lite-email-parser 1.0.1 → 1.0.2

package/README.md

@@ -0,0 +1,55 @@
+# Lite email parser
+
+Simple library to remove signature and replies and extract attachments and inline attachments from an email.
+
+## Import
+
+`const { parseEmail } = require('lite-email-parser');`
+
+`import { parseEmail } from 'lite-email-parser';`
+
+
+## Usage
+
+### No file upload
+
+```
+// Example: Simple usage without uploading attachments
+const { parseEmail } = require('lite-email-parser');
+const fs = require('fs');
+
+// Load your email (string or Buffer accepted here)
+const email = fs.readFileSync('email.eml');
+
+const result = await parseEmail(email);
+
+// lastMessage: Pure HTML content with replies/signatures stripped
+console.log(result.lastMessage);
+
+// attachments: Array of IFile objects
+console.log(result.attachments);
+
+// inlineAttachments: Array of IFile objects embedded in the email
+console.log(result.inlineAttachments);
+```
+
+### File upload
+
+```
+const { parseEmail } = require('lite-email-parser');
+const fs = require('fs');
+
+// Load your email (string or Buffer accepted here)
+const email = fs.readFileSync('email.eml');
+
+const result = await parseEmail(email);
+
+// lastMessage: Pure HTML content with replies/signatures stripped
+console.log(result.lastMessage);
+
+// attachments: Array of IFile objects
+console.log(result.attachments);
+
+// inlineAttachments: Array of IFile objects embedded in the email
+console.log(result.inlineAttachments);
+```

package/dist/cjs/index.d.ts

@@ -0,0 +1,3 @@
+export { parseEmail, replaceSrc } from './parser.js';
+export type { IFile, ParseEmailResult } from './types.js';
+//# sourceMappingURL=index.d.ts.map

package/dist/cjs/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,UAAU,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AAGpD,YAAY,EAAE,KAAK,EAAE,gBAAgB,EAAE,MAAM,YAAY,CAAA"}

package/dist/cjs/index.js

@@ -0,0 +1,7 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.replaceSrc = exports.parseEmail = void 0;
+// Public API
+var parser_js_1 = require("./parser.js");
+Object.defineProperty(exports, "parseEmail", { enumerable: true, get: function () { return parser_js_1.parseEmail; } });
+Object.defineProperty(exports, "replaceSrc", { enumerable: true, get: function () { return parser_js_1.replaceSrc; } });

package/dist/cjs/package.json

@@ -0,0 +1 @@
+{"type":"commonjs"}

package/dist/cjs/parser.d.ts

@@ -0,0 +1,4 @@
+import { IFile, ParseEmailResult } from './types.js';
+export declare function parseEmail(buffer: Buffer | string): Promise<ParseEmailResult>;
+export declare function replaceSrc(html: string, files: IFile[]): string;
+//# sourceMappingURL=parser.d.ts.map

package/dist/cjs/parser.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"parser.d.ts","sourceRoot":"","sources":["../../src/parser.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,KAAK,EAAE,gBAAgB,EAAE,MAAM,YAAY,CAAC;AAErD,wBAAsB,UAAU,CAC9B,MAAM,EAAE,MAAM,GAAG,MAAM,GACtB,OAAO,CAAC,gBAAgB,CAAC,CAkB3B;AAuDD,wBAAgB,UAAU,CAAC,IAAI,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK,EAAE,GAAG,MAAM,CAW/D"}

