Deuxfleurs/eml-codec
15
1
Fork 0
Code Issues 10 Pull requests Projects Releases 2 Packages Wiki Activity

Access headers as key/values #24

Merged
quentin merged 8 commits from headers_map into main 2023-08-30 17:50:26 +00:00
Conversation 0 Commits 8 Files changed 12 +423 -339
12 changed files with 423 additions and 339 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

102
src/header.rs
View file

@ -1,55 +1,62 @@
use crate::text::misc_token::{unstructured, Unstructured};
use crate::text::whitespace::{foldable_line, obs_crlf};
use nom::{ use nom::{
branch::alt, branch::alt,
bytes::complete::{tag, tag_no_case, take_while1}, bytes::complete::{tag, take_while1},
character::complete::space0, character::complete::space0,
combinator::map, combinator::{into, recognize},
multi::{fold_many0}, multi::many0,
sequence::{pair, terminated, tuple}, sequence::{pair, terminated, tuple},
IResult, IResult,
}; };
use std::fmt;
#[derive(Debug, PartialEq)] use crate::text::misc_token::unstructured;
pub enum CompField<'a, T> { use crate::text::whitespace::{foldable_line, obs_crlf};
Known(T),
Unknown(Kv<'a>), #[derive(PartialEq, Clone)]
Bad(&'a [u8]), pub struct Kv2<'a>(pub &'a [u8], pub &'a [u8]);
impl<'a> From<(&'a [u8], &'a [u8])> for Kv2<'a> {
fn from(pair: (&'a [u8], &'a [u8])) -> Self {
Self(pair.0, pair.1)
}
} }
impl<'a> fmt::Debug for Kv2<'a> {
#[derive(Debug, PartialEq, Clone)] fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
pub struct Kv<'a>(pub &'a [u8], pub Unstructured<'a>); fmt.debug_tuple("header::Kv2")
.field(&String::from_utf8_lossy(self.0))
.field(&String::from_utf8_lossy(self.1))
pub fn header<'a, T>( .finish()
fx: impl Fn(&'a [u8]) -> IResult<&'a [u8], T> + Copy,
) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], (Vec::<T>, Vec::<Kv>, Vec<&'a [u8]>)> {
move |input| {
terminated(
fold_many0(
alt((
map(fx, CompField::Known),
map(opt_field, |(k, v)| CompField::Unknown(Kv(k, v))),
map(foldable_line, CompField::Bad),
)),
|| (Vec::<T>::new(), Vec::<Kv>::new(), Vec::<&'a [u8]>::new()),
|(mut known, mut unknown, mut bad), item| {
match item {
CompField::Known(v) => known.push(v),
CompField::Unknown(v) => unknown.push(v),
CompField::Bad(v) => bad.push(v),
};
(known, unknown, bad)
}
),
obs_crlf,
)(input)
} }
} }
pub fn field_name<'a>(name: &'static [u8]) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], &'a [u8]> { #[derive(Debug, PartialEq, Clone)]
move |input| terminated(tag_no_case(name), tuple((space0, tag(b":"), space0)))(input) pub enum Field<'a> {
Good(Kv2<'a>),
Bad(&'a [u8]),
}
impl<'a> From<Kv2<'a>> for Field<'a> {
fn from(kv: Kv2<'a>) -> Self {
Self::Good(kv)
}
}
impl<'a> From<&'a [u8]> for Field<'a> {
fn from(bad: &'a [u8]) -> Self {
Self::Bad(bad)
}
}
/// Parse headers as key/values
pub fn header_kv(input: &[u8]) -> IResult<&[u8], Vec<Field>> {
terminated(
many0(alt((into(correct_field), into(foldable_line)))),
obs_crlf,
)(input)
}
pub fn field_any(input: &[u8]) -> IResult<&[u8], &[u8]> {
terminated(
take_while1(|c| (0x21..=0x7E).contains(&c) && c != 0x3A),
tuple((space0, tag(b":"), space0)),
)(input)
} }
/// Optional field /// Optional field
@ -61,15 +68,6 @@ pub fn field_name<'a>(name: &'static [u8]) -> impl Fn(&'a [u8]) -> IResult<&'a [
/// %d59-126 ; characters not including /// %d59-126 ; characters not including
/// ; ":". /// ; ":".
/// ``` /// ```
pub fn opt_field(input: &[u8]) -> IResult<&[u8], (&[u8], Unstructured)> { pub fn correct_field(input: &[u8]) -> IResult<&[u8], Kv2> {
terminated( terminated(into(pair(field_any, recognize(unstructured))), obs_crlf)(input)
pair(
terminated(
take_while1(|c| (0x21..=0x7E).contains(&c) && c != 0x3A),
tuple((space0, tag(b":"), space0)),
),
unstructured,
),
obs_crlf,
)(input)
} }

