diff --git a/src/header.rs b/src/header.rs index dac6d2d..5c889a3 100644 --- a/src/header.rs +++ b/src/header.rs @@ -1,4 +1,3 @@ -use std::fmt; use nom::{ branch::alt, bytes::complete::{tag, take_while1}, @@ -8,9 +7,10 @@ use nom::{ sequence::{pair, terminated, tuple}, IResult, }; +use std::fmt; -use crate::text::whitespace::{foldable_line, obs_crlf}; use crate::text::misc_token::unstructured; +use crate::text::whitespace::{foldable_line, obs_crlf}; #[derive(PartialEq, Clone)] pub struct Kv2<'a>(pub &'a [u8], pub &'a [u8]); @@ -47,13 +47,8 @@ impl<'a> From<&'a [u8]> for Field<'a> { /// Parse headers as key/values pub fn header_kv(input: &[u8]) -> IResult<&[u8], Vec> { terminated( - many0( - alt(( - into(correct_field), - into(foldable_line), - )) - ), - obs_crlf + many0(alt((into(correct_field), into(foldable_line)))), + obs_crlf, )(input) } @@ -74,11 +69,5 @@ pub fn field_any(input: &[u8]) -> IResult<&[u8], &[u8]> { /// ; ":". /// ``` pub fn correct_field(input: &[u8]) -> IResult<&[u8], Kv2> { - terminated( - into(pair( - field_any, - recognize(unstructured), - )), - obs_crlf, - )(input) + terminated(into(pair(field_any, recognize(unstructured))), obs_crlf)(input) } diff --git a/src/imf/field.rs b/src/imf/field.rs index 17a48f9..0a7766d 100644 --- a/src/imf/field.rs +++ b/src/imf/field.rs @@ -46,25 +46,27 @@ impl<'a> TryFrom<&header::Field<'a>> for Field<'a> { type Error = (); fn try_from(f: &header::Field<'a>) -> Result { let content = match f { - header::Field::Good(header::Kv2(key, value)) => match key.to_ascii_lowercase().as_slice() { - b"date" => map(date, Field::Date)(value), - b"from" => map(mailbox_list, Field::From)(value), - b"sender" => map(mailbox, Field::Sender)(value), - b"reply-to" => map(address_list, Field::ReplyTo)(value), - b"to" => map(address_list, Field::To)(value), - b"cc" => map(address_list, Field::Cc)(value), - b"bcc" => map(nullable_address_list, Field::Bcc)(value), - b"message-id" => map(msg_id, Field::MessageID)(value), - b"in-reply-to" => map(msg_list, Field::InReplyTo)(value), - b"references" => map(msg_list, Field::References)(value), - b"subject" => map(unstructured, Field::Subject)(value), - b"comments" => map(unstructured, Field::Comments)(value), - b"keywords" => map(phrase_list, Field::Keywords)(value), - b"return-path" => map(return_path, Field::ReturnPath)(value), - b"received" => map(received_log, Field::Received)(value), - b"mime-version" => map(version, Field::MIMEVersion)(value), - _ => return Err(()), - }, + header::Field::Good(header::Kv2(key, value)) => { + match key.to_ascii_lowercase().as_slice() { + b"date" => map(date, Field::Date)(value), + b"from" => map(mailbox_list, Field::From)(value), + b"sender" => map(mailbox, Field::Sender)(value), + b"reply-to" => map(address_list, Field::ReplyTo)(value), + b"to" => map(address_list, Field::To)(value), + b"cc" => map(address_list, Field::Cc)(value), + b"bcc" => map(nullable_address_list, Field::Bcc)(value), + b"message-id" => map(msg_id, Field::MessageID)(value), + b"in-reply-to" => map(msg_list, Field::InReplyTo)(value), + b"references" => map(msg_list, Field::References)(value), + b"subject" => map(unstructured, Field::Subject)(value), + b"comments" => map(unstructured, Field::Comments)(value), + b"keywords" => map(phrase_list, Field::Keywords)(value), + b"return-path" => map(return_path, Field::ReturnPath)(value), + b"received" => map(received_log, Field::Received)(value), + b"mime-version" => map(version, Field::MIMEVersion)(value), + _ => return Err(()), + } + } _ => return Err(()), }; diff --git a/src/imf/mod.rs b/src/imf/mod.rs index 32eb925..b920c49 100644 --- a/src/imf/mod.rs +++ b/src/imf/mod.rs @@ -1,5 +1,4 @@ /// Parse and represent IMF (Internet Message Format) headers (RFC822, RFC5322) - pub mod address; pub mod datetime; pub mod field; @@ -8,10 +7,7 @@ pub mod mailbox; pub mod mime; pub mod trace; -use nom::{ - combinator::map, - IResult, -}; +use nom::{combinator::map, IResult}; use crate::header; use crate::imf::address::AddressRef; @@ -61,7 +57,8 @@ pub struct Imf<'a> { } impl<'a> Imf<'a> { pub fn with_kv(mut self, v: Vec>) -> Self { - self.kv = v; self + self.kv = v; + self } } @@ -94,12 +91,15 @@ impl<'a> FromIterator> 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::() + map(header::header_kv, |fields| { + fields + .iter() + .flat_map(Field::try_from) + .into_iter() + .collect::() })(input) } - #[cfg(test)] mod tests { use super::*; diff --git a/src/lib.rs b/src/lib.rs index edb327a..bd79650 100644 --- a/src/lib.rs +++ b/src/lib.rs @@ -18,7 +18,7 @@ pub mod text; /// Manipulate buffer of bytes mod pointers; -use nom::{IResult, combinator::into}; +use nom::{combinator::into, IResult}; /// Parse a whole email including its (MIME) body /// @@ -37,7 +37,7 @@ use nom::{IResult, combinator::into}; /// * `msg` - The parsed message /// /// # Examples -/// +/// /// ``` /// let input = br#"Date: 7 Mar 2023 08:00:00 +0200 /// From: deuxfleurs@example.com @@ -57,7 +57,9 @@ use nom::{IResult, combinator::into}; /// ); /// ``` pub fn parse_message(input: &[u8]) -> IResult<&[u8], part::composite::Message> { - into(part::composite::message(mime::MIME::::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 diff --git a/src/mime/field.rs b/src/mime/field.rs index 2489516..8736b47 100644 --- a/src/mime/field.rs +++ b/src/mime/field.rs @@ -45,7 +45,10 @@ impl<'a> TryFrom<&header::Field<'a>> for Content<'a> { type Error = (); fn try_from(f: &header::Field<'a>) -> Result { let content = match f { - header::Field::Good(header::Kv2(key, value)) => match key.to_ascii_lowercase().as_slice() { + 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), @@ -107,7 +110,10 @@ This is a multipart message. .as_bytes(); assert_eq!( - map(header::header_kv, |k| k.iter().flat_map(Content::try_from).collect())(fullmail), + map(header::header_kv, |k| k + .iter() + .flat_map(Content::try_from) + .collect())(fullmail), Ok(( &b"This is a multipart message.\n\n"[..], vec![ diff --git a/src/mime/mod.rs b/src/mime/mod.rs index 5b9a46a..542d083 100644 --- a/src/mime/mod.rs +++ b/src/mime/mod.rs @@ -1,4 +1,4 @@ -/// Parsed and represent an email character set +/// Parsed and represent an email character set pub mod charset; /// MIME specific headers @@ -13,17 +13,17 @@ pub mod r#type; use std::fmt; use std::marker::PhantomData; +use crate::header; use crate::imf::identification::MessageID; use crate::mime::field::Content; use crate::mime::mechanism::Mechanism; use crate::mime::r#type::{AnyType, NaiveType}; -use crate::header; use crate::text::misc_token::Unstructured; //Multipart, Message, Text, Binary}; #[derive(Debug, PartialEq, Clone)] pub struct MIME<'a, T> { - pub interpreted_type: T, - pub fields: NaiveMIME<'a> + pub interpreted_type: