time/parsing/combinator/rfc/
rfc2822.rs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
//! Rules defined in [RFC 2822].
//!
//! [RFC 2822]: https://datatracker.ietf.org/doc/html/rfc2822

use crate::parsing::combinator::rfc::rfc2234::wsp;
use crate::parsing::combinator::{ascii_char, one_or_more, zero_or_more};
use crate::parsing::ParsedItem;

/// Consume the `fws` rule.
// The full rule is equivalent to /\r\n[ \t]+|[ \t]+(?:\r\n[ \t]+)*/
pub(crate) fn fws(mut input: &[u8]) -> Option<ParsedItem<'_, ()>> {
    if let [b'\r', b'\n', rest @ ..] = input {
        one_or_more(wsp)(rest)
    } else {
        input = one_or_more(wsp)(input)?.into_inner();
        while let [b'\r', b'\n', rest @ ..] = input {
            input = one_or_more(wsp)(rest)?.into_inner();
        }
        Some(ParsedItem(input, ()))
    }
}

/// Consume the `cfws` rule.
// The full rule is equivalent to any combination of `fws` and `comment` so long as it is not empty.
pub(crate) fn cfws(input: &[u8]) -> Option<ParsedItem<'_, ()>> {
    one_or_more(|input| fws(input).or_else(|| comment(input)))(input)
}

/// Consume the `comment` rule.
fn comment(mut input: &[u8]) -> Option<ParsedItem<'_, ()>> {
    input = ascii_char::<b'('>(input)?.into_inner();
    input = zero_or_more(fws)(input).into_inner();
    while let Some(rest) = ccontent(input) {
        input = rest.into_inner();
        input = zero_or_more(fws)(input).into_inner();
    }
    input = ascii_char::<b')'>(input)?.into_inner();

    Some(ParsedItem(input, ()))
}

/// Consume the `ccontent` rule.
fn ccontent(input: &[u8]) -> Option<ParsedItem<'_, ()>> {
    ctext(input)
        .or_else(|| quoted_pair(input))
        .or_else(|| comment(input))
}

/// Consume the `ctext` rule.
#[allow(clippy::unnecessary_lazy_evaluations)] // rust-lang/rust-clippy#8522
fn ctext(input: &[u8]) -> Option<ParsedItem<'_, ()>> {
    no_ws_ctl(input).or_else(|| match input {
        [33..=39 | 42..=91 | 93..=126, rest @ ..] => Some(ParsedItem(rest, ())),
        _ => None,
    })
}

/// Consume the `quoted_pair` rule.
fn quoted_pair(mut input: &[u8]) -> Option<ParsedItem<'_, ()>> {
    input = ascii_char::<b'\\'>(input)?.into_inner();
    input = text(input).into_inner();

    // If nothing is parsed by `text`, this means by hit the `obs-text` rule and nothing matched.
    // This is technically a success, and we used to check the `obs-qp` rule to ensure everything
    // possible was consumed. After further analysis, it was determined that this check was
    // unnecessary due to `obs-text` wholly subsuming `obs-qp` in this context. For this reason, if
    // `text` fails to parse anything, we consider it a success without further consideration.

    Some(ParsedItem(input, ()))
}

/// Consume the `no_ws_ctl` rule.
const fn no_ws_ctl(input: &[u8]) -> Option<ParsedItem<'_, ()>> {
    match input {
        [1..=8 | 11..=12 | 14..=31 | 127, rest @ ..] => Some(ParsedItem(rest, ())),
        _ => None,
    }
}

/// Consume the `text` rule.
fn text<'a>(input: &'a [u8]) -> ParsedItem<'a, ()> {
    let new_text = |input: &'a [u8]| match input {
        [1..=9 | 11..=12 | 14..=127, rest @ ..] => Some(ParsedItem(rest, ())),
        _ => None,
    };

    let obs_char = |input: &'a [u8]| match input {
        // This is technically allowed, but consuming this would mean the rest of the string is
        // eagerly consumed without consideration for where the comment actually ends.
        [b')', ..] => None,
        [0..=9 | 11..=12 | 14..=127, rest @ ..] => Some(rest),
        _ => None,
    };

    let obs_text = |mut input| {
        input = zero_or_more(ascii_char::<b'\n'>)(input).into_inner();
        input = zero_or_more(ascii_char::<b'\r'>)(input).into_inner();
        while let Some(rest) = obs_char(input) {
            input = rest;
            input = zero_or_more(ascii_char::<b'\n'>)(input).into_inner();
            input = zero_or_more(ascii_char::<b'\r'>)(input).into_inner();
        }

        ParsedItem(input, ())
    };

    new_text(input).unwrap_or_else(|| obs_text(input))
}