package/dist/cjs/parser.js

@@ -0,0 +1,204 @@
+"use strict";
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.parseEmail = parseEmail;
+exports.replaceSrc = replaceSrc;
+const mailparser_1 = require("mailparser");
+const cheerio = __importStar(require("cheerio"));
+async function parseEmail(buffer) {
+    const parsedEmail = await (0, mailparser_1.simpleParser)(buffer);
+    let content = parsedEmail.html;
+    if (!content) {
+        content = parsedEmail.text || '';
+    }
+    const cleanedContent = extractFirstMessage(content);
+    const attachments = await attachmentsToFiles(parsedEmail.attachments || []);
+    const inlineAttachments = inlineAttachmentsToFiles(cleanedContent);
+    return {
+        lastMessage: cleanedContent,
+        attachments,
+        inlineAttachments,
+    };
+}
+async function attachmentsToFiles(attach) {
+    const files = [];
+    for (const att of attach) {
+        const filename = att.filename || new Date().toDateString();
+        const contentType = att.contentType || 'application/octet-stream';
+        const attachment = {
+            name: filename,
+            type: contentType,
+            size: att.size,
+            buffer: att.content,
+        };
+        files.push(attachment);
+    }
+    return files;
+}
+function inlineAttachmentsToFiles(content) {
+    const files = [];
+    const $ = cheerio.load(content);
+    // Base64 embedded images — extract them as IFile entries
+    const base64Images = $('img[src^="data:"]').toArray();
+    for (const el of base64Images) {
+        const src = $(el).attr('src');
+        if (!src)
+            continue;
+        const matches = src.match(/^data:(image\/(\w+));base64,(.*)$/);
+        if (matches) {
+            const contentType = matches[1];
+            const extension = matches[2];
+            const base64Data = matches[3];
+            const buffer = Buffer.from(base64Data, 'base64');
+            files.push({
+                name: `inline-${new Date().toISOString()}.${extension}`,
+                src: src, // if we upload to cloud, replace src
+                type: contentType,
+                size: buffer.byteLength,
+                buffer,
+                originalSrc: src,
+            });
+        }
+    }
+    return files;
+}
+// Replace src attributes with the uploaded file URLs
+function replaceSrc(html, files) {
+    const $ = cheerio.load(html);
+    for (const file of files) {
+        if (file.originalSrc && file.src) {
+            $(`img[src="${file.originalSrc}"]`).attr('src', file.src);
+        }
+    }
+    return $.html();
+}
+function extractFirstMessage(html) {
+    const $ = cheerio.load(html);
+    // Remove text dividers (e.g. ----- Original Message -----) and everything after them
+    let dividerNode = null;
+    let isTextNode = false;
+    function findDividerNode(parent) {
+        const contents = $(parent).contents();
+        for (let i = 0; i < contents.length; i++) {
+            const node = contents[i];
+            if (node.type === 'text') {
+                const text = $(node).text();
+                const match = text.match(/-{5,}.*?-{5,}/);
+                if (match) {
+                    const index = text.indexOf(match[0]);
+                    node.data = text.substring(0, index);
+                    dividerNode = node;
+                    isTextNode = true;
+                    return true;
+                }
+            }
+            else if (node.type === 'tag') {
+                if (node.name !== 'style' && node.name !== 'script') {
+                    const clone = $(node).clone();
+                    clone.find('style, script').remove();
+                    if (/-{5,}.*?-{5,}/.test(clone.text())) {
+                        const foundInChild = findDividerNode(node);
+                        if (!foundInChild) {
+                            dividerNode = node;
+                            isTextNode = false;
+                        }
+                        return true;
+                    }
+                }
+            }
+        }
+        return false;
+    }
+    findDividerNode($('body'));
+    if (dividerNode) {
+        let current = $(dividerNode);
+        let isFirst = true;
+        while (current.length > 0 && !current.is('body')) {
+            current.nextAll().remove();
+            const parent = current.parent();
+            if (isFirst) {
+                if (!isTextNode) {
+                    current.remove();
+                }
+                isFirst = false;
+            }
+            current = parent;
+        }
+    }
+    // Gmail signature
+    $('span.gmail_signature_prefix').remove();
+    $('div.gmail_signature').remove();
+    // Gmail quoted replies
+    $('.gmail_chip').remove();
+    $('.gmail_quote').nextAll().remove();
+    $('.gmail_quote').remove();
+    // Outlook signature / forwarded messages
+    $('#divRplyFwdMsg').nextAll().remove();
+    $('#divRplyFwdMsg').remove();
+    // Outlook signature marker: <div id="Signature">
+    $('div#Signature').remove();
+    // Apple Mail signature
+    // Apple Mail wraps signatures in <div dir="ltr" class="AppleMailSignature">
+    $('div.AppleMailSignature').remove();
+    // Apple Mail quoted replies
+    // Apple Mail wraps quoted content in <blockquote type="cite">
+    $('blockquote[type="cite"]').nextAll().remove();
+    $('blockquote[type="cite"]').remove();
+    // Fallback for other email providers
+    // Matches any div whose id or class contains "signature" (case-insensitive)
+    $('div')
+        .filter((_, el) => {
+        const id = ($(el).attr('id') || '').toLowerCase();
+        const cls = ($(el).attr('class') || '').toLowerCase();
+        return id.includes('signature') || cls.includes('signature');
+    })
+        .remove();
+    // Fallback for quoted replies from other email providers
+    // Matches any div whose id or class contains "quote" or "reply" (case-insensitive)
+    $('div')
+        .filter((_, el) => {
+        const id = ($(el).attr('id') || '').toLowerCase();
+        const cls = ($(el).attr('class') || '').toLowerCase();
+        return (id.includes('quote') ||
+            id.includes('reply') ||
+            cls.includes('quote') ||
+            cls.includes('reply'));
+    })
+        .each((_, el) => {
+        $(el).nextAll().remove();
+        $(el).remove();
+    });
+    return $('body').html() || '';
+}

