@doclo/core 0.2.1 → 0.2.3

package/dist/index.d.ts

@@ -1,5 +1,5 @@
-import { P as ProviderVendor, A as AccessMethod } from './validation-B8GRTtww.js';
-export { G as AggregatedMetrics, B as BBox, r as CategorizeNodeConfig, t as ChunkMetadata, v as ChunkNodeConfig, u as ChunkOutput, n as CitationConfig, k as CitationSourceType, w as CombineNodeConfig, U as CompatibilityRule, C as ConsensusConfig, e as ConsensusMetadata, d as ConsensusRunResult, D as DocumentIR, b as DocumentIRExtras, y as EnhancedExtractionSchema, s as ExtractInputMode, E as ExtractNodeConfig, a0 as ExtractedImage, m as FieldCitation, F as FieldVotingDetails, H as FlowContext, a8 as FlowExecutionError, h as FlowInput, i as FlowInputValidation, j as FlowResult, a4 as FlowStepLocation, a9 as FlowValidationError, I as IRLine, a as IRPage, X as JSONSchemaNode, c as LLMJsonProvider, L as LLMProvider, _ as LanguageOptions, l as LineCitation, g as MaybeWithConsensusMetadata, M as MultimodalInput, aa as NODE_COMPATIBILITY_MATRIX, J as NodeCtx, Q as NodeDef, K as NodeTypeInfo, T as NodeTypeName, N as NormalizedBBox, O as OCRProvider, a1 as OCRProviderOptions, x as OutputNodeConfig, o as OutputWithCitations, f as OutputWithConsensus, Z as PageRangeOptions, p as ParseNodeConfig, Y as ProcessingMode, a3 as ProviderCitation, am as ProviderIdentity, aj as RESERVED_VARIABLES, R as ReasoningConfig, $ as SegmentationResult, S as SplitDocument, q as SplitNodeConfig, z as StepMetric, V as VLMProvider, a2 as VLMProviderOptions, W as ValidationResult, a5 as aggregateMetrics, ah as canStartForEachItemFlow, aq as createIdentity, al as extractErrorMessage, ad as getCompatibleTargets, ac as getNodeTypeInfo, ab as getNodeTypeName, ae as getSuggestedConnections, ag as getValidForEachStarters, ap as isLocalEndpoint, a6 as node, ao as parseProviderString, ak as protectReservedVariables, a7 as runPipeline, an as toProviderString, ai as validateJson, af as validateNodeConnection } from './validation-B8GRTtww.js';
+import { P as ProviderVendor, A as AccessMethod } from './validation-C_RN-Xqr.js';
+export { G as AggregatedMetrics, B as BBox, r as CategorizeNodeConfig, t as ChunkMetadata, v as ChunkNodeConfig, u as ChunkOutput, n as CitationConfig, k as CitationSourceType, w as CombineNodeConfig, W as CompatibilityRule, C as ConsensusConfig, e as ConsensusMetadata, d as ConsensusRunResult, D as DocumentIR, b as DocumentIRExtras, y as EnhancedExtractionSchema, s as ExtractInputMode, E as ExtractNodeConfig, a1 as ExtractedImage, m as FieldCitation, F as FieldVotingDetails, H as FlowContext, a9 as FlowExecutionError, h as FlowInput, i as FlowInputValidation, j as FlowResult, a5 as FlowStepLocation, aa as FlowValidationError, I as IRLine, a as IRPage, Y as JSONSchemaNode, c as LLMJsonProvider, L as LLMProvider, $ as LanguageOptions, l as LineCitation, g as MaybeWithConsensusMetadata, M as MultimodalInput, ab as NODE_COMPATIBILITY_MATRIX, J as NodeCtx, Q as NodeDef, K as NodeTypeInfo, U as NodeTypeName, N as NormalizedBBox, O as OCRProvider, a2 as OCRProviderOptions, x as OutputNodeConfig, o as OutputWithCitations, f as OutputWithConsensus, _ as PageRangeOptions, p as ParseNodeConfig, Z as ProcessingMode, a4 as ProviderCitation, an as ProviderIdentity, ak as RESERVED_VARIABLES, R as ReasoningConfig, a0 as SegmentationResult, S as SplitDocument, q as SplitNodeConfig, z as StepMetric, T as TextResponse, V as VLMProvider, a3 as VLMProviderOptions, X as ValidationResult, a6 as aggregateMetrics, ai as canStartForEachItemFlow, ar as createIdentity, am as extractErrorMessage, ae as getCompatibleTargets, ad as getNodeTypeInfo, ac as getNodeTypeName, af as getSuggestedConnections, ah as getValidForEachStarters, aq as isLocalEndpoint, a7 as node, ap as parseProviderString, al as protectReservedVariables, a8 as runPipeline, ao as toProviderString, aj as validateJson, ag as validateNodeConnection } from './validation-C_RN-Xqr.js';
 export { getDocumentPageCount, getPDFPageCount, getPageCountMetadata, getTotalPageCount, splitPDFIntoChunks } from './pdf-utils.js';
 
 /**

package/dist/index.js

@@ -1204,22 +1204,21 @@ function validateFlowInputFormat(input, acceptedFormats) {
 
 // src/pdf-utils.ts
 import { PDFDocument } from "pdf-lib";
-async function getPDFPageCount(dataUrl) {
-  const base64Match = dataUrl.match(/^data:application\/pdf;base64,(.+)$/);
-  if (!base64Match) {
-    throw new Error("Invalid PDF data URL format. Expected: data:application/pdf;base64,{base64data}");
+function extractPDFBase64(input) {
+  const dataUrlMatch = input.match(/^data:application\/pdf;base64,(.+)$/);
+  if (dataUrlMatch) {
+    return dataUrlMatch[1];
   }
-  const base64Data = base64Match[1];
+  return input;
+}
+async function getPDFPageCount(input) {
+  const base64Data = extractPDFBase64(input);
   const pdfBytes = base64ToArrayBuffer(base64Data);
   const pdfDoc = await PDFDocument.load(pdfBytes);
   return pdfDoc.getPageCount();
 }
-async function splitPDFIntoChunks(dataUrl, pageRanges) {
-  const base64Match = dataUrl.match(/^data:application\/pdf;base64,(.+)$/);
-  if (!base64Match) {
-    throw new Error("Invalid PDF data URL format. Expected: data:application/pdf;base64,{base64data}");
-  }
-  const base64Data = base64Match[1];
+async function splitPDFIntoChunks(input, pageRanges) {
+  const base64Data = extractPDFBase64(input);
   const pdfBytes = base64ToArrayBuffer(base64Data);
   const pdfDoc = await PDFDocument.load(pdfBytes);
   const totalPages = pdfDoc.getPageCount();