127
src/imf/field.rs
View file

@ -1,21 +1,14 @@
use chrono::{DateTime, FixedOffset}; use chrono::{DateTime, FixedOffset};
use nom::{ use nom::combinator::map;
branch::alt,
combinator::map,
sequence::{preceded, terminated},
IResult,
};
use crate::header::{field_name, header}; use crate::header;
use crate::imf::address::{address_list, mailbox_list, nullable_address_list, AddressList}; use crate::imf::address::{address_list, mailbox_list, nullable_address_list, AddressList};
use crate::imf::datetime::section as date; use crate::imf::datetime::section as date;
use crate::imf::identification::{msg_id, msg_list, MessageID, MessageIDList}; use crate::imf::identification::{msg_id, msg_list, MessageID, MessageIDList};
use crate::imf::mailbox::{mailbox, AddrSpec, MailboxList, MailboxRef}; use crate::imf::mailbox::{mailbox, AddrSpec, MailboxList, MailboxRef};
use crate::imf::mime::{version, Version}; use crate::imf::mime::{version, Version};
use crate::imf::trace::{received_log, return_path, ReceivedLog}; use crate::imf::trace::{received_log, return_path, ReceivedLog};
use crate::imf::Imf;
use crate::text::misc_token::{phrase_list, unstructured, PhraseList, Unstructured}; use crate::text::misc_token::{phrase_list, unstructured, PhraseList, Unstructured};
use crate::text::whitespace::obs_crlf;
#[derive(Debug, PartialEq)] #[derive(Debug, PartialEq)]
pub enum Field<'a> { pub enum Field<'a> {
@ -49,94 +42,34 @@ pub enum Field<'a> {
MIMEVersion(Version), MIMEVersion(Version),
} }
impl<'a> TryFrom<&header::Field<'a>> for Field<'a> {
pub fn field(input: &[u8]) -> IResult<&[u8], Field> { type Error = ();
terminated( fn try_from(f: &header::Field<'a>) -> Result<Self, Self::Error> {
alt(( let content = match f {
preceded(field_name(b"date"), map(date, Field::Date)), header::Field::Good(header::Kv2(key, value)) => {
preceded(field_name(b"from"), map(mailbox_list, Field::From)), match key.to_ascii_lowercase().as_slice() {
preceded(field_name(b"sender"), map(mailbox, Field::Sender)), b"date" => map(date, Field::Date)(value),
preceded(field_name(b"reply-to"), map(address_list, Field::ReplyTo)), b"from" => map(mailbox_list, Field::From)(value),
preceded(field_name(b"to"), map(address_list, Field::To)), b"sender" => map(mailbox, Field::Sender)(value),
preceded(field_name(b"cc"), map(address_list, Field::Cc)), b"reply-to" => map(address_list, Field::ReplyTo)(value),
preceded(field_name(b"bcc"), map(nullable_address_list, Field::Bcc)), b"to" => map(address_list, Field::To)(value),
preceded(field_name(b"message-id"), map(msg_id, Field::MessageID)), b"cc" => map(address_list, Field::Cc)(value),
preceded(field_name(b"in-reply-to"), map(msg_list, Field::InReplyTo)), b"bcc" => map(nullable_address_list, Field::Bcc)(value),
preceded(field_name(b"references"), map(msg_list, Field::References)), b"message-id" => map(msg_id, Field::MessageID)(value),
preceded(field_name(b"subject"), map(unstructured, Field::Subject)), b"in-reply-to" => map(msg_list, Field::InReplyTo)(value),
preceded(field_name(b"comments"), map(unstructured, Field::Comments)), b"references" => map(msg_list, Field::References)(value),
preceded(field_name(b"keywords"), map(phrase_list, Field::Keywords)), b"subject" => map(unstructured, Field::Subject)(value),
preceded( b"comments" => map(unstructured, Field::Comments)(value),
field_name(b"return-path"), b"keywords" => map(phrase_list, Field::Keywords)(value),
map(return_path, Field::ReturnPath), b"return-path" => map(return_path, Field::ReturnPath)(value),
), b"received" => map(received_log, Field::Received)(value),
preceded(field_name(b"received"), map(received_log, Field::Received)), b"mime-version" => map(version, Field::MIMEVersion)(value),
preceded( _ => return Err(()),
field_name(b"mime-version"),
map(version, Field::MIMEVersion),
),
)),
obs_crlf,
)(input)
}
pub fn imf(input: &[u8]) -> IResult<&[u8], Imf> {
map(header(field), |(known, unknown, bad)| {
let mut imf = Imf::from_iter(known);
imf.header_ext = unknown;
imf.header_bad = bad;
imf
})(input)
}
#[cfg(test)]
mod tests {
use super::*;
use crate::imf::address::*;
use crate::imf::mailbox::*;
use crate::text::misc_token::*;
use chrono::{FixedOffset, TimeZone};
#[test]
fn test_header() {
let fullmail = b"Date: 7 Mar 2023 08:00:00 +0200
From: someone@example.com
To: someone_else@example.com
Subject: An RFC 822 formatted message
This is the plain text body of the message. Note the blank line
between the header information and the body of the message.";
assert_eq!(
imf(fullmail),
Ok((
&b"This is the plain text body of the message. Note the blank line\nbetween the header information and the body of the message."[..],
Imf {
date: Some(FixedOffset::east_opt(2 * 3600).unwrap().with_ymd_and_hms(2023, 3, 7, 8, 0, 0).unwrap()),
from: vec![MailboxRef {
name: None,
addrspec: AddrSpec {
local_part: LocalPart(vec![LocalPartToken::Word(Word::Atom(&b"someone"[..]))]),
domain: Domain::Atoms(vec![&b"example"[..], &b"com"[..]]),
}
}],
to: vec![AddressRef::Single(MailboxRef {
name: None,
addrspec: AddrSpec {
local_part: LocalPart(vec![LocalPartToken::Word(Word::Atom(&b"someone_else"[..]))]),
domain: Domain::Atoms(vec![&b"example"[..], &b"com"[..]]),
}
})],
subject: Some(Unstructured(vec![
UnstrToken::Plain(&b"An"[..]),
UnstrToken::Plain(&b"RFC"[..]),
UnstrToken::Plain(&b"822"[..]),
UnstrToken::Plain(&b"formatted"[..]),
UnstrToken::Plain(&b"message"[..]),
])),
..Imf::default()
} }
)), }
) _ => return Err(()),
};
content.map(|(_, content)| content).or(Err(()))
} }
} }

81
src/imf/mod.rs
View file

@ -1,5 +1,4 @@
/// Parse and represent IMF (Internet Message Format) headers (RFC822, RFC5322) /// Parse and represent IMF (Internet Message Format) headers (RFC822, RFC5322)
pub mod address; pub mod address;
pub mod datetime; pub mod datetime;
pub mod field; pub mod field;
@ -8,13 +7,15 @@ pub mod mailbox;
pub mod mime; pub mod mime;
pub mod trace; pub mod trace;
use nom::{combinator::map, IResult};
use crate::header;
use crate::imf::address::AddressRef; use crate::imf::address::AddressRef;
use crate::imf::field::Field; use crate::imf::field::Field;
use crate::imf::identification::MessageID; use crate::imf::identification::MessageID;
use crate::imf::mailbox::{AddrSpec, MailboxRef}; use crate::imf::mailbox::{AddrSpec, MailboxRef};
use crate::imf::mime::Version; use crate::imf::mime::Version;
use crate::imf::trace::ReceivedLog; use crate::imf::trace::ReceivedLog;
use crate::header;
use crate::text::misc_token::{PhraseList, Unstructured}; use crate::text::misc_token::{PhraseList, Unstructured};
use chrono::{DateTime, FixedOffset}; use chrono::{DateTime, FixedOffset};
@ -51,17 +52,13 @@ pub struct Imf<'a> {
// MIME // MIME
pub mime_version: Option<Version>, pub mime_version: Option<Version>,
// Junk // Raw fields
pub header_ext: Vec<header::Kv<'a>>, pub kv: Vec<header::Field<'a>>,
pub header_bad: Vec<&'a [u8]>,
} }
impl<'a> Imf<'a> { impl<'a> Imf<'a> {
pub fn with_opt(mut self, opt: Vec<header::Kv<'a>>) -> Self { pub fn with_kv(mut self, v: Vec<header::Field<'a>>) -> Self {
self.header_ext = opt; self self.kv = v;
} self
pub fn with_bad(mut self, bad: Vec<&'a [u8]>) -> Self {
self.header_bad = bad; self
} }
} }
@ -92,3 +89,65 @@ impl<'a> FromIterator<Field<'a>> for Imf<'a> {
}) })
} }
} }
pub fn imf(input: &[u8]) -> IResult<&[u8], Imf> {
map(header::header_kv, |fields| {
fields
.iter()
.flat_map(Field::try_from)
.into_iter()
.collect::<Imf>()
})(input)
}
#[cfg(test)]
mod tests {
use super::*;
use crate::imf::address::*;
use crate::imf::mailbox::*;
use crate::text::misc_token::*;
use chrono::{FixedOffset, TimeZone};
#[test]
fn test_header() {
let fullmail = b"Date: 7 Mar 2023 08:00:00 +0200
From: someone@example.com
To: someone_else@example.com
Subject: An RFC 822 formatted message
This is the plain text body of the message. Note the blank line
between the header information and the body of the message.";
assert_eq!(
imf(fullmail),
Ok((
&b"This is the plain text body of the message. Note the blank line\nbetween the header information and the body of the message."[..],
Imf {
date: Some(FixedOffset::east_opt(2 * 3600).unwrap().with_ymd_and_hms(2023, 3, 7, 8, 0, 0).unwrap()),
from: vec![MailboxRef {
name: None,
addrspec: AddrSpec {
local_part: LocalPart(vec![LocalPartToken::Word(Word::Atom(&b"someone"[..]))]),
domain: Domain::Atoms(vec![&b"example"[..], &b"com"[..]]),
}
}],
to: vec![AddressRef::Single(MailboxRef {
name: None,
addrspec: AddrSpec {
local_part: LocalPart(vec![LocalPartToken::Word(Word::Atom(&b"someone_else"[..]))]),
domain: Domain::Atoms(vec![&b"example"[..], &b"com"[..]]),
}
})],
subject: Some(Unstructured(vec![
UnstrToken::Plain(&b"An"[..]),
UnstrToken::Plain(&b"RFC"[..]),
UnstrToken::Plain(&b"822"[..]),
UnstrToken::Plain(&b"formatted"[..]),
UnstrToken::Plain(&b"message"[..]),
])),
..Imf::default()
}
)),
)
}
}

