MCPcopy Index your code
hub / github.com/simstudioai/sim / splitContent

Method splitContent

apps/sim/lib/chunkers/docs-chunker.ts:193–211  ·  view source on GitHub ↗

Returns both chunks and cleaned content so header extraction uses aligned positions.

(
    content: string
  )

Source from the content-addressed store, hash-verified

191
192 /** Returns both chunks and cleaned content so header extraction uses aligned positions. */
193 private async splitContent(
194 content: string
195 ): Promise<{ chunks: string[]; cleanedContent: string }> {
196 const cleanedContent = this.cleanContent(content)
197
198 const tableBoundaries = this.detectTableBoundaries(cleanedContent)
199
200 const chunks = await this.textChunker.chunk(cleanedContent)
201
202 const processedChunks = this.mergeTableChunks(
203 chunks.map((chunk) => chunk.text),
204 tableBoundaries,
205 cleanedContent
206 )
207
208 const finalChunks = this.enforceSizeLimit(processedChunks)
209
210 return { chunks: finalChunks, cleanedContent }
211 }
212
213 private cleanContent(content: string): string {
214 return content

Callers 1

chunkMdxFileMethod · 0.95

Calls 5

cleanContentMethod · 0.95
detectTableBoundariesMethod · 0.95
mergeTableChunksMethod · 0.95
enforceSizeLimitMethod · 0.95
chunkMethod · 0.45

Tested by

no test coverage detected