eml-codec/src/text/misc_token.rs

225 lines
5.8 KiB
Rust
Raw Permalink Normal View History

2023-07-18 23:25:10 +02:00
use nom::{
branch::alt,
2023-07-20 09:41:10 +02:00
bytes::complete::{tag, take_while1},
2023-07-18 23:25:10 +02:00
character::complete::space0,
2023-07-19 10:41:51 +02:00
combinator::{map, opt},
2023-07-20 09:41:10 +02:00
multi::{many0, many1, separated_list1},
2023-07-23 16:37:47 +02:00
sequence::preceded,
2023-07-18 23:25:10 +02:00
IResult,
};
2023-07-24 12:37:30 +02:00
use std::fmt;
2023-07-18 23:25:10 +02:00
use crate::text::{
ascii,
2023-07-23 16:37:47 +02:00
encoding::{self, encoded_word},
quoted::{quoted_string, QuotedString},
whitespace::{fws, is_obs_no_ws_ctl},
words::{atom, is_vchar, mime_atom},
2023-07-18 23:25:10 +02:00
};
#[derive(Debug, PartialEq, Default)]
2023-07-20 09:41:10 +02:00
pub struct PhraseList<'a>(pub Vec<Phrase<'a>>);
pub fn phrase_list(input: &[u8]) -> IResult<&[u8], PhraseList> {
map(separated_list1(tag(","), phrase), PhraseList)(input)
2023-07-19 22:27:59 +02:00
}
2023-07-18 23:25:10 +02:00
2023-07-24 22:08:13 +02:00
#[derive(Debug, PartialEq, Clone)]
2023-07-21 18:31:56 +02:00
pub enum MIMEWord<'a> {
Quoted(QuotedString<'a>),
Atom(&'a [u8]),
}
2023-07-22 11:53:26 +02:00
impl Default for MIMEWord<'static> {
fn default() -> Self {
Self::Atom(&[])
}
}
2023-07-21 18:31:56 +02:00
impl<'a> MIMEWord<'a> {
pub fn to_string(&self) -> String {
match self {
2023-07-22 09:47:20 +02:00
Self::Quoted(v) => v.to_string(),
2023-07-23 16:37:47 +02:00
Self::Atom(v) => encoding_rs::UTF_8
.decode_without_bom_handling(v)
.0
.to_string(),
2023-07-21 18:31:56 +02:00
}
}
}
pub fn mime_word(input: &[u8]) -> IResult<&[u8], MIMEWord> {
alt((
2023-07-23 16:37:47 +02:00
map(quoted_string, MIMEWord::Quoted),
2023-07-21 18:31:56 +02:00
map(mime_atom, MIMEWord::Atom),
))(input)
}
2023-07-23 18:35:13 +02:00
#[derive(PartialEq)]
2023-07-18 23:25:10 +02:00
pub enum Word<'a> {
2023-07-19 15:28:17 +02:00
Quoted(QuotedString<'a>),
2023-07-18 23:25:10 +02:00
Encoded(encoding::EncodedWord<'a>),
Atom(&'a [u8]),
}
2023-07-19 12:09:23 +02:00
2023-07-23 17:14:16 +02:00
impl<'a> ToString for Word<'a> {
fn to_string(&self) -> String {
2023-07-18 23:25:10 +02:00
match self {
Word::Quoted(v) => v.to_string(),
Word::Encoded(v) => v.to_string(),
2023-07-23 16:37:47 +02:00
Word::Atom(v) => encoding_rs::UTF_8
.decode_without_bom_handling(v)
.0
.to_string(),
2023-07-18 23:25:10 +02:00
}
}
}
2023-07-23 18:35:13 +02:00
impl<'a> fmt::Debug for Word<'a> {
2023-07-23 17:14:16 +02:00
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
2023-07-24 12:37:30 +02:00
fmt.debug_tuple("Word")
.field(&format_args!("\"{}\"", self.to_string()))
.finish()
2023-07-23 17:14:16 +02:00
}
2023-07-23 18:35:13 +02:00
}
2023-07-18 23:25:10 +02:00
/// Word
///
/// ```abnf
/// word = atom / quoted-string
/// ```
pub fn word(input: &[u8]) -> IResult<&[u8], Word> {
alt((
2023-07-24 09:24:38 +02:00
map(quoted_string, Word::Quoted),
map(encoded_word, Word::Encoded),
map(atom, Word::Atom),
2023-07-18 23:25:10 +02:00
))(input)
}
2023-07-23 18:35:13 +02:00
#[derive(PartialEq)]
2023-07-18 23:25:10 +02:00
pub struct Phrase<'a>(pub Vec<Word<'a>>);
2023-07-19 12:09:23 +02:00
2023-07-23 17:14:16 +02:00
impl<'a> ToString for Phrase<'a> {
fn to_string(&self) -> String {
2023-07-23 16:37:47 +02:00
self.0
.iter()
.map(|v| v.to_string())
.collect::<Vec<String>>()
.join(" ")
2023-07-18 23:25:10 +02:00
}
}
2023-07-23 18:35:13 +02:00
impl<'a> fmt::Debug for Phrase<'a> {