package/dist/cjs/types.d.ts

@@ -0,0 +1,14 @@
+export interface IFile {
+    name: string;
+    src?: string;
+    type: string;
+    size: number;
+    buffer: Buffer;
+    originalSrc?: string;
+}
+export interface ParseEmailResult {
+    lastMessage: string;
+    attachments: IFile[];
+    inlineAttachments: IFile[];
+}
+//# sourceMappingURL=types.d.ts.map

package/dist/cjs/types.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../src/types.ts"],"names":[],"mappings":"AAAA,MAAM,WAAW,KAAK;IACpB,IAAI,EAAE,MAAM,CAAC;IACb,GAAG,CAAC,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,MAAM,CAAC;IACf,WAAW,CAAC,EAAE,MAAM,CAAC;CACtB;AAED,MAAM,WAAW,gBAAgB;IAC/B,WAAW,EAAE,MAAM,CAAC;IACpB,WAAW,EAAE,KAAK,EAAE,CAAC;IACrB,iBAAiB,EAAE,KAAK,EAAE,CAAC;CAC5B"}

package/dist/cjs/types.js

@@ -0,0 +1,2 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });

package/dist/index.cjs

@@ -0,0 +1,190 @@
+var __create = Object.create;
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __getProtoOf = Object.getPrototypeOf;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toESM = (mod, isNodeMode, target) => (target = mod != null ? __create(__getProtoOf(mod)) : {}, __copyProps(
+  // If the importer is in node compatibility mode or this is not an ESM
+  // file that has been converted to a CommonJS file using a Babel-
+  // compatible transform (i.e. "__esModule" has not been set), then set
+  // "default" to the CommonJS "module.exports" for node compatibility.
+  isNodeMode || !mod || !mod.__esModule ? __defProp(target, "default", { value: mod, enumerable: true }) : target,
+  mod
+));
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/index.ts
+var index_exports = {};
+__export(index_exports, {
+  parseEmail: () => parseEmail
+});
+module.exports = __toCommonJS(index_exports);
+
+// src/parser.ts
+var import_mailparser = require("mailparser");
+var cheerio = __toESM(require("cheerio"), 1);
+async function parseEmail(buffer, uploader, uploadInlineFilesWithSrc) {
+  const parsedEmail = await (0, import_mailparser.simpleParser)(buffer);
+  let content = parsedEmail.html;
+  if (!content) {
+    content = parsedEmail.text || "";
+  }
+  let cleanedContent = extractFirstMessage(content);
+  let attachments = [];
+  let inlineAttachments = [];
+  const pendingAttachments = await attachmentsToFiles(parsedEmail.attachments);
+  const pendingInline = inlineAttachmentsToFiles(cleanedContent);
+  if (uploader) {
+    attachments = await uploadFiles(pendingAttachments, uploader);
+    if (uploadInlineFilesWithSrc) {
+      inlineAttachments = await uploadFiles(pendingInline, uploader);
+      cleanedContent = replaceSrc(inlineAttachments, cleanedContent);
+    } else {
+      inlineAttachments = pendingInline;
+    }
+  }
+  return {
+    attachments,
+    inlineAttachments,
+    lastMessage: cleanedContent
+  };
+}
+async function attachmentsToFiles(attach) {
+  const files = [];
+  for (const att of attach) {
+    const filename = att.filename || (/* @__PURE__ */ new Date()).toDateString();
+    const contentType = att.contentType || "application/octet-stream";
+    const attachment = {
+      name: filename,
+      type: contentType,
+      size: att.size,
+      buffer: att.content
+    };
+    files.push(attachment);
+  }
+  return files;
+}
+function inlineAttachmentsToFiles(content) {
+  const files = [];
+  const $ = cheerio.load(content);
+  const base64Images = $('img[src^="data:"]').toArray();
+  for (const el of base64Images) {
