Skip to content

fix: forward List-Unsubscribe headers through SMTP proxy#428

Open
mattstein wants to merge 1 commit into
usesend:mainfrom
mattstein:smtp-list-unsubscribe-passthrough
Open

fix: forward List-Unsubscribe headers through SMTP proxy#428
mattstein wants to merge 1 commit into
usesend:mainfrom
mattstein:smtp-list-unsubscribe-passthrough

Conversation

@mattstein

@mattstein mattstein commented Jul 13, 2026

Copy link
Copy Markdown
Contributor

This PR modifies the SMTP proxy to allow List-Unsubscribe and List-Unsubscribe-Post headers to forwarded to the useSend API.

Mailparser transforms List-* headers into its own internal list structure and drops them from the headers map, so this reads them headerLines, splits on :, and trims + canonicalizes the names.

I’ve been using useSend for sending all kinds of email, most recently including Listmonk campaigns for the first time. mail-tester.com flagged missing List-Unsubscribe and List-Unsubscribe-Post headers as negative score impacts, even though Listmonk adds them. There’s no easy way to use a useSend driver rather than SMTP with Listmonk, so this adjustment at least lets the proxy pass them through.


Summary by cubic

Forward List-Unsubscribe and List-Unsubscribe-Post headers through the SMTP proxy to the useSend API so unsubscribe metadata is preserved and deliverability scores improve.

  • Bug Fixes
    • Read raw headerLines to extract these headers since mailparser moves List-* into an internal structure and drops them from the headers map.
    • Canonicalize names and include them in the headers field sent to /api/v1/emails.

Written for commit 71022ba. Summary will update on new commits.

Review in cubic

Summary by CodeRabbit

  • Bug Fixes
    • Preserved List-Unsubscribe and List-Unsubscribe-Post headers when forwarding outgoing emails.
    • Improved compatibility with email clients and services that rely on unsubscribe headers.

@vercel

vercel Bot commented Jul 13, 2026

Copy link
Copy Markdown

@mattstein is attempting to deploy a commit to the kmkoushik's projects Team on Vercel.

A member of the Team first needs to authorize it.

@coderabbitai

coderabbitai Bot commented Jul 13, 2026

Copy link
Copy Markdown
Contributor

Review Change Stack

Walkthrough

The SMTP server now defines the forwarded list headers it supports, normalizes header names, and extracts non-empty values from raw parsed header lines. During onData, the extracted headers are added to the emailObject passed to sendEmailToUseSend.

Possibly related PRs

Suggested reviewers: kmkoushik

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name Status Explanation Resolution
Docstring Coverage ⚠️ Warning Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. Write docstrings for the functions missing them to satisfy the coverage threshold.
✅ Passed checks (4 passed)
Check name Status Explanation
Description Check ✅ Passed Check skipped - CodeRabbit’s high-level summary is enabled.
Title check ✅ Passed The title clearly matches the main change: forwarding List-Unsubscribe-related headers through the SMTP proxy.
Linked Issues check ✅ Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check ✅ Passed Check skipped because no linked issues were found for this pull request.

Warning

There were issues while running some tools. Please review the errors and either fix the tool's configuration or disable the tool if it's a critical failure.

🔧 ESLint

If the error stems from missing dependencies, add them to the package.json file. For unrecoverable errors (e.g., due to private dependencies), disable the tool in the CodeRabbit configuration.

apps/smtp-server/src/server.ts

ESLint skipped: missing config or dependency (missing-dependency). The ESLint configuration references a package that is not available in the sandbox.


Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@coderabbitai coderabbitai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@apps/smtp-server/src/server.ts`:
- Around line 43-45: Update the header parsing flow around canonicalHeaderName
to unfold folded header values before storing them in result: replace embedded
CRLF followed by whitespace with a single space, then trim the normalized value
and preserve the existing non-empty check and forwarding behavior.
🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

  • Push a commit to this branch (recommended)
  • Create a new PR with the fixes

ℹ️ Review info
⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: d474c6d7-6116-49ab-a625-d8a13c9e582f

📥 Commits

Reviewing files that changed from the base of the PR and between 89cba7f and 71022ba.

