@volter/twin-gmail 1.0.0 → 1.0.2

package/dist/src/engine/mail.js

@@ -1,89 +1,19 @@
-/** Headers and body split at the first empty line; folded headers unfolded, names lowercased (RFC 5322 §2.2.3). */
-// source: https://www.rfc-editor.org/rfc/rfc5322 "Unfolding is accomplished by simply removing any CRLF that is immediately followed by WSP"
-function split(raw) {
-    const at = raw.search(/\r?\n\r?\n/);
-    const head = at < 0 ? raw : raw.slice(0, at);
-    const body = at < 0 ? '' : raw.slice(at).replace(/^\r?\n\r?\n/, '');
-    const headers = {};
-    for (const line of head.replace(/\r?\n[ \t]+/g, ' ').split(/\r?\n/)) {
-        const i = line.indexOf(':');
-        if (i > 0)
-            headers[line.slice(0, i).trim().toLowerCase()] ??= utf8(bytesOf(line.slice(i + 1).trim()));
-    }
-    return { headers, body };
-}
-const param = (value, name) => {
-    const m = new RegExp(`;\\s*${name}\\*?=\\s*(?:"([^"]*)"|([^;\\s]+))`, 'i').exec(value);
-    const v = m?.[1] ?? m?.[2];
-    return v?.startsWith("utf-8''") || v?.startsWith("UTF-8''") ? decodeURIComponent(v.slice(7)) : v;
-};
-const bytesOf = (s) => Uint8Array.from(s, (c) => c.charCodeAt(0) & 0xff);
-const utf8 = (b) => new TextDecoder().decode(b);
-const textFor = (b, charset) => { try {
-    return new TextDecoder(charset ?? 'utf-8').decode(b);
-}
-catch {
-    return utf8(b);
-} };
-/** A part's body decoded by its Content-Transfer-Encoding: base64 and quoted-printable to bytes (RFC 2045 §6). */
-function decoded(part) {
-    const enc = (part.headers['content-transfer-encoding'] ?? '7bit').toLowerCase();
-    if (enc === 'base64') {
-        try {
-            return bytesOf(atob(part.body.replace(/\s+/g, '')));
-        }
-        catch {
-            return bytesOf(part.body);
-        }
-    }
-    if (enc === 'quoted-printable')
-        return bytesOf(part.body.replace(/=\r?\n/g, '').replace(/=([0-9A-Fa-f]{2})/g, (_, h) => String.fromCharCode(Number.parseInt(h, 16))).split('').map((c) => (c.charCodeAt(0) > 0xff ? '?' : c)).join(''));
-    return bytesOf(part.body);
-}
-/** Encoded words in a header (`=?UTF-8?B?…?=`, `=?UTF-8?Q?…?=`), decoded (RFC 2047). */
-// source: https://www.rfc-editor.org/rfc/rfc2047 "encoded-word = \"=?\" charset \"?\" encoding \"?\" encoded-text \"?=\""
-function words(value) {
-    return value.replace(/=\?([^?]+)\?([BbQq])\?([^?]*)\?=(\s+(?==\?))?/g, (_, _cs, e, t) => {
-        try {
-            return textFor(e.toUpperCase() === 'B' ? bytesOf(atob(t)) : bytesOf(t.replace(/_/g, ' ').replace(/=([0-9A-Fa-f]{2})/g, (_m, h) => String.fromCharCode(Number.parseInt(h, 16)))), _cs);
-        }
-        catch {
-            return t;
-        }
-    });
-}
-const addresses = (v) => (v ? words(v).split(/,(?=(?:[^"]*"[^"]*")*[^"]*$)/).map((a) => (/<([^>]+)>/.exec(a)?.[1] ?? a).trim()).filter(Boolean) : []);
-/** The message's parts: a multipart's each (recursively), else the one. */
-function leaves(part) {
-    const type = part.headers['content-type'] ?? 'text/plain';
-    const boundary = /^multipart\//i.test(type) ? param(type, 'boundary') : undefined;
-    if (!boundary)
-        return [part];
-    const pieces = part.body.split(new RegExp(`\\r?\\n?--${boundary.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')}(?:--)?[ \\t]*\\r?\\n?`));
-    return pieces.slice(1, -1).filter((p) => p.trim()).flatMap((p) => leaves(split(p)));
-}
+// A message as RFC 5322 and MIME make of it (its headers, its text and HTML, its attachments), from the bytes Gmail's
+// receiving server accepted, read by the kernel's one mail reader (world-core readMailMessage): pure, so the server's
+// front keeps it in each recipient's mailbox (semantics/shared.ts `deliver`). The data arrives with its dot-stuffing
+// undone (../stream.ts); the World keeps the bytes as sent.