+    const src = $(el).attr("src");
+    if (!src) continue;
+    const matches = src.match(/^data:(image\/(\w+));base64,(.*)$/);
+    if (matches) {
+      const contentType = matches[1];
+      const extension = matches[2];
+      const base64Data = matches[3];
+      const buffer = Buffer.from(base64Data, "base64");
+      files.push({
+        name: `inline-${(/* @__PURE__ */ new Date()).toISOString()}.${extension}`,
+        src,
+        // keep the data URI as the src
+        type: contentType,
+        size: buffer.byteLength,
+        buffer,
+        originalSrc: src
+      });
+    }
+  }
+  return files;
+}
+async function uploadFiles(files, uploader) {
+  const failedFiles = [];
+  const uploadedFiles = [];
+  for (const file of files) {
+    try {
+      const uploadedFile = await uploader({
+        buffer: file.buffer || file.src || "",
+        contentType: file.type,
+        filename: file.name || "unnamed"
+      });
+      const { buffer, ...uploadedFileObj } = file;
+      uploadedFiles.push({
+        ...uploadedFileObj,
+        src: uploadedFile
+      });
+    } catch (err) {
+      console.error(`Failed to upload file: ${file.name}`, err);
+      failedFiles.push(file);
+    }
+  }
+  if (failedFiles.length > 0) {
+    throw new Error(`Failed to upload: ${failedFiles.map((f) => f.name).join(", ")}`);
+  }
+  return uploadedFiles;
+}
+function replaceSrc(inlineAttachments, html) {
+  const $ = cheerio.load(html);
+  for (const file of inlineAttachments) {
+    if (file.originalSrc) {
+      $(`img[src="${file.originalSrc}"]`).attr("src", file.src);
+    }
+  }
+  return $.html();
+}
+function extractFirstMessage(html) {
+  const $ = cheerio.load(html);
+  $("span.gmail_signature_prefix").remove();
+  $("div.gmail_signature").remove();
+  $(".gmail_chip").remove();
+  $(".gmail_quote").nextAll().remove();
+  $(".gmail_quote").remove();
+  $("#divRplyFwdMsg").nextAll().remove();
+  $("#divRplyFwdMsg").remove();
+  $("div#Signature").remove();
+  $("div.AppleMailSignature").remove();
+  $('blockquote[type="cite"]').nextAll().remove();
+  $('blockquote[type="cite"]').remove();
+  $("div").filter((_, el) => {
+    const id = ($(el).attr("id") || "").toLowerCase();
+    const cls = ($(el).attr("class") || "").toLowerCase();
+    return id.includes("signature") || cls.includes("signature");
+  }).remove();
+  $("div").filter((_, el) => {
+    const id = ($(el).attr("id") || "").toLowerCase();
+    const cls = ($(el).attr("class") || "").toLowerCase();
+    return id.includes("quote") || id.includes("reply") || cls.includes("quote") || cls.includes("reply");
+  }).each((_, el) => {
+    $(el).nextAll().remove();
+    $(el).remove();
+  });
+  $("body").find("div, p, font, span, td").filter(function() {
+    if ($(this).children().not("br").length > 0) {
+      return false;
+    }
+    const elementText = $(this).text().toLowerCase();
+    return [
+      "this e-mail and any attachments may contain confidential",
+      "sent from"
+    ].some((text) => elementText.includes(text));
+  }).remove();
+  $("div").filter((_, el) => {
+    if ($(el).closest("li").length > 0) {
+      return false;
+    }
+    if ($(el).find("img").length > 0) {
+      return false;
+    }
+    return $(el).text().trim().length === 0 && $(el).children().not("br").length === 0;
+  }).remove();
+  return $("body").html() || "";
+}
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  parseEmail
+});

