eml-codec

⚠️ This is currently only a decoder (ie. a parser), encoding is not yet implemented.

eml-codec is a child project of Aerogramme, a distributed and encrypted IMAP server developped by the non-profit organization Deuxfleurs. Its aim is to be a swiss army knife to handle emails, whether it is to build an IMAP/JMAP server, a mail filter (like an antispam), or a mail client.

Example

let input = br#"Date: 7 Mar 2023 08:00:00 +0200
From: deuxfleurs@example.com
To: someone_else@example.com
Subject: An RFC 822 formatted message
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii

This is the plain text body of the message. Note the blank line
between the header information and the body of the message."#;

let (_, email) = eml_codec::parse_message(input).unwrap();
println!(
    "{} just sent you an email with subject \"{}\"",
    email.imf.from[0].to_string(),
    email.imf.subject.unwrap().to_string(),
);

See more examples in the examples/ folder

About the name

This library does not aim at implementing a specific RFC, but to be a swiss-army knife to decode and encode ("codec") what is generaly considered an email (generally abbreviated "eml"), hence the name: eml-codec.

Goals

• Maintainability - modifying the code does not create regression and is possible for someone exterior to the project.
• Compatibility - always try to parse something, do not panic or return an error.
• Exhaustivity - serve as a common project to encode knowledge about emails (existing mime types, existing headers, etc.).
• Type safe - do not manipulate only strings/bytes but leverage Rust type system instead so you benefit from its safety checks at compile time.

See more about this library goals in the doc/ folder

Missing / known bugs

Current known limitations/bugs:

• Part transfer-decoding is not implemented yet
• Internationalized headers (UTF-8) is not implemented yet
• Resent Header Fields are not implemented
• Return-Path/Received headers might be hard to use as their order is important, and it's currently lost in the final datastructure.
• Datetime parsing of invalid date might return None instead of falling back to the bad_body field
• Comments contained in the email headers are dropped during parsing
• No support is provided for message/external-body (read data from local computer) and message/partial (aggregate multiple fragmented emails) as they seem obsolete and dangerous to implement.

Design

High-level overview of the datastructures (inspired by the UML class diagram conventions):

Testing strategy

Currently this crate has some unit tests on most of its parsing functions. It is also tested as part of Aerogramme, its parent project where it handles email parsing. In this project, eml-codec parsing capabilities are compared to Dovecot, Cyrus, Maddy and other IMAP servers.

It is planned to test it on large email datasets (like Enron, jpbush, mailing lists, etc.) but it's not done yet. Fuzzing the library would also be interesting, probably to detect crashing due to stack overflow for example due to the infinite recursivity of MIME.

RFC and IANA references

RFC

IANA

State of the art / alternatives

The following review is not an objective, neutral, impartial review. Instead, it's a temptative to explain why I wrote this library. If you find something outdated or objectively wrong, feel free to open a PR or an issue to fix it. In no case, I think eml-codec is superior, it's just another approach to the problem, and I see it as another stone to the edifice.

mail_parser, mailparse and rust-email are 3 handwritten parsers. Such handwritten parsers do not encourage separation of concerns: mail_parser and mailparse feature large functions with hundreds of lines with a high cylomatic complexity. Due to this complex logic, I have failed to debug/patch such code in the past. rust-email code is easier to read but its mime part implementation is marked as unstable. mail_parser is used in the IMAP/JMAP/SMTP server project stalwartlabs/mail-server and rust-email is used in the email-based chat application Deltachat (however rust-email MIME parsed is not used, a custom MIME parser is reimplemented in Delta Chat instead). It must be noted that mail_parser supports a large amount of extensions (UTF-8 headers, UTF-7 encoding, continuation, many custom fields, etc.) and would better cope with malformed emails than other libraries. A goal of eml_codec is to be open to contribution and maintainable over time, which is made possible trough the parser combinator pattern that encourages writing small, reusable, independently testable functions.

rustyknife uses the same design pattern (parser combinator) and the same library (nom) as eml_codec. However, rustyknife is more targeted to SMTP servers (MTA) than IMAP (MDA) and email clients (MUA). It thus only supports parsing headers and not emails' body. Also, an acquaintance warned me that this library is a bit slow, it might be due to the fact that the library does some processing while parsing the email (like rebuilding and allocating strings). If it happens that this part is not used later, the allocation/processing has been wasted. A goal of eml_codec is to produce an AST of the email with as few processing as possible, so that the parsing remains efficient, and then the allocation/processing is made lazily, on demand, when the corresponding function is called. It is often referred as zero-copy.

Support

eml-codec, as part of the Aerogramme project, was funded through the NGI Assure Fund, a fund established by NLnet with financial support from the European Commission's Next Generation Internet programme, under the aegis of DG Communications Networks, Content and Technology under grant agreement No 957073.

License

eml-codec Copyright (C) The eml-codec Contributors

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.