📒 Files selected for processing (1)
  • apps/smtp-server/src/server.ts

Comment on lines +43 to +45
const value = line.slice(colonIndex + 1).trim();
if (value.length > 0) {
result[canonicalHeaderName(key)] = value;

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -euo pipefail

printf '\n== server.ts outline ==\n'
ast-grep outline apps/smtp-server/src/server.ts --view expanded || true

printf '\n== searches ==\n'
rg -n "canonicalHeaderName|headerLines|mailparser|List-Unsubscribe|unfold|CRLF|replace\\(/\\\\r\\?\\\\n" apps/smtp-server/src -S || true

printf '\n== targeted snippet ==\n'
sed -n '1,140p' apps/smtp-server/src/server.ts

Repository: usesend/useSend

Length of output: 5972


🏁 Script executed:

#!/bin/bash
set -euo pipefail

printf '\n== package refs ==\n'
rg -n '"mailparser"|simpleParser|headerLines' package.json pnpm-lock.yaml package-lock.json yarn.lock apps -S || true

printf '\n== test files under smtp-server ==\n'
fd -t f '.*(test|spec)\.(ts|tsx|js|jsx)$' apps/smtp-server . || true

printf '\n== smtp-server tree ==\n'
fd -t f apps/smtp-server || true

Repository: usesend/useSend

Length of output: 3154


🌐 Web query:

mailparser simpleParser headerLines folded header values line property raw headers unfolding RFC 5322

💡 Result:

In the context of the mailparser library, the headerLines property and RFC 5322 specifications relate to how raw email headers are handled, specifically regarding folding and unfolding [1][2]. RFC 5322 defines header folding as the process of splitting a long header field body into multiple lines by inserting a CRLF (Carriage Return + Line Feed) followed by WSP (white space) [2]. Conversely, unfolding is the process of reversing this to reconstruct the original, single-line header field by removing the CRLF and subsequent white space [2][3]. This ensures that headers can be evaluated without length restrictions [2]. In mailparser, the headerLines event provides access to these raw header segments [1]. When you listen for this event, the library emits an array of objects, where each object contains two properties [1]: - key: The lowercase header name [1]. - line: The raw content of that header line as it appears in the message, including any folding [1]. While mailparser processes these raw lines into a structured Map of decoded header values for easier use (e.g., parsing dates, email addresses, or list headers), the headerLines event exposes the unparsed, potentially folded lines for developers who need to access the raw data exactly as it was received [1]. Top results: [1][2][4]

Citations:


🌐 Web query:

site:github.com nodemailer/mailparser headerLines line folded header raw as it appears in the message

💡 Result:

In nodemailer/mailparser, the headerLines property is an array of objects that provides access to the header information, where each object contains the key and the line (the raw, folded header string as it appeared in the original message) [1][2]. When parsing an email, mailparser stores individual header lines in this array [2]. Because the library performs parsing and transformations on headers, headerLines serves as a reliable way to access the original, verbatim header lines, including their folding (multi-line) format [1][2]. If you are using the MailParser class directly, you can access headerLines after the parsing process has begun or finished [1]. As of recent updates, you can also listen for the headerLines event to gain access to these raw lines during the parsing stream [1]. Example usage: const { simpleParser } = require('mailparser'); simpleParser(source).then(parsed => { // Access the raw folded lines via parsed.headerLines parsed.headerLines.forEach(header => { console.log(Key: ${header.key}); console.log(Raw Line: ${header.line}); }); }); Note that mailparser is designed to be a streaming parser [3], and while it provides structured header objects (which are often Maps), the headerLines array is the standard mechanism provided by the library to retrieve the exact, raw, folded header content for further analysis or manual reconstruction [1][2].

Citations:


Unfold folded List-Unsubscribe header values before forwarding. headerLines preserves raw folding, so .trim() leaves embedded CRLF + WSP intact. Unfold the value before storing it; otherwise valid unsubscribe headers can be forwarded malformed.

🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@apps/smtp-server/src/server.ts` around lines 43 - 45, Update the header
parsing flow around canonicalHeaderName to unfold folded header values before
storing them in result: replace embedded CRLF followed by whitespace with a
single space, then trim the normalized value and preserve the existing non-empty
check and forwarding behavior.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant