markitdown-ts 0.0.6 → 0.0.8

package/README.md

@@ -50,7 +50,7 @@ try {
   const result = await markitdown.convert("https://arxiv.org/pdf/2308.08155v2.pdf");
 
   if (result) {
-    console.log(result.text_content);
+    console.log(result.markdown);
   }
 } catch (error) {
   console.error("Conversion failed:", error);
@@ -158,6 +158,8 @@ class MarkItDown {
 export type ConverterResult =
   | {
       title: string | null;
+      markdown: string;
+      /** @deprecated Use `markdown` instead. */
       text_content: string;
     }
   | null

package/dist/index.cjs

@@ -8,7 +8,7 @@ const TurndownService = require('turndown');
 const turndownPluginGfm = require('@joplin/turndown-plugin-gfm');
 const xmldom = require('@xmldom/xmldom');
 const url = require('url');
-const pdfTs = require('pdf-ts');
+const pdfParse = require('pdf-parse');
 const Mammoth = require('mammoth');
 const XLSX = require('xlsx');
 const childProcess = require('child_process');
@@ -63,6 +63,7 @@ class PlainTextConverter {
     }
     return {
       title: null,
+      markdown: content,
       text_content: content
     };
   }
@@ -176,6 +177,7 @@ class HtmlConverter {
     }
     return {
       title: doc.title,
+      markdown: webpageText,
       text_content: webpageText
     };
   }
@@ -239,10 +241,7 @@ class RSSConverter {
           mdText += this._parseContent(entryContent);
         }
       }
-      return {
-        title,
-        text_content: mdText
-      };
+      return { title, markdown: mdText, text_content: mdText };
     } catch (error) {
       console.error("Atom Parsing Error:", error);
       return null;
@@ -290,10 +289,7 @@ class RSSConverter {
           mdText += this._parseContent(content);
         }
       }
-      return {
-        title: channelTitle,
-        text_content: mdText
-      };
+      return { title: channelTitle, markdown: mdText, text_content: mdText };
     } catch (error) {
       console.error("RSS Parsing Error:", error);
       return null;
@@ -363,10 +359,7 @@ class WikipediaConverter {
     } else {
       webpageText = new CustomTurnDown().convert_soup(doc);
     }
-    return {
-      title: mainTitle,
-      text_content: webpageText
-    };
+    return { title: mainTitle, markdown: webpageText, text_content: webpageText };
   }
 }
 
@@ -492,10 +485,7 @@ ${transcriptText}
       }
     }
     const finalTitle = title ? title : doc.title;
-    return {
-      title: finalTitle,
-      text_content: webpageText
-    };
+    return { title: finalTitle, markdown: webpageText, text_content: webpageText };
   }
   _get(metadata, keys, default_value) {
     for (const k of keys) {
@@ -573,10 +563,7 @@ ${sourceLines.join("")}
       }
       const mdText = mdOutput.join("\n\n");
       title = notebookContent.metadata?.title || title;
-      return {
-        title,
-        text_content: mdText
-      };
+      return { title, markdown: mdText, text_content: mdText };
     } catch (e) {
       console.error("Error converting .ipynb file:", e);
       throw new Error(`Error converting .ipynb file: ${e}`);
@@ -637,10 +624,7 @@ class BingSerpConverter {
     const webpageText = `## A Bing search for '${query}' found the following results:
 