+import { mailAddressList, mailWords, readMailMessage } from '@volter/world-core';
+/** An address field's mailboxes: each member's `<addr>`, or the member as written. */
+const addresses = (v) => mailAddressList(v).map((a) => /<([^>]+)>/.exec(a)?.[1] ?? a);
 export function parse(data) {
-    const top = split(data);
-    let text = null;
-    let html = null;
-    const attachments = [];
-    for (const leaf of leaves(top)) {
-        const type = (leaf.headers['content-type'] ?? 'text/plain').split(';')[0].trim().toLowerCase();
-        const disposition = leaf.headers['content-disposition'] ?? '';
-        const filename = param(disposition, 'filename') ?? param(leaf.headers['content-type'] ?? '', 'name');
-        const bytes = decoded(leaf);
-        if (!/^attachment/i.test(disposition) && !filename && type === 'text/plain' && text === null) {
-            text = textFor(bytes, param(leaf.headers['content-type'] ?? '', 'charset'));
-            continue;
-        }
-        if (!/^attachment/i.test(disposition) && !filename && type === 'text/html' && html === null) {
-            html = textFor(bytes, param(leaf.headers['content-type'] ?? '', 'charset'));
-            continue;
-        }
-        let bin = '';
-        for (const b of bytes)
-            bin += String.fromCharCode(b);
-        attachments.push({ filename: filename ? words(filename) : null, contentType: type, size: bytes.length, contentId: leaf.headers['content-id']?.replace(/^<|>$/g, '') ?? null, base64: btoa(bin) });
-    }
-    return { headers: top.headers, subject: words(top.headers.subject ?? ''), from: addresses(top.headers.from)[0] ?? '', to: addresses(top.headers.to), cc: addresses(top.headers.cc), text, html, attachments };
+    const { headers, text, html, attachments } = readMailMessage(data);
+    return {
+        headers, subject: mailWords(headers.subject ?? ''), from: addresses(headers.from)[0] ?? '', to: addresses(headers.to), cc: addresses(headers.cc), text, html,
+        attachments: attachments.map(({ filename, contentType, contentId, bytes }) => {
+            let bin = '';
+            for (const b of bytes)
+                bin += String.fromCharCode(b);
+            return { filename, contentType, size: bytes.length, contentId, base64: btoa(bin) };
+        }),
+    };
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@volter/twin-gmail",
-  "version": "1.0.0",
+  "version": "1.0.2",
   "description": "Gmail twin: gmail.com's receiving SMTP server for the World's Google accounts, and the mail.google.com webmail where the person reads their mail.",
   "author": "Volter (https://github.com/volter-ai)",
   "license": "Apache-2.0",
@@ -29,7 +29,7 @@
   ],
   "peerDependencies": {
     "@volter/twin-googleoauth": "^1.1.2",
-    "@volter/world-core": "^3.0.185"
+    "@volter/world-core": "^3.0.190"
   },
   "engines": {
     "node": ">=22.3"
@@ -49,6 +49,6 @@
     }
   },
   "volterRelease": {
-    "commit": "c41b6615f37f0e7d69753fd85ff9568d9f4c8174"
+    "commit": "71c14672a241845f01d16a370c666e263af87f55"
   }
 }

package/src/engine/mail.ts

@@ -1,76 +1,23 @@
 // A message as RFC 5322 and MIME make of it (its headers, its text and HTML, its attachments), from the bytes Gmail's
-// receiving server accepted: pure, so the server's front keeps it in each recipient's mailbox (semantics/shared.ts
-// `deliver`). The data arrives with its dot-stuffing undone (../stream.ts); the World keeps the bytes as sent.