T, + pub fields: NaiveMIME<'a>, } impl<'a> Default for MIME<'a, r#type::DeductibleText> { fn default() -> Self { @@ -80,9 +80,8 @@ impl<'a> fmt::Debug for NaiveMIME<'a> { impl<'a> FromIterator> for NaiveMIME<'a> { fn from_iter>>(it: I) -> Self { - it.into_iter().fold( - NaiveMIME::default(), - |mut section, field| { + it.into_iter() + .fold(NaiveMIME::default(), |mut section, field| { match field { Content::Type(v) => section.ctype = Some(v), Content::TransferEncoding(v) => section.transfer_encoding = v, @@ -90,25 +89,29 @@ impl<'a> FromIterator> for NaiveMIME<'a> { Content::Description(v) => section.description = Some(v), }; section - }, - ) + }) } } impl<'a> NaiveMIME<'a> { pub fn with_kv(mut self, fields: Vec>) -> Self { - self.kv = fields; self + self.kv = fields; + self } pub fn with_raw(mut self, raw: &'a [u8]) -> Self { - self.raw = raw; self + self.raw = raw; + self } pub fn to_interpreted(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 { fn default_type() -> AnyType; } diff --git a/src/mime/type.rs b/src/mime/type.rs index a1ed74a..7dafb29 100644 --- a/src/mime/type.rs +++ b/src/mime/type.rs @@ -1,4 +1,3 @@ -use std::fmt; use nom::{ bytes::complete::tag, combinator::{map, opt}, @@ -6,11 +5,12 @@ use nom::{ sequence::{preceded, terminated, tuple}, IResult, }; +use std::fmt; use crate::mime::charset::EmailCharset; +use crate::mime::{AnyMIME, NaiveMIME, MIME}; use crate::text::misc_token::{mime_word, MIMEWord}; use crate::text::words::mime_atom; -use crate::mime::{AnyMIME, MIME, NaiveMIME}; // --------- NAIVE TYPE #[derive(PartialEq, Clone)] @@ -92,16 +92,27 @@ impl<'a> From<&'a NaiveType<'a>> for AnyType { impl<'a> AnyType { pub fn to_mime(self, fields: NaiveMIME<'a>) -> AnyMIME<'a> { - match self { - Self::Multipart(interpreted_type) => AnyMIME::Mult(MIME:: { interpreted_type, fields }), - Self::Message(interpreted_type) => AnyMIME::Msg(MIME:: { interpreted_type, fields }), - Self::Text(interpreted_type) => AnyMIME::Txt(MIME:: { interpreted_type, fields }), - Self::Binary(interpreted_type) => AnyMIME::Bin(MIME:: { interpreted_type, fields }), - } + match self { + Self::Multipart(interpreted_type) => AnyMIME::Mult(MIME:: { + interpreted_type, + fields, + }), + Self::Message(interpreted_type) => AnyMIME::Msg(MIME:: { + interpreted_type, + fields, + }), + Self::Text(interpreted_type) => AnyMIME::Txt(MIME:: { + interpreted_type, + fields, + }), + Self::Binary(interpreted_type) => AnyMIME::Bin(MIME:: { + interpreted_type, + fields, + }), + } } } - #[derive(Debug, PartialEq, Clone)] pub enum Deductible { Inferred(T), @@ -158,7 +169,8 @@ impl ToString for MultipartSubtype { Self::Parallel => "parallel", Self::Report => "report", Self::Unknown => "mixed", - }.into() + } + .into() } } impl<'a> From<&NaiveType<'a>> for MultipartSubtype { @@ -174,8 +186,6 @@ impl<'a> From<&NaiveType<'a>> for MultipartSubtype { } } - - #[derive(Debug, PartialEq, Default, Clone)] pub enum MessageSubtype { #[default] @@ -191,7 +201,8 @@ impl ToString for MessageSubtype { Self::Partial => "partial", Self::External => "external", Self::Unknown => "rfc822", - }.into() + } + .into() } } @@ -203,17 +214,25 @@ pub struct Message { impl<'a> From<&NaiveType<'a>> for Message { fn from(nt: &NaiveType<'a>) -> Self { match nt.sub.to_ascii_lowercase().as_slice() { - b"rfc822" => Self { subtype: MessageSubtype::RFC822 }, - b"partial" => Self { subtype: MessageSubtype::Partial }, - b"external" => Self { subtype: MessageSubtype::External }, - _ => Self { subtype: MessageSubtype::Unknown }, + b"rfc822" => Self { + subtype: MessageSubtype::RFC822, + }, + b"partial" => Self { + subtype: MessageSubtype::Partial, + }, + b"external" => Self { + subtype: MessageSubtype::External, + }, + _ => Self { + subtype: MessageSubtype::Unknown, + }, } } } impl From> for Message { fn from(d: Deductible) -> Self { match d { - Deductible::Inferred(t) | Deductible::Explicit(t) => t + Deductible::Inferred(t) | Deductible::Explicit(t) => t, } } } @@ -240,7 +259,7 @@ impl<'a> From<&NaiveType<'a>> for Text { impl From> for Text { fn from(d: Deductible) -> Self { match d { - Deductible::Inferred(t) | Deductible::Explicit(t) => t + Deductible::Inferred(t) | Deductible::Explicit(t) => t, } } } @@ -257,7 +276,8 @@ impl ToString for TextSubtype { match self { Self::Plain | Self::Unknown => "plain", Self::Html => "html", - }.into() + } + .into() } } impl<'a> From<&NaiveType<'a>> for TextSubtype { @@ -277,8 +297,8 @@ pub struct Binary {} mod tests { use super::*; use crate::mime::charset::EmailCharset; - use crate::text::quoted::QuotedString; use crate::mime::r#type::Deductible; + use crate::text::quoted::QuotedString; #[test] fn test_parameter() { @@ -336,7 +356,12 @@ mod tests { let (rest, nt) = naive_type(b"message/rfc822").unwrap(); 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] diff --git a/src/part/composite.rs b/src/part/composite.rs index f1bae97..fabe5f5 100644 --- a/src/part/composite.rs +++ b/src/part/composite.rs @@ -1,12 +1,12 @@ -use std::fmt; use nom::IResult; +use std::fmt; use crate::header; use crate::imf; use crate::mime; use crate::part::{self, AnyPart}; -use crate::text::boundary::{boundary, Delimiter}; use crate::pointers; +use crate::text::boundary::{boundary, Delimiter}; //--- Multipart #[derive(PartialEq)] @@ -21,8 +21,14 @@ impl<'a> fmt::Debug for Multipart<'a> { 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)) + .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() } } @@ -76,7 +82,10 @@ pub fn multipart<'a>( mime: m.clone(), children: mparts, 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()..], + ), }, )) } @@ -93,19 +102,21 @@ pub fn multipart<'a>( .into_iter() .collect::(); - let mime = mime - .with_kv(fields) - .with_raw(raw_hdrs); + let mime = mime.with_kv(fields).with_raw(raw_hdrs); (input_eom, mime) - }, + } Err(_) => (input, mime::NaiveMIME::default()), }; // interpret mime according to context let mime = match m.interpreted_type.subtype { - mime::r#type::MultipartSubtype::Digest => naive_mime.to_interpreted::().into(), - _ => naive_mime.to_interpreted::().into(), + mime::r#type::MultipartSubtype::Digest => naive_mime + .to_interpreted::() + .into(), + _ => naive_mime + .to_interpreted::() + .into(), }; // parse raw part @@ -168,7 +179,10 @@ pub fn message<'a>( let imf = imf.with_kv(headers); // interpret headers to choose a mime type - let in_mime = naive_mime.with_raw(raw_headers).to_interpreted::().into(); + let in_mime = naive_mime + .with_raw(raw_headers) + .to_interpreted::() + .into(); //--------------- // parse a part following this mime specification @@ -183,7 +197,9 @@ pub fn message<'a>( Message { mime: m.clone(), imf, - raw_part, raw_headers, raw_body, + raw_part, + raw_headers, + raw_body, child: Box::new(part), }, )) @@ -196,7 +212,7 @@ mod tests { use crate::part::discrete::Text; use crate::part::AnyPart; 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 chrono::{FixedOffset, TimeZone}; @@ -265,7 +281,7 @@ It DOES end with a linebreak. body: &b"This is implicitly typed plain US-ASCII text.