 ${results.join("\n\n")}`;
-    return {
-      title: doc.title,
-      text_content: webpageText
-    };
+    return { title: doc.title, markdown: webpageText, text_content: webpageText };
   }
   _decodeBase64Url(encodedUrl) {
     let u = encodedUrl.slice(2).trim() + "==";
@@ -670,11 +654,10 @@ class PdfConverter {
   }
   async _convert(pdfContent) {
     try {
-      const textContent = await pdfTs.pdfToText(pdfContent);
-      return {
-        title: null,
-        text_content: textContent
-      };
+      const parser = new pdfParse.PDFParse({ data: pdfContent });
+      const result = await parser.getText();
+      await parser.destroy();
+      return { title: null, markdown: result.text, text_content: result.text };
     } catch (error) {
       console.error("PDF Parsing Error:", error);
       return null;
@@ -730,12 +713,9 @@ class XlsxConverter extends HtmlConverter {
         mdContent += `## ${sheetName}
 `;
         let htmlContent = XLSX__namespace.utils.sheet_to_html(workbook.Sheets[sheetName]);
-        mdContent += (await this._convert(htmlContent))?.text_content.trim() + "\n\n";
+        mdContent += (await this._convert(htmlContent))?.markdown.trim() + "\n\n";
       }
-      return {
-        title: workbook?.Props?.Title || "Untitled",
-        text_content: mdContent
-      };
+      return { title: workbook?.Props?.Title || "Untitled", markdown: mdContent, text_content: mdContent };
     } catch (e) {
       console.error(e);
       return null;
@@ -825,10 +805,7 @@ ${transcript === "" ? "[No speech detected]" : transcript}`;
     } else {
       mdContent += "\n\n### Audio Transcript:\n[Audio transcription is not supported for Buffer inputs in this version.]";
     }
-    return {
-      title: null,
-      text_content: mdContent.trim()
-    };
+    return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
   }
   // TODO: Add speech to text
   async _transcribeAudio(_) {
@@ -895,10 +872,7 @@ ${transcript == "" ? "[No speech detected]" : transcript}`;
     } else {
       mdContent += "\n\n### Audio Transcript:\n[Audio conversion and transcription are not supported for Buffer inputs.]";
     }
-    return {
-      title: null,
-      text_content: mdContent.trim()
-    };
+    return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
   }
 }
 
@@ -950,10 +924,7 @@ class ImageConverter extends MediaConverter {
 ${(await this._getLLMDescription(imageBuffer, options)).trim()}
 `;
     }
-    return {
-      title: null,
-      text_content: mdContent.trim()
-    };
+    return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
   }
   async _getLLMDescription(imageBuffer, options) {
     if (!options.llmPrompt || options.llmPrompt.trim() === "") {
@@ -989,6 +960,7 @@ class ZipConverter {
     if (!parentConverters) {
       return {
         title: null,
+        markdown: `[ERROR] No converters available to process zip contents from: ${source}`,
         text_content: `[ERROR] No converters available to process zip contents from: ${source}`
       };
     }
@@ -1026,7 +998,7 @@ class ZipConverter {
               mdResults.push(`
 ## File: ${relativePath}
 
-${result.text_content}
+${result.markdown}
 
 `);
               break;
@@ -1050,19 +1022,18 @@ ${result.text_content}
         inputStream.pipe(parser);
       });
       mdContent += mdResults.join("");
-      return {
-        title: null,
-        text_content: mdContent.trim()
-      };
+      return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
     } catch (error) {
       if (error.message.includes("invalid signature")) {
         return {
           title: null,
+          markdown: `[ERROR] Invalid or corrupted zip file: ${source}`,
           text_content: `[ERROR] Invalid or corrupted zip file: ${source}`
         };
       }
       return {
         title: null,
+        markdown: `[ERROR] Failed to process zip file ${source}: ${String(error)}`,
         text_content: `[ERROR] Failed to process zip file ${source}: ${String(error)}`
       };
     }
@@ -1180,8 +1151,8 @@ class MarkItDown {
           error = e;
         }
         if (res != null) {
-          res.text_content = res.text_content.replace(/(?:\r\n|\r|\n)/g, "\n").trim();
-          res.text_content = res.text_content.replace(/\n{3,}/g, "\n\n");
+          res.markdown = res.markdown.replace(/(?:\r\n|\r|\n)/g, "\n").trim();
+          res.markdown = res.markdown.replace(/\n{3,}/g, "\n\n");
           return res;
         }
       }

package/dist/index.d.cts

@@ -3,6 +3,8 @@ import mammoth from 'mammoth';
 
 type ConverterResult = {
     title: string | null;
+    markdown: string;
+    /** @deprecated Use `markdown` instead. */
     text_content: string;
 } | null | undefined;
 type ConverterOptions = {

package/dist/index.d.mts

@@ -3,6 +3,8 @@ import mammoth from 'mammoth';
 
 type ConverterResult = {
     title: string | null;
+    markdown: string;
+    /** @deprecated Use `markdown` instead. */
     text_content: string;
 } | null | undefined;
 type ConverterOptions = {