10
src/lib.rs
View file

@ -18,7 +18,7 @@ pub mod text;
/// Manipulate buffer of bytes /// Manipulate buffer of bytes
mod pointers; mod pointers;
use nom::{IResult, combinator::into}; use nom::{combinator::into, IResult};
/// Parse a whole email including its (MIME) body /// Parse a whole email including its (MIME) body
/// ///
@ -37,7 +37,7 @@ use nom::{IResult, combinator::into};
/// * `msg` - The parsed message /// * `msg` - The parsed message
/// ///
/// # Examples /// # Examples
/// ///
/// ``` /// ```
/// let input = br#"Date: 7 Mar 2023 08:00:00 +0200 /// let input = br#"Date: 7 Mar 2023 08:00:00 +0200
/// From: deuxfleurs@example.com /// From: deuxfleurs@example.com
@ -57,7 +57,9 @@ use nom::{IResult, combinator::into};
/// ); /// );
/// ``` /// ```
pub fn parse_message(input: &[u8]) -> IResult<&[u8], part::composite::Message> { pub fn parse_message(input: &[u8]) -> IResult<&[u8], part::composite::Message> {
into(part::composite::message(mime::MIME::<mime::r#type::DeductibleMessage>::default()))(input) into(part::composite::message(mime::MIME::<
mime::r#type::DeductibleMessage,
>::default()))(input)
} }
/// Only extract the headers of the email that are part of the Internet Message Format spec /// Only extract the headers of the email that are part of the Internet Message Format spec
@ -98,5 +100,5 @@ pub fn parse_message(input: &[u8]) -> IResult<&[u8], part::composite::Message> {
/// ); /// );
/// ``` /// ```
pub fn parse_imf(input: &[u8]) -> IResult<&[u8], imf::Imf> { pub fn parse_imf(input: &[u8]) -> IResult<&[u8], imf::Imf> {
imf::field::imf(input) imf::imf(input)
} }

61
src/mime/field.rs
View file

@ -1,16 +1,10 @@
use nom::{ use nom::combinator::map;
branch::alt,
combinator::map,
sequence::{preceded, terminated},
IResult,
};
use crate::header::{field_name}; use crate::header;
use crate::imf::identification::{msg_id, MessageID}; use crate::imf::identification::{msg_id, MessageID};
use crate::mime::mechanism::{mechanism, Mechanism}; use crate::mime::mechanism::{mechanism, Mechanism};
use crate::mime::r#type::{naive_type, NaiveType}; use crate::mime::r#type::{naive_type, NaiveType};
use crate::text::misc_token::{unstructured, Unstructured}; use crate::text::misc_token::{unstructured, Unstructured};
use crate::text::whitespace::obs_crlf;
#[derive(Debug, PartialEq)] #[derive(Debug, PartialEq)]
pub enum Content<'a> { pub enum Content<'a> {
@ -47,38 +41,38 @@ impl<'a> Content<'a> {
} }
} }
/* impl<'a> TryFrom<&header::Field<'a>> for Content<'a> {
pub fn to_mime<'a, T: WithDefaultType>(list: Vec<Content<'a>>) -> AnyMIMEWithDefault<'a, T> { type Error = ();
list.into_iter().collect::<AnyMIMEWithDefault<T>>() fn try_from(f: &header::Field<'a>) -> Result<Self, Self::Error> {
}*/ let content = match f {
header::Field::Good(header::Kv2(key, value)) => match key
.to_ascii_lowercase()
.as_slice()
{
b"content-type" => map(naive_type, Content::Type)(value),
b"content-transfer-encoding" => map(mechanism, Content::TransferEncoding)(value),
b"content-id" => map(msg_id, Content::ID)(value),
b"content-description" => map(unstructured, Content::Description)(value),
_ => return Err(()),
},
_ => return Err(()),
};
pub fn content(input: &[u8]) -> IResult<&[u8], Content> { //@TODO check that the full value is parsed, otherwise maybe log an error ?!
terminated( content.map(|(_, content)| content).or(Err(()))
alt(( }
preceded(field_name(b"content-type"), map(naive_type, Content::Type)),
preceded(
field_name(b"content-transfer-encoding"),
map(mechanism, Content::TransferEncoding),
),
preceded(field_name(b"content-id"), map(msg_id, Content::ID)),
preceded(
field_name(b"content-description"),
map(unstructured, Content::Description),
),
)),
obs_crlf,
)(input)
} }
#[cfg(test)] #[cfg(test)]
mod tests { mod tests {
use super::*; use super::*;
use crate::header::{header}; use crate::header;
use crate::mime::charset::EmailCharset; //use crate::mime::charset::EmailCharset;
use crate::mime::r#type::*; use crate::mime::r#type::*;
use crate::text::misc_token::MIMEWord; use crate::text::misc_token::MIMEWord;
use crate::text::quoted::QuotedString; use crate::text::quoted::QuotedString;
/*
#[test] #[test]
fn test_content_type() { fn test_content_type() {
let (rest, content) = let (rest, content) =
@ -96,7 +90,7 @@ mod tests {
} else { } else {
panic!("Expected Content::Type, got {:?}", content); panic!("Expected Content::Type, got {:?}", content);
} }
} }*/
#[test] #[test]
fn test_header() { fn test_header() {
@ -116,7 +110,10 @@ This is a multipart message.
.as_bytes(); .as_bytes();
assert_eq!( assert_eq!(
map(header(content), |(k, _, _)| k)(fullmail), map(header::header_kv, |k| k
.iter()
.flat_map(Content::try_from)
.collect())(fullmail),
Ok(( Ok((
&b"This is a multipart message.\n\n"[..], &b"This is a multipart message.\n\n"[..],
vec![ vec![