-type Part = { headers: Record<string, string>; body: string };
+// receiving server accepted, read by the kernel's one mail reader (world-core readMailMessage): pure, so the server's
+// front keeps it in each recipient's mailbox (semantics/shared.ts `deliver`). The data arrives with its dot-stuffing
+// undone (../stream.ts); the World keeps the bytes as sent.
+import { mailAddressList, mailWords, readMailMessage } from '@volter/world-core';
+
 type Attachment = { filename: string | null; contentType: string; size: number; contentId: string | null; base64: string };
 type Parsed = { headers: Record<string, string>; subject: string; from: string; to: string[]; cc: string[]; text: string | null; html: string | null; attachments: Attachment[] };
 
-/** Headers and body split at the first empty line; folded headers unfolded, names lowercased (RFC 5322 §2.2.3). */
-// source: https://www.rfc-editor.org/rfc/rfc5322 "Unfolding is accomplished by simply removing any CRLF that is immediately followed by WSP"
-function split(raw: string): Part {
-  const at = raw.search(/\r?\n\r?\n/);
-  const head = at < 0 ? raw : raw.slice(0, at);
-  const body = at < 0 ? '' : raw.slice(at).replace(/^\r?\n\r?\n/, '');
-  const headers: Record<string, string> = {};
-  for (const line of head.replace(/\r?\n[ \t]+/g, ' ').split(/\r?\n/)) {
-    const i = line.indexOf(':');
-    if (i > 0) headers[line.slice(0, i).trim().toLowerCase()] ??= utf8(bytesOf(line.slice(i + 1).trim()));
-  }
-  return { headers, body };
-}
-
-const param = (value: string, name: string): string | undefined => {
-  const m = new RegExp(`;\\s*${name}\\*?=\\s*(?:"([^"]*)"|([^;\\s]+))`, 'i').exec(value);
-  const v = m?.[1] ?? m?.[2];
-  return v?.startsWith("utf-8''") || v?.startsWith("UTF-8''") ? decodeURIComponent(v.slice(7)) : v;
-};
-
-const bytesOf = (s: string): Uint8Array => Uint8Array.from(s, (c) => c.charCodeAt(0) & 0xff);
-const utf8 = (b: Uint8Array): string => new TextDecoder().decode(b);
-const textFor = (b: Uint8Array, charset: string | undefined): string => { try { return new TextDecoder(charset ?? 'utf-8').decode(b); } catch { return utf8(b); } };
-
-/** A part's body decoded by its Content-Transfer-Encoding: base64 and quoted-printable to bytes (RFC 2045 §6). */
-function decoded(part: Part): Uint8Array {
-  const enc = (part.headers['content-transfer-encoding'] ?? '7bit').toLowerCase();
-  if (enc === 'base64') { try { return bytesOf(atob(part.body.replace(/\s+/g, ''))); } catch { return bytesOf(part.body); } }
-  if (enc === 'quoted-printable') return bytesOf(part.body.replace(/=\r?\n/g, '').replace(/=([0-9A-Fa-f]{2})/g, (_, h: string) => String.fromCharCode(Number.parseInt(h, 16))).split('').map((c) => (c.charCodeAt(0) > 0xff ? '?' : c)).join(''));
-  return bytesOf(part.body);
-}
-
-/** Encoded words in a header (`=?UTF-8?B?…?=`, `=?UTF-8?Q?…?=`), decoded (RFC 2047). */
-// source: https://www.rfc-editor.org/rfc/rfc2047 "encoded-word = \"=?\" charset \"?\" encoding \"?\" encoded-text \"?=\""
-function words(value: string): string {
-  return value.replace(/=\?([^?]+)\?([BbQq])\?([^?]*)\?=(\s+(?==\?))?/g, (_, _cs: string, e: string, t: string) => {