package/dist/index.d.ts

@@ -3,6 +3,8 @@ import mammoth from 'mammoth';
 
 type ConverterResult = {
     title: string | null;
+    markdown: string;
+    /** @deprecated Use `markdown` instead. */
     text_content: string;
 } | null | undefined;
 type ConverterOptions = {

package/dist/index.mjs

@@ -8,7 +8,7 @@ import TurndownService from 'turndown';
 import turndownPluginGfm from '@joplin/turndown-plugin-gfm';
 import { DOMParser } from '@xmldom/xmldom';
 import { URL as URL$1 } from 'url';
-import { pdfToText } from 'pdf-ts';
+import { PDFParse } from 'pdf-parse';
 import Mammoth from 'mammoth';
 import * as XLSX from 'xlsx';
 import * as childProcess from 'child_process';
@@ -35,6 +35,7 @@ class PlainTextConverter {
     }
     return {
       title: null,
+      markdown: content,
       text_content: content
     };
   }
@@ -148,6 +149,7 @@ class HtmlConverter {
     }
     return {
       title: doc.title,
+      markdown: webpageText,
       text_content: webpageText
     };
   }
@@ -211,10 +213,7 @@ class RSSConverter {
           mdText += this._parseContent(entryContent);
         }
       }
-      return {
-        title,
-        text_content: mdText
-      };
+      return { title, markdown: mdText, text_content: mdText };
     } catch (error) {
       console.error("Atom Parsing Error:", error);
       return null;
@@ -262,10 +261,7 @@ class RSSConverter {
           mdText += this._parseContent(content);
         }
       }
-      return {
-        title: channelTitle,
-        text_content: mdText
-      };
+      return { title: channelTitle, markdown: mdText, text_content: mdText };
     } catch (error) {
       console.error("RSS Parsing Error:", error);
       return null;
@@ -335,10 +331,7 @@ class WikipediaConverter {
     } else {
       webpageText = new CustomTurnDown().convert_soup(doc);
     }
-    return {
-      title: mainTitle,
-      text_content: webpageText
-    };
+    return { title: mainTitle, markdown: webpageText, text_content: webpageText };
   }
 }
 
@@ -464,10 +457,7 @@ ${transcriptText}
       }
     }
     const finalTitle = title ? title : doc.title;
-    return {
-      title: finalTitle,
-      text_content: webpageText
-    };
+    return { title: finalTitle, markdown: webpageText, text_content: webpageText };
   }
   _get(metadata, keys, default_value) {
     for (const k of keys) {
@@ -545,10 +535,7 @@ ${sourceLines.join("")}
       }
       const mdText = mdOutput.join("\n\n");
       title = notebookContent.metadata?.title || title;
-      return {
-        title,
-        text_content: mdText
-      };
+      return { title, markdown: mdText, text_content: mdText };
     } catch (e) {
       console.error("Error converting .ipynb file:", e);
       throw new Error(`Error converting .ipynb file: ${e}`);
@@ -609,10 +596,7 @@ class BingSerpConverter {
     const webpageText = `## A Bing search for '${query}' found the following results:
 
 ${results.join("\n\n")}`;
-    return {
-      title: doc.title,
-      text_content: webpageText
-    };
+    return { title: doc.title, markdown: webpageText, text_content: webpageText };
   }
   _decodeBase64Url(encodedUrl) {
     let u = encodedUrl.slice(2).trim() + "==";
@@ -642,11 +626,10 @@ class PdfConverter {
   }
   async _convert(pdfContent) {
     try {
-      const textContent = await pdfToText(pdfContent);
-      return {
-        title: null,
-        text_content: textContent
-      };
+      const parser = new PDFParse({ data: pdfContent });
+      const result = await parser.getText();
+      await parser.destroy();
+      return { title: null, markdown: result.text, text_content: result.text };
     } catch (error) {
       console.error("PDF Parsing Error:", error);
       return null;
@@ -702,12 +685,9 @@ class XlsxConverter extends HtmlConverter {
         mdContent += `## ${sheetName}
 `;
         let htmlContent = XLSX.utils.sheet_to_html(workbook.Sheets[sheetName]);
-        mdContent += (await this._convert(htmlContent))?.text_content.trim() + "\n\n";
+        mdContent += (await this._convert(htmlContent))?.markdown.trim() + "\n\n";
       }
-      return {
-        title: workbook?.Props?.Title || "Untitled",
-        text_content: mdContent
-      };
+      return { title: workbook?.Props?.Title || "Untitled", markdown: mdContent, text_content: mdContent };
     } catch (e) {
       console.error(e);
       return null;
@@ -797,10 +777,7 @@ ${transcript === "" ? "[No speech detected]" : transcript}`;
     } else {
       mdContent += "\n\n### Audio Transcript:\n[Audio transcription is not supported for Buffer inputs in this version.]";
     }
-    return {
-      title: null,
-      text_content: mdContent.trim()
-    };
+    return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
   }
   // TODO: Add speech to text
   async _transcribeAudio(_) {
@@ -867,10 +844,7 @@ ${transcript == "" ? "[No speech detected]" : transcript}`;
     } else {
       mdContent += "\n\n### Audio Transcript:\n[Audio conversion and transcription are not supported for Buffer inputs.]";
     }
-    return {
-      title: null,
-      text_content: mdContent.trim()
-    };
+    return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
   }
 }
 