54
src/mime/mod.rs
View file

@ -1,4 +1,4 @@
/// Parsed and represent an email character set /// Parsed and represent an email character set
pub mod charset; pub mod charset;
/// MIME specific headers /// MIME specific headers
@ -10,19 +10,20 @@ pub mod mechanism;
/// Content-Type representation /// Content-Type representation
pub mod r#type; pub mod r#type;
use std::fmt;
use std::marker::PhantomData; use std::marker::PhantomData;
use crate::header;
use crate::imf::identification::MessageID; use crate::imf::identification::MessageID;
use crate::mime::field::Content; use crate::mime::field::Content;
use crate::mime::mechanism::Mechanism; use crate::mime::mechanism::Mechanism;
use crate::mime::r#type::{AnyType, NaiveType}; use crate::mime::r#type::{AnyType, NaiveType};
use crate::header;
use crate::text::misc_token::Unstructured; //Multipart, Message, Text, Binary}; use crate::text::misc_token::Unstructured; //Multipart, Message, Text, Binary};
#[derive(Debug, PartialEq, Clone)] #[derive(Debug, PartialEq, Clone)]
pub struct MIME<'a, T> { pub struct MIME<'a, T> {
pub interpreted_type: T, pub interpreted_type: T,
pub fields: NaiveMIME<'a> pub fields: NaiveMIME<'a>,
} }
impl<'a> Default for MIME<'a, r#type::DeductibleText> { impl<'a> Default for MIME<'a, r#type::DeductibleText> {
fn default() -> Self { fn default() -> Self {
@ -55,22 +56,32 @@ impl<'a, T: WithDefaultType> From<AnyMIMEWithDefault<'a, T>> for AnyMIME<'a> {
} }
} }
#[derive(Debug, PartialEq, Default, Clone)] #[derive(PartialEq, Default, Clone)]
pub struct NaiveMIME<'a> { pub struct NaiveMIME<'a> {
pub ctype: Option<NaiveType<'a>>, pub ctype: Option<NaiveType<'a>>,
pub transfer_encoding: Mechanism<'a>, pub transfer_encoding: Mechanism<'a>,
pub id: Option<MessageID<'a>>, pub id: Option<MessageID<'a>>,
pub description: Option<Unstructured<'a>>, pub description: Option<Unstructured<'a>>,
pub header_ext: Vec<header::Kv<'a>>, pub kv: Vec<header::Field<'a>>,
pub header_bad: Vec<&'a [u8]>,
pub raw: &'a [u8], pub raw: &'a [u8],
} }
impl<'a> fmt::Debug for NaiveMIME<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("NaiveMime")
.field("ctype", &self.ctype)
.field("transfer_encoding", &self.transfer_encoding)
.field("id", &self.id)
.field("description", &self.description)
.field("kv", &self.kv)
.field("raw", &String::from_utf8_lossy(self.raw))
.finish()
}
}
impl<'a> FromIterator<Content<'a>> for NaiveMIME<'a> { impl<'a> FromIterator<Content<'a>> for NaiveMIME<'a> {
fn from_iter<I: IntoIterator<Item = Content<'a>>>(it: I) -> Self { fn from_iter<I: IntoIterator<Item = Content<'a>>>(it: I) -> Self {
it.into_iter().fold( it.into_iter()
NaiveMIME::default(), .fold(NaiveMIME::default(), |mut section, field| {
|mut section, field| {
match field { match field {
Content::Type(v) => section.ctype = Some(v), Content::Type(v) => section.ctype = Some(v),
Content::TransferEncoding(v) => section.transfer_encoding = v, Content::TransferEncoding(v) => section.transfer_encoding = v,
@ -78,28 +89,29 @@ impl<'a> FromIterator<Content<'a>> for NaiveMIME<'a> {
Content::Description(v) => section.description = Some(v), Content::Description(v) => section.description = Some(v),
}; };
section section
}, })
)
} }
} }
impl<'a> NaiveMIME<'a> { impl<'a> NaiveMIME<'a> {
pub fn with_opt(mut self, opt: Vec<header::Kv<'a>>) -> Self { pub fn with_kv(mut self, fields: Vec<header::Field<'a>>) -> Self {
self.header_ext = opt; self self.kv = fields;
} self
pub fn with_bad(mut self, bad: Vec<&'a [u8]>) -> Self {
self.header_bad = bad; self
} }
pub fn with_raw(mut self, raw: &'a [u8]) -> Self { pub fn with_raw(mut self, raw: &'a [u8]) -> Self {
self.raw = raw; self self.raw = raw;
self
} }
pub fn to_interpreted<T: WithDefaultType>(self) -> AnyMIME<'a> { pub fn to_interpreted<T: WithDefaultType>(self) -> AnyMIME<'a> {
self.ctype.as_ref().map(|c| c.to_type()).unwrap_or(T::default_type()).to_mime(self).into() self.ctype
.as_ref()
.map(|c| c.to_type())
.unwrap_or(T::default_type())
.to_mime(self)
.into()
} }
} }
pub trait WithDefaultType { pub trait WithDefaultType {
fn default_type() -> AnyType; fn default_type() -> AnyType;
} }