package/dist/index.d.ts

@@ -1,2 +1,3 @@
-export * from './parser.js';
+export { parseEmail, replaceSrc } from './parser.js';
+export type { IFile, ParseEmailResult } from './types.js';
 //# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,cAAc,aAAa,CAAA"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,UAAU,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AAGpD,YAAY,EAAE,KAAK,EAAE,gBAAgB,EAAE,MAAM,YAAY,CAAA"}

package/dist/index.js

@@ -1 +1,2 @@
-export * from './parser.js';
+// Public API
+export { parseEmail, replaceSrc } from './parser.js';

package/dist/parser.d.ts

@@ -1,30 +1,4 @@
-import { ParsedMail } from "mailparser";
-export interface IFile {
-    name?: string;
-    url: string;
-    type?: string;
-    size?: number;
-    tags?: Record<string, string>;
-}
-export interface UploadFileOptions {
-    buffer: Buffer | string;
-    contentType: string;
-    filename: string;
-}
-export type UploadFunction = (file: UploadFileOptions) => Promise<string>;
-export declare function processInlineImages(html: string, cidMap: Map<string, string>, uploader: UploadFunction): Promise<string>;
-export declare function processEmailAttachments(parsedEmail: ParsedMail, uploader: UploadFunction): Promise<{
-    attachments: IFile[];
-    inlineMap: Map<string, string>;
-}>;
-export declare function extractFirstMessage(html: string): string;
-export interface ParseEmailOptions {
-    buffer: Buffer | string;
-    uploader: UploadFunction;
-}
-export interface ParseEmailResult {
-    attachments: IFile[];
-    response: string;
-}
-export declare function parseEmail(options: ParseEmailOptions): Promise<ParseEmailResult>;
+import { IFile, ParseEmailResult } from './types.js';
+export declare function parseEmail(buffer: Buffer | string): Promise<ParseEmailResult>;
+export declare function replaceSrc(html: string, files: IFile[]): string;
 //# sourceMappingURL=parser.d.ts.map

package/dist/parser.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"parser.d.ts","sourceRoot":"","sources":["../src/parser.ts"],"names":[],"mappings":"AAAA,OAAO,EAAgB,UAAU,EAAE,MAAM,YAAY,CAAC;AAItD,MAAM,WAAW,KAAK;IAClB,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,GAAG,EAAE,MAAM,CAAC;IACZ,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,IAAI,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CACjC;AAED,MAAM,WAAW,iBAAiB;IAC9B,MAAM,EAAE,MAAM,GAAG,MAAM,CAAC;IACxB,WAAW,EAAE,MAAM,CAAC;IACpB,QAAQ,EAAE,MAAM,CAAC;CACpB;AAED,MAAM,MAAM,cAAc,GAAG,CAAC,IAAI,EAAE,iBAAiB,KAAK,OAAO,CAAC,MAAM,CAAC,CAAC;AAE1E,wBAAsB,mBAAmB,CACrC,IAAI,EAAE,MAAM,EACZ,MAAM,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,EAC3B,QAAQ,EAAE,cAAc,GACzB,OAAO,CAAC,MAAM,CAAC,CA8CjB;AAED,wBAAsB,uBAAuB,CACzC,WAAW,EAAE,UAAU,EACvB,QAAQ,EAAE,cAAc,GACzB,OAAO,CAAC;IAAE,WAAW,EAAE,KAAK,EAAE,CAAC;IAAC,SAAS,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;CAAE,CAAC,CA6CnE;AAED,wBAAgB,mBAAmB,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CA0FxD;AAED,MAAM,WAAW,iBAAiB;IAC9B,MAAM,EAAE,MAAM,GAAG,MAAM,CAAC;IACxB,QAAQ,EAAE,cAAc,CAAC;CAC5B;AAED,MAAM,WAAW,gBAAgB;IAC7B,WAAW,EAAE,KAAK,EAAE,CAAC;IACrB,QAAQ,EAAE,MAAM,CAAC;CACpB;AAED,wBAAsB,UAAU,CAAC,OAAO,EAAE,iBAAiB,GAAG,OAAO,CAAC,gBAAgB,CAAC,CAmBtF"}
+{"version":3,"file":"parser.d.ts","sourceRoot":"","sources":["../src/parser.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,KAAK,EAAE,gBAAgB,EAAE,MAAM,YAAY,CAAC;AAErD,wBAAsB,UAAU,CAC9B,MAAM,EAAE,MAAM,GAAG,MAAM,GACtB,OAAO,CAAC,gBAAgB,CAAC,CAkB3B;AAuDD,wBAAgB,UAAU,CAAC,IAAI,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK,EAAE,GAAG,MAAM,CAW/D"}