-    try { return textFor(e.toUpperCase() === 'B' ? bytesOf(atob(t)) : bytesOf(t.replace(/_/g, ' ').replace(/=([0-9A-Fa-f]{2})/g, (_m, h: string) => String.fromCharCode(Number.parseInt(h, 16)))), _cs); } catch { return t; }
-  });
-}
-
-const addresses = (v: string | undefined): string[] => (v ? words(v).split(/,(?=(?:[^"]*"[^"]*")*[^"]*$)/).map((a) => (/<([^>]+)>/.exec(a)?.[1] ?? a).trim()).filter(Boolean) : []);
-
-/** The message's parts: a multipart's each (recursively), else the one. */
-function leaves(part: Part): Part[] {
-  const type = part.headers['content-type'] ?? 'text/plain';
-  const boundary = /^multipart\//i.test(type) ? param(type, 'boundary') : undefined;
-  if (!boundary) return [part];
-  const pieces = part.body.split(new RegExp(`\\r?\\n?--${boundary.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')}(?:--)?[ \\t]*\\r?\\n?`));
-  return pieces.slice(1, -1).filter((p) => p.trim()).flatMap((p) => leaves(split(p)));
-}
+/** An address field's mailboxes: each member's `<addr>`, or the member as written. */
+const addresses = (v: string | undefined): string[] => mailAddressList(v).map((a) => /<([^>]+)>/.exec(a)?.[1] ?? a);
 
 export function parse(data: string): Parsed {
-  const top = split(data);
-  let text: string | null = null;
-  let html: string | null = null;
-  const attachments: Attachment[] = [];
-  for (const leaf of leaves(top)) {
-    const type = (leaf.headers['content-type'] ?? 'text/plain').split(';')[0]!.trim().toLowerCase();
-    const disposition = leaf.headers['content-disposition'] ?? '';
-    const filename = param(disposition, 'filename') ?? param(leaf.headers['content-type'] ?? '', 'name');
-    const bytes = decoded(leaf);
-    if (!/^attachment/i.test(disposition) && !filename && type === 'text/plain' && text === null) { text = textFor(bytes, param(leaf.headers['content-type'] ?? '', 'charset')); continue; }
-    if (!/^attachment/i.test(disposition) && !filename && type === 'text/html' && html === null) { html = textFor(bytes, param(leaf.headers['content-type'] ?? '', 'charset')); continue; }
-    let bin = '';
-    for (const b of bytes) bin += String.fromCharCode(b);
-    attachments.push({ filename: filename ? words(filename) : null, contentType: type, size: bytes.length, contentId: leaf.headers['content-id']?.replace(/^<|>$/g, '') ?? null, base64: btoa(bin) });
-  }
-  return { headers: top.headers, subject: words(top.headers.subject ?? ''), from: addresses(top.headers.from)[0] ?? '', to: addresses(top.headers.to), cc: addresses(top.headers.cc), text, html, attachments };
+  const { headers, text, html, attachments } = readMailMessage(data);
+  return {
+    headers, subject: mailWords(headers.subject ?? ''), from: addresses(headers.from)[0] ?? '', to: addresses(headers.to), cc: addresses(headers.cc), text, html,
+    attachments: attachments.map(({ filename, contentType, contentId, bytes }) => {
+      let bin = '';
+      for (const b of bytes) bin += String.fromCharCode(b);
+      return { filename, contentType, size: bytes.length, contentId, base64: btoa(bin) };
+    }),
+  };
 }