90
src/mime/type.rs
View file

@ -5,19 +5,29 @@ use nom::{
sequence::{preceded, terminated, tuple}, sequence::{preceded, terminated, tuple},
IResult, IResult,
}; };
use std::fmt;
use crate::mime::charset::EmailCharset; use crate::mime::charset::EmailCharset;
use crate::mime::{AnyMIME, NaiveMIME, MIME};
use crate::text::misc_token::{mime_word, MIMEWord}; use crate::text::misc_token::{mime_word, MIMEWord};
use crate::text::words::mime_atom; use crate::text::words::mime_atom;
use crate::mime::{AnyMIME, MIME, NaiveMIME};
// --------- NAIVE TYPE // --------- NAIVE TYPE
#[derive(Debug, PartialEq, Clone)] #[derive(PartialEq, Clone)]
pub struct NaiveType<'a> { pub struct NaiveType<'a> {
pub main: &'a [u8], pub main: &'a [u8],
pub sub: &'a [u8], pub sub: &'a [u8],
pub params: Vec<Parameter<'a>>, pub params: Vec<Parameter<'a>>,
} }
impl<'a> fmt::Debug for NaiveType<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("mime::NaiveType")
.field("main", &String::from_utf8_lossy(self.main))
.field("sub", &String::from_utf8_lossy(self.sub))
.field("params", &self.params)
.finish()
}
}
impl<'a> NaiveType<'a> { impl<'a> NaiveType<'a> {
pub fn to_type(&self) -> AnyType { pub fn to_type(&self) -> AnyType {
self.into() self.into()
@ -30,11 +40,20 @@ pub fn naive_type(input: &[u8]) -> IResult<&[u8], NaiveType> {
)(input) )(input)
} }
#[derive(Debug, PartialEq, Clone)] #[derive(PartialEq, Clone)]
pub struct Parameter<'a> { pub struct Parameter<'a> {
pub name: &'a [u8], pub name: &'a [u8],
pub value: MIMEWord<'a>, pub value: MIMEWord<'a>,
} }
impl<'a> fmt::Debug for Parameter<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("mime::Parameter")
.field("name", &String::from_utf8_lossy(self.name))
.field("value", &self.value)
.finish()
}
}
pub fn parameter(input: &[u8]) -> IResult<&[u8], Parameter> { pub fn parameter(input: &[u8]) -> IResult<&[u8], Parameter> {
map( map(
tuple((mime_atom, tag(b"="), mime_word)), tuple((mime_atom, tag(b"="), mime_word)),
@ -73,16 +92,27 @@ impl<'a> From<&'a NaiveType<'a>> for AnyType {
impl<'a> AnyType { impl<'a> AnyType {
pub fn to_mime(self, fields: NaiveMIME<'a>) -> AnyMIME<'a> { pub fn to_mime(self, fields: NaiveMIME<'a>) -> AnyMIME<'a> {
match self { match self {
Self::Multipart(interpreted_type) => AnyMIME::Mult(MIME::<Multipart> { interpreted_type, fields }), Self::Multipart(interpreted_type) => AnyMIME::Mult(MIME::<Multipart> {
Self::Message(interpreted_type) => AnyMIME::Msg(MIME::<DeductibleMessage> { interpreted_type, fields }), interpreted_type,
Self::Text(interpreted_type) => AnyMIME::Txt(MIME::<DeductibleText> { interpreted_type, fields }), fields,
Self::Binary(interpreted_type) => AnyMIME::Bin(MIME::<Binary> { interpreted_type, fields }), }),
} Self::Message(interpreted_type) => AnyMIME::Msg(MIME::<DeductibleMessage> {
interpreted_type,
fields,
}),
Self::Text(interpreted_type) => AnyMIME::Txt(MIME::<DeductibleText> {
interpreted_type,
fields,
}),
Self::Binary(interpreted_type) => AnyMIME::Bin(MIME::<Binary> {
interpreted_type,
fields,
}),
}
} }
} }
#[derive(Debug, PartialEq, Clone)] #[derive(Debug, PartialEq, Clone)]
pub enum Deductible<T: Default> { pub enum Deductible<T: Default> {
Inferred(T), Inferred(T),
@ -139,7 +169,8 @@ impl ToString for MultipartSubtype {
Self::Parallel => "parallel", Self::Parallel => "parallel",
Self::Report => "report", Self::Report => "report",
Self::Unknown => "mixed", Self::Unknown => "mixed",
}.into() }
.into()
} }
} }
impl<'a> From<&NaiveType<'a>> for MultipartSubtype { impl<'a> From<&NaiveType<'a>> for MultipartSubtype {
@ -155,8 +186,6 @@ impl<'a> From<&NaiveType<'a>> for MultipartSubtype {
} }
} }
#[derive(Debug, PartialEq, Default, Clone)] #[derive(Debug, PartialEq, Default, Clone)]
pub enum MessageSubtype { pub enum MessageSubtype {
#[default] #[default]
@ -172,7 +201,8 @@ impl ToString for MessageSubtype {
Self::Partial => "partial", Self::Partial => "partial",
Self::External => "external", Self::External => "external",
Self::Unknown => "rfc822", Self::Unknown => "rfc822",
}.into() }
.into()
} }
} }
@ -184,17 +214,25 @@ pub struct Message {
impl<'a> From<&NaiveType<'a>> for Message { impl<'a> From<&NaiveType<'a>> for Message {
fn from(nt: &NaiveType<'a>) -> Self { fn from(nt: &NaiveType<'a>) -> Self {
match nt.sub.to_ascii_lowercase().as_slice() { match nt.sub.to_ascii_lowercase().as_slice() {
b"rfc822" => Self { subtype: MessageSubtype::RFC822 }, b"rfc822" => Self {
b"partial" => Self { subtype: MessageSubtype::Partial }, subtype: MessageSubtype::RFC822,
b"external" => Self { subtype: MessageSubtype::External }, },
_ => Self { subtype: MessageSubtype::Unknown }, b"partial" => Self {
subtype: MessageSubtype::Partial,
},
b"external" => Self {
subtype: MessageSubtype::External,
},
_ => Self {
subtype: MessageSubtype::Unknown,
},
} }
} }
} }
impl From<Deductible<Message>> for Message { impl From<Deductible<Message>> for Message {
fn from(d: Deductible<Message>) -> Self { fn from(d: Deductible<Message>) -> Self {
match d { match d {
Deductible::Inferred(t) | Deductible::Explicit(t) => t Deductible::Inferred(t) | Deductible::Explicit(t) => t,
} }
} }
} }
@ -221,7 +259,7 @@ impl<'a> From<&NaiveType<'a>> for Text {
impl From<Deductible<Text>> for Text { impl From<Deductible<Text>> for Text {
fn from(d: Deductible<Text>) -> Self { fn from(d: Deductible<Text>) -> Self {
match d { match d {
Deductible::Inferred(t) | Deductible::Explicit(t) => t Deductible::Inferred(t) | Deductible::Explicit(t) => t,
} }
} }
} }
@ -238,7 +276,8 @@ impl ToString for TextSubtype {
match self { match self {
Self::Plain | Self::Unknown => "plain", Self::Plain | Self::Unknown => "plain",
Self::Html => "html", Self::Html => "html",
}.into() }
.into()
} }
} }
impl<'a> From<&NaiveType<'a>> for TextSubtype { impl<'a> From<&NaiveType<'a>> for TextSubtype {
@ -258,8 +297,8 @@ pub struct Binary {}
mod tests { mod tests {
use super::*; use super::*;
use crate::mime::charset::EmailCharset; use crate::mime::charset::EmailCharset;
use crate::text::quoted::QuotedString;
use crate::mime::r#type::Deductible; use crate::mime::r#type::Deductible;
use crate::text::quoted::QuotedString;
#[test] #[test]
fn test_parameter() { fn test_parameter() {
@ -317,7 +356,12 @@ mod tests {
let (rest, nt) = naive_type(b"message/rfc822").unwrap(); let (rest, nt) = naive_type(b"message/rfc822").unwrap();
assert_eq!(rest, &[]); assert_eq!(rest, &[]);
assert_eq!(nt.to_type(), AnyType::Message(Deductible::Explicit(Message { subtype: MessageSubtype::RFC822 }))); assert_eq!(
nt.to_type(),
AnyType::Message(Deductible::Explicit(Message {
subtype: MessageSubtype::RFC822
}))
);
} }
#[test] #[test]

