rfc2047-decoder 1.1.2

use charset::Charset;
use std::{convert::TryFrom, result};

use crate::lexer::{encoded_word, Token, Tokens};

/// All errors which the parser can throw.
#[derive(thiserror::Error, Debug, Clone, PartialEq)]
pub enum Error {
    #[error("cannot parse encoding: encoding is bigger than a char")]
    ParseEncodingTooBigError,
    #[error("cannot parse encoding: encoding is empty")]
    ParseEncodingEmptyError,
    #[error("cannot parse encoding {0}: B or Q is expected")]
    ParseEncodingError(char),
}

type Result<T> = result::Result<T, Error>;

pub type ClearText = Vec<u8>;
pub type ParsedEncodedWords = Vec<ParsedEncodedWord>;

#[derive(Debug, Clone, PartialEq, Eq, Hash)]
pub enum Encoding {
    B,
    Q,
}

impl Encoding {
    pub const B_CHAR: char = 'b';
    pub const Q_CHAR: char = 'q';
    pub const MAX_LENGTH: usize = 1;
}

impl TryFrom<Vec<u8>> for Encoding {
    type Error = Error;

    fn try_from(token: Vec<u8>) -> Result<Self> {
        if token.len() > Self::MAX_LENGTH {
            return Err(Error::ParseEncodingTooBigError);
        }

        let encoding = token.first().ok_or(Error::ParseEncodingEmptyError)?;
        let encoding = *encoding as char;

        match encoding.to_ascii_lowercase() {
            Encoding::Q_CHAR => Ok(Self::Q),
            Encoding::B_CHAR => Ok(Self::B),
            _ => Err(Error::ParseEncodingError(encoding)),
        }
    }
}

#[derive(Debug, Clone, PartialEq, Hash)]
pub enum ParsedEncodedWord {
    ClearText(ClearText),
    EncodedWord {
        charset: Option<Charset>,
        encoding: Encoding,
        encoded_text: Vec<u8>,
    },
}

impl ParsedEncodedWord {
    pub fn convert_encoded_word(encoded_word: encoded_word::EncodedWord) -> Result<Self> {
        let encoding = Encoding::try_from(encoded_word.encoding)?;
        let charset = Charset::for_label(&encoded_word.charset);

        Ok(Self::EncodedWord {
            charset,
            encoding,
            encoded_text: encoded_word.encoded_text,
        })
    }
}

pub fn run(tokens: Tokens) -> Result<ParsedEncodedWords> {
    let parsed_encoded_words = convert_tokens_to_encoded_words(tokens)?;
    Ok(parsed_encoded_words)
}

fn convert_tokens_to_encoded_words(tokens: Tokens) -> Result<ParsedEncodedWords> {
    tokens
        .into_iter()
        .map(|token: Token| match token {
            Token::ClearText(clear_text) => Ok(ParsedEncodedWord::ClearText(clear_text)),
            Token::EncodedWord(encoded_word) => {
                ParsedEncodedWord::convert_encoded_word(encoded_word)
            }
        })
        .collect()
}

#[cfg(test)]
mod tests {
    use charset::Charset;

    use crate::{
        lexer,
        parser::{self, Encoding, ParsedEncodedWord},
        RecoverStrategy,
    };

    /// Example taken from:
    /// https://datatracker.ietf.org/doc/html/rfc2047#section-8
    ///
    /// `From` field
    #[test]
    fn test_parse1() {
        let message = "=?US-ASCII?Q?Keith_Moore?=".as_bytes();
        let tokens = lexer::run(&message, RecoverStrategy::Abort).unwrap();
        let parsed = parser::run(tokens).unwrap();

        let expected = vec![ParsedEncodedWord::EncodedWord {
            charset: Charset::for_label("US-ASCII".as_bytes()),
            encoding: Encoding::Q,
            encoded_text: "Keith_Moore".as_bytes().to_vec(),
        }];

        assert_eq!(parsed, expected);
    }

    /// Example taken from:
    /// https://datatracker.ietf.org/doc/html/rfc2047#section-8
    ///
    /// `To` field
    #[test]
    fn test_parse2() {
        let message = "=?ISO-8859-1?Q?Keld_J=F8rn_Simonsen?=".as_bytes();
        let tokens = lexer::run(&message, RecoverStrategy::Abort).unwrap();
        let parsed = parser::run(tokens).unwrap();

        let expected = vec![ParsedEncodedWord::EncodedWord {
            charset: Charset::for_label("ISO-8859-1".as_bytes()),
            encoding: Encoding::Q,
            encoded_text: "Keld_J=F8rn_Simonsen".as_bytes().to_vec(),
        }];

        assert_eq!(parsed, expected);
    }

    /// Example taken from:
    /// https://datatracker.ietf.org/doc/html/rfc2047#section-8
    ///
    /// `CC` field
    #[test]
    fn test_parse3() {
        let message = "=?ISO-8859-1?Q?Andr=E9?=".as_bytes();
        let tokens = lexer::run(&message, RecoverStrategy::Abort).unwrap();
        let parsed = parser::run(tokens).unwrap();

        let expected = vec![ParsedEncodedWord::EncodedWord {
            charset: Charset::for_label("ISO-8859-1".as_bytes()),
            encoding: Encoding::Q,
            encoded_text: "Andr=E9".as_bytes().to_vec(),
        }];

        assert_eq!(parsed, expected);
    }

    /// Example taken from:
    /// https://datatracker.ietf.org/doc/html/rfc2047#section-8
    ///
    /// `Subject` field
    #[test]
    fn test_parse4() {
        let message = "=?ISO-8859-1?B?SWYgeW91IGNhbiByZWFkIHRoaXMgeW8=?=".as_bytes();
        let tokens = lexer::run(&message, RecoverStrategy::Abort).unwrap();
        let parsed = parser::run(tokens).unwrap();

        let expected = vec![ParsedEncodedWord::EncodedWord {
            charset: Charset::for_label("ISO-8859-1".as_bytes()),
            encoding: Encoding::B,
            encoded_text: "SWYgeW91IGNhbiByZWFkIHRoaXMgeW8=".as_bytes().to_vec(),
        }];

        assert_eq!(parsed, expected);
    }
}