser is the default NodeParser. It splits documents into TextNodes using a splitter, by default SentenceSplitter
Implements
Constructors
constructor
• new SimpleNodeParser(init?): SimpleNodeParser
Parameters
| Name | Type |
|---|---|
init? | Object |
init.chunkOverlap? | number |
init.chunkSize? | number |
init.includeMetadata? | boolean |
init.includePrevNextRel? | boolean |
init.textSplitter? | SentenceSplitter |
Returns
Defined in
packages/core/src/nodeParsers/SimpleNodeParser.ts:24
Properties
includeMetadata
• includeMetadata: boolean
Whether to include metadata in the nodes.
Defined in
packages/core/src/nodeParsers/SimpleNodeParser.ts:18
includePrevNextRel
• includePrevNextRel: boolean
Whether to include previous and next relationships in the nodes.
Defined in
packages/core/src/nodeParsers/SimpleNodeParser.ts:22
textSplitter
• textSplitter: SentenceSplitter
The text splitter to use.
Defined in
packages/core/src/nodeParsers/SimpleNodeParser.ts:14
Methods
getNodesFromDocuments
▸ getNodesFromDocuments(documents): TextNode<Metadata>[]
Generate Node objects from documents
Parameters
| Name | Type |
|---|---|
documents | BaseNode<Metadata>[] |
Returns
Implementation of
NodeParser.getNodesFromDocuments
Defined in
packages/core/src/nodeParsers/SimpleNodeParser.ts:58
transform
▸ transform(nodes, _options?): Promise<BaseNode<Metadata>[]>
Parameters
| Name | Type |
|---|---|
nodes | BaseNode<Metadata>[] |
_options? | any |
Returns
Implementation of
Defined in
packages/core/src/nodeParsers/SimpleNodeParser.ts:41
fromDefaults
▸ fromDefaults(init?): SimpleNodeParser
Parameters
| Name | Type |
|---|---|
init? | Object |
init.chunkOverlap? | number |
init.chunkSize? | number |
init.includeMetadata? | boolean |
init.includePrevNextRel? | boolean |