137
src/part/composite.rs
View file

@ -1,20 +1,37 @@
use nom::IResult; use nom::IResult;
use std::fmt;
use crate::header::{header, self}; use crate::header;
use crate::imf; use crate::imf;
use crate::mime; use crate::mime;
use crate::part::{self, AnyPart, field::MixedField}; use crate::part::{self, AnyPart};
use crate::text::boundary::{boundary, Delimiter};
use crate::pointers; use crate::pointers;
use crate::text::boundary::{boundary, Delimiter};
//--- Multipart //--- Multipart
#[derive(Debug, PartialEq)] #[derive(PartialEq)]
pub struct Multipart<'a> { pub struct Multipart<'a> {
pub mime: mime::MIME<'a, mime::r#type::Multipart>, pub mime: mime::MIME<'a, mime::r#type::Multipart>,
pub children: Vec<AnyPart<'a>>, pub children: Vec<AnyPart<'a>>,
pub raw_part_inner: &'a [u8], pub raw_part_inner: &'a [u8],
pub raw_part_outer: &'a [u8], pub raw_part_outer: &'a [u8],
} }
impl<'a> fmt::Debug for Multipart<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("part::Multipart")
.field("mime", &self.mime)
.field("children", &self.children)
.field(
"raw_part_inner",
&String::from_utf8_lossy(self.raw_part_inner),
)
.field(
"raw_part_outer",
&String::from_utf8_lossy(self.raw_part_outer),
)
.finish()
}
}
impl<'a> Multipart<'a> { impl<'a> Multipart<'a> {
pub fn preamble(&self) -> &'a [u8] { pub fn preamble(&self) -> &'a [u8] {
pointers::parsed(self.raw_part_outer, self.raw_part_inner) pointers::parsed(self.raw_part_outer, self.raw_part_inner)
@ -65,7 +82,10 @@ pub fn multipart<'a>(
mime: m.clone(), mime: m.clone(),
children: mparts, children: mparts,
raw_part_inner: pointers::parsed(inner_orig, inp), raw_part_inner: pointers::parsed(inner_orig, inp),
raw_part_outer: pointers::parsed(outer_orig, &outer_orig[outer_orig.len()..]), raw_part_outer: pointers::parsed(
outer_orig,
&outer_orig[outer_orig.len()..],
),
}, },
)) ))
} }
@ -73,24 +93,30 @@ pub fn multipart<'a>(
}; };
// parse mime headers, otherwise pick default mime // parse mime headers, otherwise pick default mime
let (input, naive_mime) = match header(mime::field::content)(input) { let (input, naive_mime) = match header::header_kv(input) {
Ok((input_eom, (known, unknown, bad))) => { Ok((input_eom, fields)) => {
let raw_hdrs = pointers::parsed(input, input_eom); let raw_hdrs = pointers::parsed(input, input_eom);
let mime = known let mime = fields
.iter()
.flat_map(mime::field::Content::try_from)
.into_iter() .into_iter()
.collect::<mime::NaiveMIME>() .collect::<mime::NaiveMIME>();
.with_opt(unknown)
.with_bad(bad) let mime = mime.with_kv(fields).with_raw(raw_hdrs);
.with_raw(raw_hdrs);
(input_eom, mime) (input_eom, mime)
}, }
Err(_) => (input, mime::NaiveMIME::default()), Err(_) => (input, mime::NaiveMIME::default()),
}; };
// interpret mime according to context // interpret mime according to context
let mime = match m.interpreted_type.subtype { let mime = match m.interpreted_type.subtype {
mime::r#type::MultipartSubtype::Digest => naive_mime.to_interpreted::<mime::WithDigestDefault>().into(), mime::r#type::MultipartSubtype::Digest => naive_mime
_ => naive_mime.to_interpreted::<mime::WithGenericDefault>().into(), .to_interpreted::<mime::WithDigestDefault>()
.into(),
_ => naive_mime
.to_interpreted::<mime::WithGenericDefault>()
.into(),
}; };
// parse raw part // parse raw part
@ -109,7 +135,7 @@ pub fn multipart<'a>(
//--- Message //--- Message
#[derive(Debug, PartialEq)] #[derive(PartialEq)]
pub struct Message<'a> { pub struct Message<'a> {
pub mime: mime::MIME<'a, mime::r#type::DeductibleMessage>, pub mime: mime::MIME<'a, mime::r#type::DeductibleMessage>,
pub imf: imf::Imf<'a>, pub imf: imf::Imf<'a>,
@ -119,6 +145,18 @@ pub struct Message<'a> {
pub raw_headers: &'a [u8], pub raw_headers: &'a [u8],
pub raw_body: &'a [u8], pub raw_body: &'a [u8],
} }
impl<'a> fmt::Debug for Message<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("part::Message")
.field("mime", &self.mime)
.field("imf", &self.imf)
.field("child", &self.child)
.field("raw_part", &String::from_utf8_lossy(self.raw_part))
.field("raw_headers", &String::from_utf8_lossy(self.raw_headers))
.field("raw_body", &String::from_utf8_lossy(self.raw_body))
.finish()
}
}
pub fn message<'a>( pub fn message<'a>(
m: mime::MIME<'a, mime::r#type::DeductibleMessage>, m: mime::MIME<'a, mime::r#type::DeductibleMessage>,
@ -127,23 +165,27 @@ pub fn message<'a>(
let orig = input; let orig = input;
// parse header fields // parse header fields
let (input, (known, unknown, bad)): (_, (Vec::<MixedField>, Vec<header::Kv>, Vec<&[u8]>)) = let (input, headers) = header::header_kv(input)?;
header(part::field::mixed_field)(input)?;
// extract raw parts 1/2 // extract raw parts 1/2
let raw_headers = pointers::parsed(orig, input); let raw_headers = pointers::parsed(orig, input);
let body_orig = input; let body_orig = input;
//---------------
// aggregate header fields // aggregate header fields
let (naive_mime, imf) = part::field::sections(known); let (naive_mime, imf) = part::field::split_and_build(&headers);
// attach bad headers to imf // Bind headers to the IMF object
let imf = imf.with_opt(unknown).with_bad(bad); let imf = imf.with_kv(headers);
// interpret headers to choose a mime type // interpret headers to choose a mime type
let in_mime = naive_mime.with_raw(raw_headers).to_interpreted::<mime::WithGenericDefault>().into(); let in_mime = naive_mime
.with_raw(raw_headers)
.to_interpreted::<mime::WithGenericDefault>()
