@heripo/pdf-parser 0.1.35 → 0.1.37

package/dist/index.js

@@ -590,6 +590,9 @@ var DoclingServer = class _DoclingServer {
     this.venvPath = venvPath;
     this.port = port;
   }
+  logger;
+  venvPath;
+  port;
   async isPortInUse() {
     try {
       const result = await spawnAsync("lsof", ["-ti", `:${this.port}`]);
@@ -736,6 +739,8 @@ var PythonEnvironment = class {
     this.logger = logger;
     this.venvPath = venvPath;
   }
+  logger;
+  venvPath;
   async setup() {
     await this.checkPythonVersion();
     await this.setupPythonEnvironment();
@@ -989,6 +994,7 @@ var PdfTextExtractor = class _PdfTextExtractor {
   constructor(logger) {
     this.logger = logger;
   }
+  logger;
   /**
    * Safely extract page texts: returns the map on success, `undefined` on
    * failure. Failures are logged as a single warning and do not throw, so
@@ -1141,6 +1147,8 @@ var PdfLanguageDetector = class {
     this.pageRenderer = pageRenderer;
     this.textExtractor = textExtractor ?? new PdfTextExtractor(logger);
   }
+  logger;
+  pageRenderer;
   textExtractor;
   async detect(pdfPath, outputDir, options) {
     this.logger.info(
@@ -1357,6 +1365,8 @@ var ImagePdfFallbackError = class extends Error {
     this.originalError = originalError;
     this.fallbackError = fallbackError;
   }
+  originalError;
+  fallbackError;
   name = "ImagePdfFallbackError";
 };
 
@@ -1368,6 +1378,7 @@ var PageRenderer = class {
   constructor(logger) {
     this.logger = logger;
   }
+  logger;
   lastLoggedPercent = 0;
   /**
    * Render all pages of a PDF to individual PNG files.
@@ -1493,6 +1504,7 @@ var InvalidDocumentTypeError = class extends Error {
     );
     this.reason = reason;
   }
+  reason;
   name = "InvalidDocumentTypeError";
   code = "INVALID_DOCUMENT_TYPE";
 };
@@ -2423,6 +2435,10 @@ var ChunkedPDFConverter = class {
     this.config = config;
     this.timeout = timeout;
   }
+  logger;
+  client;
+  config;
+  timeout;
   /**
    * Convert a local PDF in chunks.
    *
@@ -2769,6 +2785,9 @@ var DoclingConversionExecutor = class {
     this.client = client;
     this.timeout = timeout;
   }
+  logger;
+  client;
+  timeout;
   /**
    * Execute a full Docling conversion pipeline.
    *
@@ -2931,6 +2950,7 @@ var ImagePdfConverter = class {
   constructor(logger) {
     this.logger = logger;
   }
+  logger;
   /**
    * Convert a PDF file to an image-based PDF.
    * Downloads the PDF from URL, converts it using ImageMagick, and returns the path.
@@ -3415,6 +3435,7 @@ var PostDoclingPageProcessor = class {
   constructor(logger) {
     this.logger = logger;
   }
+  logger;
   /**
    * Read DoclingDocument from output directory, correct text via VLM,
    * and save the corrected document back.
@@ -5221,6 +5242,8 @@ var ImageRegionSnapper = class {
     this.logger = logger;
     this.options = options;
   }
+  logger;
+  options;
   async snap(pageImagePath, pageSize, bbox) {
     const warnings = [];
     const dims = await this.getImageDimensions(pageImagePath);
@@ -5433,6 +5456,8 @@ var PictureSplitCandidateDetector = class {
     this.options = options;
     this.snapper = snapper ?? new ImageRegionSnapper(logger);
   }
+  logger;
+  options;
   snapper;
   async detect(input) {
     if (!this.isLargeEnoughForInspection(input.pictureBbox)) {
@@ -5734,6 +5759,9 @@ var ReviewAssistanceCheckpointStore = class _ReviewAssistanceCheckpointStore {
     this.checkpoint = checkpoint;
     this.now = now;
   }
+  path;
+  checkpoint;
+  now;
   static open(outputDir, runId, now) {
     const path = join12(outputDir, CHECKPOINT_FILE);
     const checkpoint = this.read(path, runId, now?.() ?? /* @__PURE__ */ new Date());
@@ -6107,6 +6135,7 @@ var ImageCropWriter = class {
     this.logger = logger;
     this.snapper = snapper ?? new ImageRegionSnapper(logger);
   }
+  logger;
   snapper;
   async writeCrop(input) {
     const imagesDir = join13(input.outputDir, "images");
@@ -6229,6 +6258,7 @@ var ReviewAssistancePatcher = class {
     this.snapper = new ImageRegionSnapper(logger);
     this.cropWriter = new ImageCropWriter(logger, this.snapper);
   }
+  logger;
   snapper;
   cropWriter;
   captionResolver = new OrphanCaptionResolver();
@@ -8640,6 +8670,7 @@ var ReviewAssistanceRunner = class {
   constructor(logger) {
     this.logger = logger;
   }
+  logger;
   async analyzeAndSave(outputDir, reportId, modelResolver, options) {
     this.emitProgress(options, {
       substage: "review-assistance:prepare",
@@ -10550,6 +10581,7 @@ var PostDoclingCorrectionPipeline = class {
   constructor(logger) {
     this.logger = logger;
   }
+  logger;
   wrapCallback(pdfPath, reportId, options, originalCallback, abortSignal) {
     const correction = normalizePDFCorrectionOptions(options.correction);
     return async (outputDir) => {
@@ -10656,6 +10688,10 @@ var PDFConverter = class {
     this.enableImagePdfFallback = enableImagePdfFallback;
     this.timeout = timeout;
   }
+  logger;
+  client;
+  enableImagePdfFallback;
+  timeout;
   documentTypeValidated = false;
   /**
    * Validate that the PDF is an archaeological investigation report.