Add script for contextual retrieval with document chunking in genaisrc

packages/sample/genaisrc/contextual-retreival.genai.mts

@@ -0,0 +1,34 @@
+script({
+    system: [],
+    files: ["src/azure-lza/azure-azure-resource-manager-bicep.pdf"],
+})
+
+const doc = env.files[0]
+const chunks = splitTextIntoChunks(doc.content, 100)
+
+console.log(
+    `Document: ${doc.filename}, ${doc.content.length} characters, ${chunks.length} chunks`
+)
+for (const chunk of chunks) {
+    console.log(`chunk: ${chunk.slice(0, 25) + "..."}`)
+    const res = await runPrompt(
+        (_) => {
+            _.def("DOCUMENT", doc, { maxTokens: 10000 })
+            $`Here is the chunk we want to situate within the whole document`
+            _.def("CHUNK", chunk)
+            _.$`Please give a short succinct context to situate this chunk 
+within the overall document for the purposes of improving search retrieval of the chunk.
+Answer only with the succinct context and nothing else. `
+        },
+        { cache: "cr" }
+    )
+}
+
+function splitTextIntoChunks(text: string, chunkSize: number): string[] {
+    const tokens = text.split(/\s+/) // Split text into tokens based on whitespace
+    const chunks = []
+    for (let i = 0; i < tokens.length; i += chunkSize) {
+        chunks.push(tokens.slice(i, i + chunkSize).join(" "))
+    }
+    return chunks
+}