2023-07-23 17:14:16 +02:00
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
2023-07-24 12:37:30 +02:00
fmt.debug_tuple("Phrase")
.field(&format_args!("\"{}\"", self.to_string()))
.finish()
2023-07-23 17:14:16 +02:00
}
2023-07-23 18:35:13 +02:00
}
2023-07-18 23:25:10 +02:00
/// Phrase
///
/// ```abnf
/// phrase = 1*word / obs-phrase
/// ```
pub fn phrase(input: &[u8]) -> IResult<&[u8], Phrase> {
2023-07-24 09:24:38 +02:00
let (input, phrase) = map(many1(word), Phrase)(input)?;
2023-07-18 23:25:10 +02:00
Ok((input, phrase))
}
/// Compatible unstructured input
///
/// ```abnf
/// obs-utext = %d0 / obs-NO-WS-CTL / VCHAR
/// ```
fn is_unstructured(c: u8) -> bool {
is_vchar(c) || is_obs_no_ws_ctl(c) || c == ascii::NULL
}
2023-07-23 12:24:46 +02:00
#[derive(Debug, PartialEq, Clone)]
2023-07-19 10:41:51 +02:00
pub enum UnstrToken<'a> {
2023-07-18 23:25:10 +02:00
Init,
Encoded(encoding::EncodedWord<'a>),
Plain(&'a [u8]),
}
2023-07-19 12:09:23 +02:00
2023-07-23 17:14:16 +02:00
impl<'a> ToString for UnstrToken<'a> {
fn to_string(&self) -> String {
2023-07-18 23:25:10 +02:00
match self {
UnstrToken::Init => "".into(),
UnstrToken::Encoded(e) => e.to_string(),
2023-07-23 16:37:47 +02:00
UnstrToken::Plain(e) => encoding_rs::UTF_8
.decode_without_bom_handling(e)
.0
.into_owned(),
2023-07-18 23:25:10 +02:00
}
}
}
2023-07-23 17:14:16 +02:00
#[derive(PartialEq, Clone)]
2023-07-18 23:25:10 +02:00
pub struct Unstructured<'a>(pub Vec<UnstrToken<'a>>);
2023-07-19 12:09:23 +02:00
2023-07-23 17:14:16 +02:00
impl<'a> ToString for Unstructured<'a> {
fn to_string(&self) -> String {
2023-07-23 16:37:47 +02:00
self.0
.iter()
.fold(
(&UnstrToken::Init, String::new()),
|(prev_token, mut result), current_token| {
match (prev_token, current_token) {
(UnstrToken::Init, v) => result.push_str(v.to_string().as_ref()),
(UnstrToken::Encoded(_), UnstrToken::Encoded(v)) => {
result.push_str(v.to_string().as_ref())
}
(_, v) => {
result.push(' ');
result.push_str(v.to_string().as_ref())
}
};
(current_token, result)
},
)
.1
2023-07-18 23:25:10 +02:00
}
}
2023-07-23 17:14:16 +02:00
impl<'a> fmt::Debug for Unstructured<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
2023-07-24 12:37:30 +02:00
fmt.debug_tuple("Unstructured")
.field(&format_args!("\"{}\"", self.to_string()))
.finish()
2023-07-23 17:14:16 +02:00
}
}
2023-07-18 23:25:10 +02:00
/// Unstructured header field body
///
/// ```abnf
/// unstructured = (*([FWS] VCHAR_SEQ) *WSP) / obs-unstruct
/// ```
pub fn unstructured(input: &[u8]) -> IResult<&[u8], Unstructured> {
2023-07-23 16:37:47 +02:00
let (input, r) = many0(preceded(
opt(fws),
alt((
2023-07-24 09:24:38 +02:00
map(encoded_word, UnstrToken::Encoded),
map(take_while1(is_unstructured), UnstrToken::Plain),
2023-07-23 16:37:47 +02:00
)),
))(input)?;
2023-07-18 23:25:10 +02:00
let (input, _) = space0(input)?;
Ok((input, Unstructured(r)))
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_phrase() {
assert_eq!(
2023-07-23 16:37:47 +02:00
phrase(b"hello world").unwrap().1.to_string(),
2023-07-19 11:03:40 +02:00
"hello world".to_string(),
2023-07-18 23:25:10 +02:00
);
assert_eq!(
2023-07-19 11:03:40 +02:00
phrase(b"salut \"le\" monde").unwrap().1.to_string(),
"salut le monde".to_string(),
2023-07-18 23:25:10 +02:00
);
2023-07-19 11:03:40 +02:00
let (rest, parsed) = phrase(b"fin\r\n du\r\nmonde").unwrap();
assert_eq!(rest, &b"\r\nmonde"[..]);
assert_eq!(parsed.to_string(), "fin du".to_string());
2023-07-18 23:25:10 +02:00
}
}