.into();
//---------------
// parse this mimetype // parse a part following this mime specification
let (input, part) = part::anypart(in_mime)(input)?; let (input, part) = part::anypart(in_mime)(input)?;
// extract raw parts 2/2 // extract raw parts 2/2
@ -155,7 +197,9 @@ pub fn message<'a>(
Message { Message {
mime: m.clone(), mime: m.clone(),
imf, imf,
raw_part, raw_headers, raw_body, raw_part,
raw_headers,
raw_body,
child: Box::new(part), child: Box::new(part),
}, },
)) ))
@ -168,7 +212,7 @@ mod tests {
use crate::part::discrete::Text; use crate::part::discrete::Text;
use crate::part::AnyPart; use crate::part::AnyPart;
use crate::text::encoding::{Base64Word, EncodedWord, QuotedChunk, QuotedWord}; use crate::text::encoding::{Base64Word, EncodedWord, QuotedChunk, QuotedWord};
use crate::text::misc_token::{Phrase, UnstrToken, Unstructured, Word, MIMEWord}; use crate::text::misc_token::{MIMEWord, Phrase, UnstrToken, Unstructured, Word};
use crate::text::quoted::QuotedString; use crate::text::quoted::QuotedString;
use chrono::{FixedOffset, TimeZone}; use chrono::{FixedOffset, TimeZone};
@ -229,12 +273,15 @@ It DOES end with a linebreak.
subtype: mime::r#type::TextSubtype::Plain, subtype: mime::r#type::TextSubtype::Plain,
charset: mime::r#type::Deductible::Inferred(mime::charset::EmailCharset::US_ASCII), charset: mime::r#type::Deductible::Inferred(mime::charset::EmailCharset::US_ASCII),
}), }),
fields: mime::NaiveMIME::default(), fields: mime::NaiveMIME {
raw: &b"\n"[..],
..mime::NaiveMIME::default()
},
}, },
body: &b"This is implicitly typed plain US-ASCII text.\nIt does NOT end with a linebreak."[..], body: &b"This is implicitly typed plain US-ASCII text.\nIt does NOT end with a linebreak."[..],
}), }),
AnyPart::Txt(Text { AnyPart::Txt(Text {
mime: mime::MIME { mime: mime::MIME {
interpreted_type: mime::r#type::Deductible::Explicit(mime::r#type::Text { interpreted_type: mime::r#type::Deductible::Explicit(mime::r#type::Text {
subtype: mime::r#type::TextSubtype::Plain, subtype: mime::r#type::TextSubtype::Plain,
charset: mime::r#type::Deductible::Explicit(mime::charset::EmailCharset::US_ASCII), charset: mime::r#type::Deductible::Explicit(mime::charset::EmailCharset::US_ASCII),
@ -250,6 +297,10 @@ It DOES end with a linebreak.
} }
] ]
}), }),
raw: &b"Content-type: text/plain; charset=us-ascii\n\n"[..],
kv: vec![
header::Field::Good(header::Kv2(&b"Content-type"[..], &b"text/plain; charset=us-ascii"[..]))
],
..mime::NaiveMIME::default() ..mime::NaiveMIME::default()
}, },
}, },
@ -455,14 +506,18 @@ OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO<br />
right: &b"www.grrrndzero.org"[..], right: &b"www.grrrndzero.org"[..],
}), }),
mime_version: Some(imf::mime::Version { major: 1, minor: 0}), mime_version: Some(imf::mime::Version { major: 1, minor: 0}),
header_ext: vec![ kv: vec![
header::Kv(&b"X-Unknown"[..], Unstructured(vec![ header::Field::Good(header::Kv2(&b"Date"[..], &b"Sat, 8 Jul 2023 07:14:29 +0200"[..])),
UnstrToken::Plain(&b"something"[..]), header::Field::Good(header::Kv2(&b"From"[..], &b"Grrrnd Zero <grrrndzero@example.org>"[..])),
UnstrToken::Plain(&b"something"[..]), header::Field::Good(header::Kv2(&b"To"[..], &b"John Doe <jdoe@machine.example>"[..])),
])) header::Field::Good(header::Kv2(&b"CC"[..], &b"=?ISO-8859-1?Q?Andr=E9?= Pirard <PIRARD@vm1.ulg.ac.be>"[..])),
], header::Field::Good(header::Kv2(&b"Subject"[..], &b"=?ISO-8859-1?B?SWYgeW91IGNhbiByZWFkIHRoaXMgeW8=?=\n =?ISO-8859-2?B?dSB1bmRlcnN0YW5kIHRoZSBleGFtcGxlLg==?="[..])),
header_bad: vec![ header::Field::Good(header::Kv2(&b"X-Unknown"[..], &b"something something"[..])),
&b"Bad entry\n on multiple lines\n"[..], header::Field::Bad(&b"Bad entry\n on multiple lines\n"[..]),
header::Field::Good(header::Kv2(&b"Message-ID"[..], &b"<NTAxNzA2AC47634Y366BAMTY4ODc5MzQyODY0ODY5@www.grrrndzero.org>"[..])),
header::Field::Good(header::Kv2(&b"MIME-Version"[..], &b"1.0"[..])),
header::Field::Good(header::Kv2(&b"Content-Type"[..], &b"multipart/alternative;\n boundary=\"b1_e376dc71bafc953c0b0fdeb9983a9956\""[..])),
header::Field::Good(header::Kv2(&b"Content-Transfer-Encoding"[..], &b"7bit"[..])),
], ],
..imf::Imf::default() ..imf::Imf::default()
}, },
@ -483,6 +538,7 @@ OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO<br />
} }
] ]
}), }),
raw: hdrs,
..mime::NaiveMIME::default() ..mime::NaiveMIME::default()
}, },
}, },
@ -507,6 +563,11 @@ OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO<br />
] ]
}), }),
transfer_encoding: mime::mechanism::Mechanism::QuotedPrintable, transfer_encoding: mime::mechanism::Mechanism::QuotedPrintable,
kv: vec![
header::Field::Good(header::Kv2(&b"Content-Type"[..], &b"text/plain; charset=utf-8"[..])),
header::Field::Good(header::Kv2(&b"Content-Transfer-Encoding"[..], &b"quoted-printable"[..])),
],
raw: &b"Content-Type: text/plain; charset=utf-8\nContent-Transfer-Encoding: quoted-printable\n\n"[..],
..mime::NaiveMIME::default() ..mime::NaiveMIME::default()
} }
}, },
@ -529,7 +590,11 @@ OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO<br />
value: MIMEWord::Atom(&b"us-ascii"[..]), value: MIMEWord::Atom(&b"us-ascii"[..]),
} }
] ]
}), }),
kv: vec![
header::Field::Good(header::Kv2(&b"Content-Type"[..], &b"text/html; charset=us-ascii"[..])),
],
raw: &b"Content-Type: text/html; charset=us-ascii\n\n"[..],
..mime::NaiveMIME::default() ..mime::NaiveMIME::default()
}, },
}, },