package/dist/parser.js

@@ -1,22 +1,39 @@
-import { simpleParser } from "mailparser";
-import * as cheerio from "cheerio";
-import { v4 as uuidv4 } from "uuid";
-export async function processInlineImages(html, cidMap, uploader) {
-    const $ = cheerio.load(html);
-    // remove width and height attributes from <img> tags
-    $('img').each((_, el) => {
-        $(el).removeAttr('width');
-        $(el).removeAttr('height');
-    });
-    // replace inline attachments
-    $('img[src^="cid:"]').each((_, el) => {
-        const src = $(el).attr('src');
-        const cid = src?.replace('cid:', '').trim();
-        if (cid && cidMap.has(cid)) {
-            $(el).attr('src', cidMap.get(cid));
-        }
-    });
-    // handle base64 images
+import { simpleParser } from 'mailparser';
+import * as cheerio from 'cheerio';
+export async function parseEmail(buffer) {
+    const parsedEmail = await simpleParser(buffer);
+    let content = parsedEmail.html;
+    if (!content) {
+        content = parsedEmail.text || '';
+    }
+    const cleanedContent = extractFirstMessage(content);
+    const attachments = await attachmentsToFiles(parsedEmail.attachments || []);
+    const inlineAttachments = inlineAttachmentsToFiles(cleanedContent);
+    return {
+        lastMessage: cleanedContent,
+        attachments,
+        inlineAttachments,
+    };
+}
+async function attachmentsToFiles(attach) {
+    const files = [];
+    for (const att of attach) {
+        const filename = att.filename || new Date().toDateString();
+        const contentType = att.contentType || 'application/octet-stream';
+        const attachment = {
+            name: filename,
+            type: contentType,
+            size: att.size,
+            buffer: att.content,
+        };
+        files.push(attachment);
+    }
+    return files;
+}
+function inlineAttachmentsToFiles(content) {
+    const files = [];
+    const $ = cheerio.load(content);
+    // Base64 embedded images — extract them as IFile entries
     const base64Images = $('img[src^="data:"]').toArray();
     for (const el of base64Images) {
         const src = $(el).attr('src');
@@ -25,65 +42,84 @@ export async function processInlineImages(html, cidMap, uploader) {
         const matches = src.match(/^data:(image\/(\w+));base64,(.*)$/);
         if (matches) {
             const contentType = matches[1];
+            const extension = matches[2];
             const base64Data = matches[3];
-            const filename = `inline-${uuidv4()}`;
             const buffer = Buffer.from(base64Data, 'base64');
-            try {
-                const url = await uploader({
-                    buffer,
-                    contentType,
-                    filename
-                });
-                $(el).attr('src', url);
-            }
-            catch (err) {
-                console.error(`Failed to upload base64 inline image: ${err}`);
-            }
+            files.push({
+                name: `inline-${new Date().toISOString()}.${extension}`,
+                src: src, // if we upload to cloud, replace src
+                type: contentType,
+                size: buffer.byteLength,
+                buffer,
+                originalSrc: src,
+            });
+        }
+    }
+    return files;
+}
+// Replace src attributes with the uploaded file URLs
+export function replaceSrc(html, files) {
+    const $ = cheerio.load(html);
+    for (const file of files) {
+        if (file.originalSrc && file.src) {
+            $(`img[src="${file.originalSrc}"]`).attr('src', file.src);
         }
     }
     return $.html();
 }
-export async function processEmailAttachments(parsedEmail, uploader) {
-    const attachments = [];
-    const inlineMap = new Map();
-    if (parsedEmail.attachments && parsedEmail.attachments.length > 0) {
-        for (const att of parsedEmail.attachments) {
-            const filename = att.filename || uuidv4();
-            const contentType = att.contentType || 'application/octet-stream';
-            try {
-                const publicUrl = await uploader({
-                    buffer: att.content,
-                    contentType,
-                    filename
-                });
-                const attachment = {
-                    name: att.filename,
-                    url: publicUrl,
-                    type: att.contentType,
-                    size: att.size,
-                };
-                attachment.tags = att.related
-                    ? { inlineAttachment: 'true' }
-                    : { attachment: 'true' };
-                attachments.push(attachment);
