---
title: "Convert Apple Mail EMLX to EML in C#"
url: https://recursiveai.org/onemail/cookbook/convert-emlx-to-eml/
description: "Extract the embedded email bytes from an Apple Mail EMLX file in C#. Write an EML without the byte-count line or Apple Mail plist."
version: "1.0.1"
---

# Convert Apple Mail EMLX to EML in C#

**OneMail** is a free, open-source email library for .NET from Recursive AI. MIT or Apache-2.0, no native dependencies.

[NuGet](https://www.nuget.org/packages/RecursiveAI.OneMail) · [GitHub](https://github.com/recursive-ai-oss/onemail-csharp) · [Docs](https://recursiveai.org/onemail/docs/) · [Download example (.zip)](https://recursiveai.org/onemail/cookbook/convert-emlx-to-eml/convert-emlx-to-eml.zip)

Extract the email from `note.emlx` in the zip. Attachments Apple Mail saved as separate files aren’t included.

```shell
dotnet add package RecursiveAI.OneMail
```

Requires [.NET 10](https://dotnet.microsoft.com/download/dotnet/10.0) (net10.0); .NET 9 and earlier aren’t supported.

```csharp
using RecursiveAI.OneMail.Sources;
using RecursiveAI.OneMail.Stores;
using RecursiveAI.OneMail.Stores.Emlx;
using var src = FileSource.Open("note.emlx");
var part = EmlxReader.Read(src);
if (part.Defects != StoreDefects.None)
    throw new InvalidDataException(
        "Bad EMLX.");
using var eml =
    EmlxReader.OpenMessage(src, part);
using var dst = File.Create("note.eml");
eml.CopyTo(dst);
```

Unzip [the example](https://recursiveai.org/onemail/cookbook/convert-emlx-to-eml/convert-emlx-to-eml.zip), then `dotnet run`.

Writes or replaces `note.eml` with the embedded message only.

## The full program

The fictional `note.emlx` contains the following text. The count line ends with LF; the message lines use CRLF. The `207` counts only the message bytes, including the two UTF-8 bytes for `é`, not the count line or the XML property list.

```text
207
From: Ada <ada@example.test>
Date: Thu, 19 Nov 2026 09:00:00 +0000
Subject: Station cafe
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit

Meet at the café.
<?xml version="1.0" encoding="UTF-8"?>
<plist version="1.0"><dict><key>flags</key><integer>1</integer></dict></plist>
```

The program copies the message bytes as they are, then parses the saved EML to check the subject and compares the output with the expected UTF-8 bytes.

```csharp
using System.Text;
using RecursiveAI.OneMail;
using RecursiveAI.OneMail.Sources;
using RecursiveAI.OneMail.Stores;
using RecursiveAI.OneMail.Stores.Emlx;

using var file = FileSource.Open("note.emlx");
var entry = EmlxReader.Read(file);
if (entry.Defects != StoreDefects.None)
    throw new InvalidDataException($"Cannot extract message: {entry.Defects}");

using (var message = EmlxReader.OpenMessage(file, entry))
using (var output = File.Create("note.eml"))
    message.CopyTo(output);

var eml = File.ReadAllBytes("note.eml");
var subject = Mail.Parse(eml).Subject;
const string expected = "From: Ada <ada@example.test>\r\n" +
    "Date: Thu, 19 Nov 2026 09:00:00 +0000\r\n" +
    "Subject: Station cafe\r\nMIME-Version: 1.0\r\n" +
    "Content-Type: text/plain; charset=utf-8\r\n" +
    "Content-Transfer-Encoding: 8bit\r\n\r\nMeet at the café.\r\n";
if (!eml.AsSpan().SequenceEqual(Encoding.UTF8.GetBytes(expected)) || subject != "Station cafe")
    throw new InvalidDataException("The extracted message differs from the input message.");

Console.WriteLine($"Subject: {subject}");
Console.WriteLine($"Wrote note.eml: {eml.Length} bytes.");
```

`dotnet run` prints:

```text
Subject: Station cafe
Wrote note.eml: 207 bytes.
```

If the count line is invalid or the message is truncated, the reader marks a defect. The example rejects it before creating the EML.

## Variations

**Read the Apple Mail flags without putting them in the EML.** Add these lines after checking `entry.Defects` in the full program.

```csharp
// After reading entry, with file still open:
var metadata = EmlxReader.ReadMetadata(file, entry);
Console.WriteLine($"Apple Mail flags: {metadata.Flags}");
```

```text
Apple Mail flags: 1
```

## Related recipes

[Parse an EML file](https://recursiveai.org/onemail/cookbook/parse-eml/) · [Extract attachments from email](https://recursiveai.org/onemail/cookbook/extract-attachments/)

---

## The short answer

Read the file with `EmlxReader.Read`, check the returned `Defects`, and copy the stream from `EmlxReader.OpenMessage` into an `.eml` file. That stream includes the embedded email’s headers and body but not Apple Mail’s metadata. Keep the `.emlx` source open until the copy finishes.

## What is EMLX format?

An Apple Mail EMLX file has three parts: a decimal byte-count line, the stated number of message bytes, and usually an XML property list (plist). The plist may record a flag word, dates, and other state that isn’t stored as part of the email. The [Library of Congress format description](https://www.loc.gov/preservation/digital/formats/fdd/fdd000615.shtml) describes the container; [OneMail’s EMLX docs](https://recursiveai.org/onemail/docs/mailboxes/) explain the ranges it returns.

An EML, by contrast, contains the message without those outer parts. The sample’s `note.eml` starts with `From:`, not `207`, and ends with the body, not the plist. Renaming the original file wouldn’t remove either extra part.

## How do I convert an Apple Mail email to EML without changing its bytes?

Use the count in the file, not a character count or an XML search. The sample’s `café` demonstrates why those differ: the `é` takes two bytes in UTF-8. Copying the byte range also means you don’t have to decode and re-encode MIME bodies to get the EML. The full program compares all 207 output bytes and checks the subject separately; a subject can parse correctly even if a bad conversion also includes an unwanted trailer.

`EmlxReader.Read` reports a bad count line as `BadByteCount`, with no message range. If the count is larger than the remaining file, it reports `TruncatedMessage` and cuts the range at the file end. Neither case is a clean conversion. The check in the snippet at the top prevents an empty or cut-off export from being written as though it were complete.

A `.partial.emlx` may refer to attachments Apple Mail stored separately. The conversion copies only the message bytes present in the input; it does not look up or reattach those files. Keep the original if you need the Apple Mail metadata too. The flags variation reads that metadata separately while leaving it out of the EML.