10
src/part/discrete.rs
View file

@ -12,10 +12,7 @@ impl<'a> fmt::Debug for Text<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result { fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("part::Text") fmt.debug_struct("part::Text")
.field("mime", &self.mime) .field("mime", &self.mime)
.field( .field("body", &String::from_utf8_lossy(self.body))
"body",
&format_args!("\"{}\"", String::from_utf8_lossy(self.body)),
)
.finish() .finish()
} }
} }
@ -30,10 +27,7 @@ impl<'a> fmt::Debug for Binary<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result { fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("part::Binary") fmt.debug_struct("part::Binary")
.field("mime", &self.mime) .field("mime", &self.mime)
.field( .field("body", &String::from_utf8_lossy(self.body))
"body",
&format_args!("\"{}\"", String::from_utf8_lossy(self.body)),
)
.finish() .finish()
} }
} }

64
src/part/field.rs
View file

@ -1,50 +1,24 @@
use nom::{branch::alt, combinator::map, IResult}; use crate::header;
use crate::imf; use crate::imf;
use crate::mime; use crate::mime;
pub enum MixedField<'a> { pub fn split_and_build<'a>(v: &Vec<header::Field<'a>>) -> (mime::NaiveMIME<'a>, imf::Imf<'a>) {
MIME(mime::field::Content<'a>), let (mimev, imfv) = v.iter().fold(
IMF(imf::field::Field<'a>), (
} Vec::<mime::field::Content>::new(),
#[allow(dead_code)] Vec::<imf::field::Field>::new(),
impl<'a> MixedField<'a> { ),
pub fn mime(&self) -> Option<&mime::field::Content<'a>> { |(mut mime, mut imf), f| {
match self { if let Ok(m) = mime::field::Content::try_from(f) {
Self::MIME(v) => Some(v), mime.push(m);
_ => None, } else if let Ok(i) = imf::field::Field::try_from(f) {
} imf.push(i);
} }
pub fn to_mime(self) -> Option<mime::field::Content<'a>> { (mime, imf)
match self { },
Self::MIME(v) => Some(v), );
_ => None,
}
}
pub fn imf(&self) -> Option<&imf::field::Field<'a>> {
match self {
Self::IMF(v) => Some(v),
_ => None,
}
}
pub fn to_imf(self) -> Option<imf::field::Field<'a>> {
match self {
Self::IMF(v) => Some(v),
_ => None,
}
}
}
pub fn sections<'a>(list: Vec<MixedField<'a>>) -> (mime::NaiveMIME<'a>, imf::Imf<'a>) { let fmime = mimev.into_iter().collect::<mime::NaiveMIME>();
let (v1, v2): (Vec<MixedField>, Vec<_>) = list.into_iter().partition(|v| v.mime().is_some()); let fimf = imfv.into_iter().collect::<imf::Imf>();
let mime = v1.into_iter().flat_map(MixedField::to_mime).collect::<mime::NaiveMIME>(); (fmime, fimf)
let imf = v2.into_iter().flat_map(MixedField::to_imf).collect::<imf::Imf>();
(mime, imf)
}
pub fn mixed_field(input: &[u8]) -> IResult<&[u8], MixedField> {
alt((
map(mime::field::content, MixedField::MIME),
map(imf::field::field, MixedField::IMF),
))(input)
} }

19
src/part/mod.rs
View file

@ -80,18 +80,19 @@ pub fn anypart<'a>(m: AnyMIME<'a>) -> impl FnOnce(&'a [u8]) -> IResult<&'a [u8],
move |input| { move |input| {
let part = match m { let part = match m {
AnyMIME::Mult(a) => multipart(a)(input) AnyMIME::Mult(a) => multipart(a)(input)
.map(|(_, multi)| .map(|(_, multi)| multi.into())
multi.into())
.unwrap_or(AnyPart::Txt(Text {
mime: mime::MIME::<mime::r#type::DeductibleText>::default(),
body: input,
})),
AnyMIME::Msg(a) => message(a)(input)
.map(|(_, msg)| msg.into())
.unwrap_or(AnyPart::Txt(Text { .unwrap_or(AnyPart::Txt(Text {
mime: mime::MIME::<mime::r#type::DeductibleText>::default(), mime: mime::MIME::<mime::r#type::DeductibleText>::default(),
body: input, body: input,
})), })),
AnyMIME::Msg(a) => {
message(a)(input)
.map(|(_, msg)| msg.into())
.unwrap_or(AnyPart::Txt(Text {
mime: mime::MIME::<mime::r#type::DeductibleText>::default(),
body: input,
}))
}
AnyMIME::Txt(a) => AnyPart::Txt(Text { AnyMIME::Txt(a) => AnyPart::Txt(Text {
mime: a, mime: a,
body: input, body: input,
@ -101,7 +102,7 @@ pub fn anypart<'a>(m: AnyMIME<'a>) -> impl FnOnce(&'a [u8]) -> IResult<&'a [u8],
body: input, body: input,
}), }),
}; };
// This function always consumes the whole input // This function always consumes the whole input
Ok((&input[input.len()..], part)) Ok((&input[input.len()..], part))
} }

7
src/text/whitespace.rs
View file

@ -22,7 +22,12 @@ use nom::{
/// \r or \n is allowed nowhere else, so we also add this support. /// \r or \n is allowed nowhere else, so we also add this support.
pub fn obs_crlf(input: &[u8]) -> IResult<&[u8], &[u8]> { pub fn obs_crlf(input: &[u8]) -> IResult<&[u8], &[u8]> {
alt((tag(ascii::CRLF), tag(ascii::CRCRLF), tag(&[ascii::CR]), tag(&[ascii::LF])))(input) alt((
tag(ascii::CRLF),
tag(ascii::CRCRLF),
tag(&[ascii::CR]),
tag(&[ascii::LF]),
))(input)
} }
/// ```abnf /// ```abnf