-                // Map Content-ID to URL (mailparser wraps CID in < >, strip them)
-                if (att.contentId) {
-                    inlineMap.set(att.contentId.replace(/[<>]/g, ''), publicUrl);
+function extractFirstMessage(html) {
+    const $ = cheerio.load(html);
+    // Remove text dividers (e.g. ----- Original Message -----) and everything after them
+    let dividerNode = null;
+    let isTextNode = false;
+    function findDividerNode(parent) {
+        const contents = $(parent).contents();
+        for (let i = 0; i < contents.length; i++) {
+            const node = contents[i];
+            if (node.type === 'text') {
+                const text = $(node).text();
+                const match = text.match(/-{5,}.*?-{5,}/);
+                if (match) {
+                    const index = text.indexOf(match[0]);
+                    node.data = text.substring(0, index);
+                    dividerNode = node;
+                    isTextNode = true;
+                    return true;
                 }
-                // Also map by filename so we can match email-provider proxy URLs
-                if (att.filename) {
-                    inlineMap.set(att.filename, publicUrl);
+            }
+            else if (node.type === 'tag') {
+                if (node.name !== 'style' && node.name !== 'script') {
+                    const clone = $(node).clone();
+                    clone.find('style, script').remove();
+                    if (/-{5,}.*?-{5,}/.test(clone.text())) {
+                        const foundInChild = findDividerNode(node);
+                        if (!foundInChild) {
+                            dividerNode = node;
+                            isTextNode = false;
+                        }
+                        return true;
+                    }
                 }
             }
-            catch (error) {
-                console.error(`Failed to upload attachment ${filename}: ${error}`);
+        }
+        return false;
+    }
+    findDividerNode($('body'));
+    if (dividerNode) {
+        let current = $(dividerNode);
+        let isFirst = true;
+        while (current.length > 0 && !current.is('body')) {
+            current.nextAll().remove();
+            const parent = current.parent();
+            if (isFirst) {
+                if (!isTextNode) {
+                    current.remove();
+                }
+                isFirst = false;
             }
+            current = parent;
         }
     }
-    return { attachments, inlineMap };
-}
-export function extractFirstMessage(html) {
-    const $ = cheerio.load(html);
     // Gmail signature
     $('span.gmail_signature_prefix').remove();
     $('div.gmail_signature').remove();
@@ -127,52 +163,5 @@ export function extractFirstMessage(html) {
         $(el).nextAll().remove();
         $(el).remove();
     });
-    // Remove confidentiality disclaimers
-    $('body')
-        .find('div, p, font, span, td')
-        .filter(function () {
-        // Skip if this element has child elements that aren't just <br>
-        if ($(this).children().not('br').length > 0) {
-            return false;
-        }
-        const elementText = $(this).text().toLowerCase();
-        return [
-            'this e-mail and any attachments may contain confidential',
-            'sent from'
-        ].some((text) => elementText.includes(text));
-    })
-        .remove();
-    // Clean empty divs
-    $('div')
-        .filter((_, el) => {
-        // Avoid removing divs that are direct children of list items, as this can break bullet points
-        if ($(el).closest('li').length > 0) {
-            return false;
-        }
-        // Avoid removing divs that contain images (inline images, etc.)
-        if ($(el).find('img').length > 0) {
-            return false;
-        }
-        return ($(el).text().trim().length === 0 &&
-            $(el).children().not('br').length === 0);
-    })
-        .remove();
     return $('body').html() || '';
 }
-export async function parseEmail(options) {
-    const { buffer, uploader } = options;
-    const parsedEmail = await simpleParser(buffer);
-    const { attachments, inlineMap } = await processEmailAttachments(parsedEmail, uploader);
-    let response = parsedEmail.html;
-    if (response) {
-        response = await processInlineImages(response, inlineMap, uploader);
-        response = extractFirstMessage(response);
-    }
-    else {
-        response = parsedEmail.text || '';
-    }
-    return {
-        attachments,
-        response,
-    };
-}

package/dist/types.d.ts

@@ -0,0 +1,14 @@
+export interface IFile {
+    name: string;
+    src?: string;
+    type: string;
+    size: number;
+    buffer: Buffer;
+    originalSrc?: string;
+}
+export interface ParseEmailResult {
+    lastMessage: string;
+    attachments: IFile[];
+    inlineAttachments: IFile[];
+}
+//# sourceMappingURL=types.d.ts.map

package/dist/types.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../src/types.ts"],"names":[],"mappings":"AAAA,MAAM,WAAW,KAAK;IACpB,IAAI,EAAE,MAAM,CAAC;IACb,GAAG,CAAC,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,MAAM,CAAC;IACf,WAAW,CAAC,EAAE,MAAM,CAAC;CACtB;AAED,MAAM,WAAW,gBAAgB;IAC/B,WAAW,EAAE,MAAM,CAAC;IACpB,WAAW,EAAE,KAAK,EAAE,CAAC;IACrB,iBAAiB,EAAE,KAAK,EAAE,CAAC;CAC5B"}

package/dist/types.js

@@ -0,0 +1 @@
+export {};

package/package.json

@@ -1,27 +1,38 @@
 {
   "name": "lite-email-parser",
-  "version": "1.0.1",
+  "version": "1.0.2",
   "description": "Parse email to get signature, replies, attachments and inline images",
-  "main": "dist/index.js",
-  "types": "dist/index.d.ts",
+  "main": "./dist/cjs/index.js",
+  "module": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": {
+      "import": {
+        "types": "./dist/index.d.ts",
+        "default": "./dist/index.js"
+      },
+      "require": {
+        "types": "./dist/cjs/index.d.ts",
+        "default": "./dist/cjs/index.js"
+      }
+    }
+  },
   "files": [
     "dist/"
   ],
   "scripts": {
-    "build": "npx tsc",
-    "start": "node ./dist/src/index.js",
+    "build": "npx tsc && npx tsc --module commonjs --outDir dist/cjs --declaration && echo '{\"type\":\"commonjs\"}' > dist/cjs/package.json",
+    "start": "node ./dist/index.js",
     "test": "npx -y tsx sample/sample.ts"
   },
   "author": "philipedc",
   "type": "module",
   "dependencies": {
     "cheerio": "^1.2.0",
-    "mailparser": "^3.9.14",
-    "uuid": "^14.0.1"
+    "mailparser": "^3.9.14"
   },
   "devDependencies": {
     "@types/mailparser": "^3.4.6",
-    "@types/uuid": "^10.0.0",
     "typescript": "^7.0.2"
   }
-}
+}

package/dist/src/index.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":""}

package/dist/src/index.js

@@ -1,2 +0,0 @@
-console.log("hi");
-export {};