speclore 0.1.4 → 0.1.6

package/dist/index.js

@@ -1411,6 +1411,13 @@ function formatCellValue(value) {
   if (typeof value === "string") return value;
   if (typeof value === "number" || typeof value === "boolean") return String(value);
   if (value instanceof Date) return value.toISOString();
+  if (typeof value === "object" && value !== null && "v" in value) {
+    const cell = value;
+    if (cell.w != null) return cell.w;
+    if (cell.v == null) return "";
+    if (typeof cell.v === "string" || typeof cell.v === "number" || typeof cell.v === "boolean") return String(cell.v);
+    return JSON.stringify(cell.v);
+  }
   return "";
 }
 var init_excel_utils = __esm({
@@ -1420,7 +1427,7 @@ var init_excel_utils = __esm({
 });
 
 // src/plugins/builtin/xlsx-reader.ts
-import ExcelJS from "exceljs";
+import { readFile, utils } from "xlsx";
 var XlsxReader;
 var init_xlsx_reader = __esm({
   "src/plugins/builtin/xlsx-reader.ts"() {
@@ -1432,27 +1439,23 @@ var init_xlsx_reader = __esm({
       canRead(source) {
         return /\.xlsx$/i.test(source) || /\.xls$/i.test(source);
       }
-      async read(source) {
-        const workbook = new ExcelJS.Workbook();
-        await workbook.xlsx.readFile(source);
+      read(source) {
+        const workbook = readFile(source);
         const requirements = [];
-        for (const sheet of workbook.worksheets) {
-          const sheetName = sheet.name;
-          const headers = [];
-          const rows = [];
-          sheet.eachRow((row, rowNumber) => {
-            if (rowNumber === 1) {
-              row.eachCell((cell, colNumber) => {
-                headers[colNumber - 1] = formatCellValue(cell.value);
-              });
-            } else {
-              const obj = {};
-              row.eachCell((cell, colNumber) => {
-                const key = headers[colNumber - 1] ?? `Col${colNumber}`;
-                obj[key] = formatCellValue(cell.value);
-              });
-              rows.push(obj);
-            }
+        for (const sheetName of workbook.SheetNames) {
+          const sheet = workbook.Sheets[sheetName];
+          if (!sheet) continue;
+          const allRows = utils.sheet_to_json(sheet, { header: 1, defval: "" });
+          if (allRows.length === 0) continue;
+          const headers = allRows[0].map((cell) => formatCellValue(cell));
+          const dataRows = allRows.slice(1);
+          const rows = dataRows.map((row) => {
+            const obj = {};
+            row.forEach((cell, colNumber) => {
+              const key = headers[colNumber] ?? `Col${colNumber + 1}`;
+              obj[key] = formatCellValue(cell);
+            });
+            return obj;
           });
           for (let i = 0; i < rows.length; i++) {
             const row = rows[i];
@@ -1471,7 +1474,7 @@ var init_xlsx_reader = __esm({
             }
           }
         }
-        return requirements;
+        return Promise.resolve(requirements);
       }
     };
   }
@@ -2783,44 +2786,51 @@ var init_docx_reader2 = __esm({
 
 // src/core/requirement-reader/xlsx-reader.ts
 import { basename as basename4, extname as extname4 } from "path";
-import ExcelJS2 from "exceljs";
-async function readXlsxFile(filePath) {
-  const workbook = new ExcelJS2.Workbook();
-  await workbook.xlsx.readFile(filePath);
+import { readFile as readFile2, utils as utils2 } from "xlsx";
+function readXlsxFile(filePath) {
+  const workbook = readFile2(filePath);
   const id = basename4(filePath, extname4(filePath)).toLowerCase().replace(/[^a-z0-9\u4e00-\u9fff-]/g, "-").replace(/-+/g, "-");
-  const sheet = workbook.worksheets[0];
+  const firstSheetName = workbook.SheetNames[0];
+  if (!firstSheetName) {
+    return Promise.reject(new Error(`No sheets found in ${filePath}`));
+  }
+  const sheet = workbook.Sheets[firstSheetName];
   if (!sheet) {
-    throw new Error(`No sheets found in ${filePath}`);
-  }
-  const title = sheet.name;
-  const headers = [];
-  const rows = [];
-  sheet.eachRow((row, rowNumber) => {
-    if (rowNumber === 1) {
-      row.eachCell((cell, colNumber) => {
-        headers[colNumber - 1] = formatCellValue(cell.value);
-      });
-    } else {
-      const obj = {};
-      row.eachCell((cell, colNumber) => {
-        const key = headers[colNumber - 1] ?? `Col${colNumber}`;
-        obj[key] = formatCellValue(cell.value);
-      });
-      rows.push(obj);
-    }
+    return Promise.reject(new Error(`Sheet "${firstSheetName}" not found in ${filePath}`));
+  }
+  const title = firstSheetName;
+  const rows = utils2.sheet_to_json(sheet, { header: 1, defval: "" });
+  if (rows.length === 0) {
+    return Promise.resolve({
+      id,
+      title,
+      description: "",
+      rawContent: "",
+      confidence: 0.85
+    });
+  }
+  const headers = rows[0].map((cell) => formatCellValue(cell));
+  const dataRows = rows.slice(1);
+  const objects = dataRows.map((row) => {
+    const obj = {};
+    row.forEach((cell, colNumber) => {
+      const key = headers[colNumber] ?? `Col${colNumber + 1}`;
+      obj[key] = formatCellValue(cell);
+    });
+    return obj;
   });
-  const textRows = rows.map((row) => {
+  const textRows = objects.map((row) => {
     const cells = Object.entries(row).map(([key, value]) => `${key}: ${value}`).join(" | ");
     return cells;
   });
   const content = textRows.join("\n");
-  return {
+  return Promise.resolve({
     id,
     title,
     description: content,
     rawContent: content,
     confidence: 0.85
-  };
+  });
 }
 var init_xlsx_reader2 = __esm({
   "src/core/requirement-reader/xlsx-reader.ts"() {
@@ -2831,10 +2841,10 @@ var init_xlsx_reader2 = __esm({
 
 // src/core/requirement-reader/pdf-reader.ts
 import { basename as basename5, extname as extname5 } from "path";
-import { readFile } from "fs/promises";
+import { readFile as readFile3 } from "fs/promises";
 async function readPdfFile(filePath) {
   const pdfParse = (await import("pdf-parse")).default;
-  const dataBuffer = await readFile(filePath);
+  const dataBuffer = await readFile3(filePath);
   const data = await pdfParse(dataBuffer);
   const id = basename5(filePath, extname5(filePath)).toLowerCase().replace(/[^a-z0-9\u4e00-\u9fff-]/g, "-").replace(/-+/g, "-");
   const lines = data.text.split("\n").filter((l) => l.trim());