Documentation
¶
Index ¶
- Constants
- func FilterChunksFilePaths(filePaths []string) []string
- func FilterConvertedFilePaths(filePaths []string) []string
- func FilterRawFilePaths(filePaths []string) []string
- func GetChunksFilePath(rawFilePath string) string
- func GetConvertedFilePath(rawFilePath string) string
- func GetMetadataFilePath(rawFilePath string) string
- func GetRawFilePathFromConvertedFilePath(convertedFilePath string) string
- func GetVectorEmbeddingsFilePath(rawFilePath string) string
- type Chunker
- type ChunkingTool
- type Chunks
- type ChunksDocument
- type ChunksFile
- type ChunksFileMetadata
- type Content
- type ConvertedDocument
- type ConvertedFile
- type ConvertedFileMetadata
- type DataSource
- type Destination
- type DocumentConverter
- type MarkdownSplitter
- type RawFileMetadata
- type RecursiveCharacterSplitter
- type S3BucketSource
- type SnowflakeInternalStage
- type TokenSplitter
Constants ¶
View Source
const ( MetadataFileSuffix = "-metadata.json" ConvertedFileSuffix = "-converted.json" ChunksFileSuffix = "-chunks.json" VectorEmbeddingsFileSuffix = "-vector-embeddings.json" )
Variables ¶
This section is empty.
Functions ¶
func FilterChunksFilePaths ¶
func FilterRawFilePaths ¶
func GetChunksFilePath ¶
func GetConvertedFilePath ¶
func GetMetadataFilePath ¶
Types ¶
type ChunkingTool ¶
type ChunkingTool string
const (
LangchainChunkingTool ChunkingTool = "langchain"
)
type ChunksDocument ¶
type ChunksDocument struct {
Metadata *ChunksFileMetadata `json:"metadata"`
Chunks *Chunks `json:"chunks"`
}
type ChunksFile ¶
type ChunksFile struct {
ConvertedDocument *ConvertedDocument `json:"convertedDocument"`
ChunksDocument *ChunksDocument `json:"chunksDocument"`
}
type ChunksFileMetadata ¶
type ChunksFileMetadata struct {
ConvertedFileMetadata *ConvertedFileMetadata `json:"convertedFileMetadata"`
ChunkingTool ChunkingTool `json:"chunkingTool"`
ChunksGeneratorConfig v1alpha1.ChunksGeneratorConfig `json:"chunksGeneratorConfig"`
}
func (*ChunksFileMetadata) Equal ¶
func (c *ChunksFileMetadata) Equal(other *ChunksFileMetadata) bool
type ConvertedDocument ¶
type ConvertedDocument struct {
Metadata *ConvertedFileMetadata `json:"metadata"`
Content *Content `json:"content"`
}
type ConvertedFile ¶
type ConvertedFile struct {
ConvertedDocument *ConvertedDocument `json:"convertedDocument"`
}
type ConvertedFileMetadata ¶
type ConvertedFileMetadata struct {
RawFilePath string `json:"rawFilePath"`
DocumentConverter DocumentConverter `json:"documentConverter"`
DoclingConfig v1alpha1.DoclingConfig `json:"doclingConfig"`
}
func (*ConvertedFileMetadata) Equal ¶
func (c *ConvertedFileMetadata) Equal(other *ConvertedFileMetadata) bool
type DataSource ¶
type Destination ¶
type DocumentConverter ¶
type DocumentConverter string
const (
DocumentConverterDocling DocumentConverter = "docling"
)
type MarkdownSplitter ¶
type MarkdownSplitter struct {
LangchainClient *langchain.Client
Config *textsplitter.Options
}
type RawFileMetadata ¶
type RecursiveCharacterSplitter ¶
type RecursiveCharacterSplitter struct {
LangchainClient *langchain.Client
Config *textsplitter.Options
}
type S3BucketSource ¶
func (*S3BucketSource) SyncFilesToFilestore ¶
func (s *S3BucketSource) SyncFilesToFilestore(ctx context.Context, fs *filestore.FileStore) ([]RawFileMetadata, error)
type SnowflakeInternalStage ¶
type SnowflakeInternalStage struct {
Client *snowflake.Client
Role string
Database string
Schema string
Stage string
}
func (*SnowflakeInternalStage) SyncFilesToDestination ¶
type TokenSplitter ¶
type TokenSplitter struct {
LangchainClient *langchain.Client
Config *textsplitter.Options
}
Click to show internal directories.
Click to hide internal directories.