@@ -922,10 +896,7 @@ class ImageConverter extends MediaConverter {
 ${(await this._getLLMDescription(imageBuffer, options)).trim()}
 `;
     }
-    return {
-      title: null,
-      text_content: mdContent.trim()
-    };
+    return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
   }
   async _getLLMDescription(imageBuffer, options) {
     if (!options.llmPrompt || options.llmPrompt.trim() === "") {
@@ -961,6 +932,7 @@ class ZipConverter {
     if (!parentConverters) {
       return {
         title: null,
+        markdown: `[ERROR] No converters available to process zip contents from: ${source}`,
         text_content: `[ERROR] No converters available to process zip contents from: ${source}`
       };
     }
@@ -998,7 +970,7 @@ class ZipConverter {
               mdResults.push(`
 ## File: ${relativePath}
 
-${result.text_content}
+${result.markdown}
 
 `);
               break;
@@ -1022,19 +994,18 @@ ${result.text_content}
         inputStream.pipe(parser);
       });
       mdContent += mdResults.join("");
-      return {
-        title: null,
-        text_content: mdContent.trim()
-      };
+      return { title: null, markdown: mdContent.trim(), text_content: mdContent.trim() };
     } catch (error) {
       if (error.message.includes("invalid signature")) {
         return {
           title: null,
+          markdown: `[ERROR] Invalid or corrupted zip file: ${source}`,
           text_content: `[ERROR] Invalid or corrupted zip file: ${source}`
         };
       }
       return {
         title: null,
+        markdown: `[ERROR] Failed to process zip file ${source}: ${String(error)}`,
         text_content: `[ERROR] Failed to process zip file ${source}: ${String(error)}`
       };
     }
@@ -1152,8 +1123,8 @@ class MarkItDown {
           error = e;
         }
         if (res != null) {
-          res.text_content = res.text_content.replace(/(?:\r\n|\r|\n)/g, "\n").trim();
-          res.text_content = res.text_content.replace(/\n{3,}/g, "\n\n");
+          res.markdown = res.markdown.replace(/(?:\r\n|\r|\n)/g, "\n").trim();
+          res.markdown = res.markdown.replace(/\n{3,}/g, "\n\n");
           return res;
         }
       }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "markitdown-ts",
-  "version": "0.0.6",
+  "version": "0.0.8",
   "description": "",
   "keywords": [],
   "homepage": "https://github.com/dead8309/markitdown-ts#readme",
@@ -40,7 +40,8 @@
     "typescript": "^5.7.2",
     "unbuild": "^3.0.1",
     "vite": "^6.0.4",
-    "vitest": "^2.1.8"
+    "vitest": "^2.1.8",
+    "zod": "^4.1.8"
   },
   "dependencies": {
     "@joplin/turndown-plugin-gfm": "^1.0.60",
@@ -49,7 +50,7 @@
     "jsdom": "^25.0.1",
     "mammoth": "^1.8.0",
     "mime-types": "^2.1.35",
-    "pdf-ts": "^0.0.2",
+    "pdf-parse": "^2.4.5",
     "turndown": "^7.2.0",
     "xlsx": "^0.18.5"
   },