andrewpietiladotcom/lib/ReferencesMimeHeader.js

import MimeHeader from "./MimeHeader.js";

const atext = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789!#$%&'*+-/=?^_`{|}~".split('');
const dtext = "!\"#$%^&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ^_`abcdefghijklmnopqrstuvwxyz{|}~".split('');

class ReferencesMimeHeader extends MimeHeader {
    constructor(key, value) {
        super(key, value);
        this._referencesValue = [];
    }
    // Message-ID: <0Q27kpOFnz1MK3ASM_vbjI8L_BYN3j5OSxIaGxerST74zBn7VZfqCowbhEWgJ0Yv4mV847u25YGjENqlsM7-15Zdus90qN7t_kj55FyuN_c=@protonmail.com>
    get referencesValues() {
        // ABNF:
        //    references      =   "References:" 1*msg-id CRLF
        //    msg-id          =   [CFWS] "<" id-left "@" id-right ">" [CFWS]
        //    id-left         =   dot-atom-text / obs-id-left
        //    obs-id-left     =   local-part
        //    id-right        =   dot-atom-text / no-fold-literal / obs-id-right
        //    obs-id-right    =   domain
        //    no-fold-literal =   "[" *dtext "]"
        //    dtext           =   %d33-90 /          ; Printable US-ASCII
        //                        %d94-126 /         ;  characters not including
        //                        obs-dtext          ;  "[", "]", or "\"
        //    dot-atom-text   =   1*atext *("." 1*atext)
        //    atext           =   ALPHA / DIGIT /    ; Printable US-ASCII
        //                        "!" / "#" /        ;  characters not including
        //                        "$" / "%" /        ;  specials.  Used for atoms.
        //                        "&" / "'" /
        //                        "*" / "+" /
        //                        "-" / "/" /
        //                        "=" / "?" /
        //                        "^" / "_" /
        //                        "`" / "{" /
        //                        "|" / "}" /
        //                        "~"
        //    dtext           =   %d33-90 /          ; Printable US-ASCII
        //                        %d94-126 /         ;  characters not including
        //                        obs-dtext          ;  "[", "]", or "\"
        //    comment         =   "(" *([FWS] ccontent) [FWS] ")"
        //    ccontent        =   ctext / quoted-pair / comment
        //    WSP             =   SP / HTAB          ; white space
        //    obs-FWS         =   1*WSP *(CRLF 1*WSP)
        //    quoted-pair     =   ("\" (VCHAR / WSP)) / obs-qp
        //    obs-qp          =   "\" (%d0 / obs-NO-WS-CTL / LF / CR)
        //    obs-NO-WS-CTL   =   %d1-8 /            ; US-ASCII control
        //                        %d11 /             ;  characters that do not
        //                        %d12 /             ;  include the carriage
        //                        %d14-31 /          ;  return, line feed, and
        //                        %d127              ;  white space characters
        // TODO: Implement utf-8 encoding and whatever other encodings we have to support.
        if ( this._referencesValue.length !== 0 ) {
            return this._referencesValue;
        }
        // TODO: Implement this state machine properly according to the above ABNF.
        var references = [];
        var state = "";
        var inCRLF = false;
        var commentDepth = 0;
        var inDquot = false;
        var idLeft = "";
        var idRight = "";
        var idRightDatEmittedRightSquacket = false;
        for (var char of this.rawValue) {
            if ( inCRLF === true && char !== "\n" ) {
                return null; // Error state.
            }
            if (state === "") {
                // CFWS
                if ( char === " ") {
                    continue;
                } else if ( char === "\t") {
                    continue;
                } else if ( char === "(" && !inDquot ) {
                    commentDepth++;
                } else if ( char === "\"" ) {
                    inDquot = !inDquot;
                } else if ( char === ")" && !inDquot ) {
                    commentDepth--;
                } else if ( char === "\r" ) {
                    inCRLF = true;
                } else if ( char === "\n" ) {
                    inCRLF = false;
                } else if ( char === "<" ) {
                    state = "id-left";
                } else {
                    return null; // Error state, we couldn't produce a proper message ID from the input.
                }
            } else if ( state === "id-left" ) {
                if ( atext.includes(char) ) {
                    idLeft += char;
                } else if ( char === "." ) {
                    if ( idLeft.endsWith(".") || idLeft.length === 0 ) {
                        return null; // Error state.
                    }
                    idLeft += char;
                } else if ( char === "@" ) {
                    if ( idLeft.endsWith(".") || idLeft.length === 0) {
                        return null; // Error state.
                    }
                    state = "id-right";
                }
            } else if ( state === "id-right" ) {
                if ( char === "[" ) {
                    if ( idRight.length !== 0 ) {
                        return null; // Error state.
                    }
                    idRight += char;
                    state = "id-right-dat";
                } else if ( char === "." ) {
                    if ( idRight.endsWith(".") || idRight.length === 0 ) {
                        return null; // Error state.
                    }
                    idRight += char;
                } else if ( char === ">" ) {
                    if ( idRight.endsWith(".") || idRight.length === 0) {
                        return null; // Error state.
                    }
                    references.push(`${idLeft}@${idRight}`);
                    idLeft = "";
                    idRight = "";
                    state = "intermediate-or-end-cfws"
                }
            } else if ( state === "id-right-dat" ) {
                if ( char === ">" && idRight.endsWith("]") ) {
                    state = "intermediate-or-end-cfws";
                    continue;
                }
                if ( dtext.includes(char) ) {
                    if ( idRightDatEmittedRightSquacket ) {
                        return null; // Error state.
                    }
                    idRight += char;
                } else if ( char === "]" ) {
                    if ( idRightDatEmittedRightSquacket ) {
                        return null; // Error state.
                    }
                    idRight += char;
                    idRightDatEmittedRightSquacket = true;
                }
            } else if ( state === "intermediate-or-end-cfws" ) {
                // CFWS
                if ( char === " ") {
                    continue;
                } else if ( char === "\t") {
                    continue;
                } else if ( char === "(" && !inDquot ) {
                    commentDepth++;
                } else if ( char === "\"" ) {
                    inDquot = !inDquot;
                } else if ( char === ")" && !inDquot ) {
                    commentDepth--;
                } else if ( char === "\r" ) {
                    inCRLF = true;
                } else if ( char === "\n" ) {
                    inCRLF = false;
                } else if ( char === "<" ) {
                    state = "id-left";
                } else {
                    return null; // Error state.
                }
            }
        }

        return references;
    }
}

export default ReferencesMimeHeader;