\nIt does NOT end with a linebreak."[..], }), AnyPart::Txt(Text { - mime: mime::MIME { + mime: mime::MIME { interpreted_type: mime::r#type::Deductible::Explicit(mime::r#type::Text { subtype: mime::r#type::TextSubtype::Plain, charset: mime::r#type::Deductible::Explicit(mime::charset::EmailCharset::US_ASCII), diff --git a/src/part/discrete.rs b/src/part/discrete.rs index 2372c77..8cfe5c3 100644 --- a/src/part/discrete.rs +++ b/src/part/discrete.rs @@ -12,10 +12,7 @@ impl<'a> fmt::Debug for Text<'a> { fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result { fmt.debug_struct("part::Text") .field("mime", &self.mime) - .field( - "body", - &String::from_utf8_lossy(self.body), - ) + .field("body", &String::from_utf8_lossy(self.body)) .finish() } } @@ -30,10 +27,7 @@ impl<'a> fmt::Debug for Binary<'a> { fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result { fmt.debug_struct("part::Binary") .field("mime", &self.mime) - .field( - "body", - &String::from_utf8_lossy(self.body), - ) + .field("body", &String::from_utf8_lossy(self.body)) .finish() } } diff --git a/src/part/field.rs b/src/part/field.rs index 2717efd..e7fd32a 100644 --- a/src/part/field.rs +++ b/src/part/field.rs @@ -4,7 +4,10 @@ use crate::mime; pub fn split_and_build<'a>(v: &Vec>) -> (mime::NaiveMIME<'a>, imf::Imf<'a>) { let (mimev, imfv) = v.iter().fold( - (Vec::::new(), Vec::::new()), + ( + Vec::::new(), + Vec::::new(), + ), |(mut mime, mut imf), f| { if let Ok(m) = mime::field::Content::try_from(f) { mime.push(m); @@ -12,9 +15,9 @@ pub fn split_and_build<'a>(v: &Vec>) -> (mime::NaiveMIME<'a>, imf.push(i); } (mime, imf) - } + }, ); - + let fmime = mimev.into_iter().collect::(); let fimf = imfv.into_iter().collect::(); (fmime, fimf) diff --git a/src/part/mod.rs b/src/part/mod.rs index 1d91f91..cc5c4b5 100644 --- a/src/part/mod.rs +++ b/src/part/mod.rs @@ -80,18 +80,19 @@ pub fn anypart<'a>(m: AnyMIME<'a>) -> impl FnOnce(&'a [u8]) -> IResult<&'a [u8], move |input| { let part = match m { AnyMIME::Mult(a) => multipart(a)(input) - .map(|(_, multi)| - multi.into()) - .unwrap_or(AnyPart::Txt(Text { - mime: mime::MIME::::default(), - body: input, - })), - AnyMIME::Msg(a) => message(a)(input) - .map(|(_, msg)| msg.into()) + .map(|(_, multi)| multi.into()) .unwrap_or(AnyPart::Txt(Text { mime: mime::MIME::::default(), body: input, })), + AnyMIME::Msg(a) => { + message(a)(input) + .map(|(_, msg)| msg.into()) + .unwrap_or(AnyPart::Txt(Text { + mime: mime::MIME::::default(), + body: input, + })) + } AnyMIME::Txt(a) => AnyPart::Txt(Text { mime: a, body: input, @@ -101,7 +102,7 @@ pub fn anypart<'a>(m: AnyMIME<'a>) -> impl FnOnce(&'a [u8]) -> IResult<&'a [u8], body: input, }), }; - + // This function always consumes the whole input Ok((&input[input.len()..], part)) } diff --git a/src/text/whitespace.rs b/src/text/whitespace.rs index c48e347..b7777c0 100644 --- a/src/text/whitespace.rs +++ b/src/text/whitespace.rs @@ -22,7 +22,12 @@ use nom::{ /// \r or \n is allowed nowhere else, so we also add this support. 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