Struct MessagePart 

pub struct MessagePart<'m> { /* private fields */ }

A MIME part: a header block and a body. Parts are numbered in document order across nested messages (MessagePart::id), and every message has a root part holding its own header fields.

Bodies are decoded on access: MessagePart::decoded for the transfer encoding, MessagePart::text for text parts (transfer encoding and charset), MessagePart::text_prefix and MessagePart::decoded_len to avoid decoding the whole body.

Implementations

impl<'m> MessagePart<'m>

pub fn id(&self) -> PartId

The id of this part in crate::Message::parts: parts of every message are numbered in document order.

pub fn message(&self) -> MessageRef<'m>

The message this part belongs to.

pub fn parent(&self) -> Option<MessagePart<'m>>

The enclosing multipart; None for the root part of a message.

pub fn kind(&self) -> PartKind<'m>

What the part contains, from its Content-Type (and Content-Disposition for PartKind::InlineBinary).

pub fn is_text(&self) -> bool

Whether the part is PartKind::Text or PartKind::Html.

pub fn is_multipart(&self) -> bool

Whether the part is PartKind::Multipart.

pub fn is_message(&self) -> bool

Whether the part is PartKind::Message.

pub fn children( &self, ) -> impl ExactSizeIterator<Item = MessagePart<'m>> + DoubleEndedIterator + use<'m>

Children of a multipart, in order.

pub fn nested(&self) -> Option<MessageRef<'m>>

The nested message of a message/rfc822 part.

pub fn headers(&self) -> Headers<'m>

The header fields of this part.

pub fn content_type(&self) -> Option<ContentType<'m>>

The last Content-Type field, when it parsed.

pub fn content_disposition(&self) -> Option<ContentType<'m>>

The last Content-Disposition field, when it parsed.

pub fn attachment_name(&self) -> Option<&'m str>

The Content-Disposition filename, else the Content-Type name.

pub fn is_content_type(&self, type_: &str, subtype: &str) -> bool

Whether the Content-Type is type_/subtype (ASCII case-insensitive).

pub fn content_id(&self) -> Option<&'m str>

The Content-ID field, without angle brackets.

pub fn content_description(&self) -> Option<&'m str>

The Content-Description field, decoded.

pub fn content_location(&self) -> Option<&'m str>

The Content-Location field.

pub fn content_language(&self) -> Option<TextList<'m>>

The languages of the Content-Language field.

pub fn content_transfer_encoding(&self) -> Option<&'m str>

The Content-Transfer-Encoding field as text: unfolded, trimmed, with encoded words decoded (the crate::HeaderForm::Text form), in the case it was written in. MessagePart::encoding gives the transfer encoding the body is decoded with.

pub fn encoding(&self) -> Encoding

The declared transfer encoding: base64, quoted-printable, or none (7bit, 8bit, binary, no field, or an unknown value).

pub fn flags(&self) -> PartFlags

What happened while the part was parsed.

pub fn has_problems(&self) -> bool

Whether the structure of this part was damaged: any of its PartFlags is set (this replaces 0.11’s is_encoding_problem). Problems found while decoding the body are reported by MessagePart::decoded_checked and MessagePart::text_checked.

pub fn in_text_body(&self) -> bool

Whether the part is in the text body of its message (MessageRef::text_body).

pub fn in_html_body(&self) -> bool

Whether the part is in the HTML body of its message (MessageRef::html_body).

pub fn is_attachment(&self) -> bool

Whether the part is in the attachments of its message (MessageRef::attachments).

pub fn is_inline(&self) -> bool

Whether the part is marked to be rendered inside the message rather than offered as a download (issue #70): Content-Disposition: inline, or no disposition on a child of multipart/related other than the first (the target of a cid: reference). Content-Disposition: attachment is never inline.

This answers the disposition question only. It does not say whether the part is shown in a body list or offered as an attachment; see MessagePart::role and the list membership accessors for that.

pub fn role(&self) -> PartRole

The classifier’s decision for this part. A part copied into the other body list of a multipart/alternative is reported as PartRole::CopiedBody, although it is also in its own list.

pub fn boundary(&self) -> Option<&'m str>

The boundary parameter of a multipart part.

pub fn delimiter(&self) -> Option<&'m str>

The boundary of the enclosing multipart, whose delimiter starts this part.

pub fn raw(&self) -> &'m [u8]

Headers and body, excluding the line break before the next delimiter.

pub fn raw_headers(&self) -> &'m [u8]

The header block, from the first header to the start of the body (the blank line included). The part ends, like any part, before the line break that precedes its delimiter; when that delimiter cuts the header block (PartFlags::NO_BLANK_LINE), or ends the part and its container right after the blank line, the header block stops there too and excludes that line break, which the last header field (crate::Header::offset_end) or the blank line still covers.

pub fn raw_body(&self) -> &'m [u8]

The body as written (still transfer-encoded), excluding the line break before the next delimiter.

pub fn offset_header(&self) -> u32

Offset of the first header of this part in the source buffer of its message (MessageRef::source).

pub fn offset_body(&self) -> u32

Offset of the body of this part in the source buffer of its message.

pub fn offset_end(&self) -> u32

Offset of the end of this part in the source buffer of its message.

pub fn decoded(&self) -> Cow<'m, [u8]>

The transfer-decoded body; borrowed when nothing needs decoding. Malformed input is decoded on a best-effort basis: bytes outside the base64 alphabet are skipped, malformed quoted-printable escapes are kept as written. MessagePart::decoded_checked also reports those problems.

pub fn decode_into(&self, out: &mut Vec<u8>)

Appends the transfer-decoded body to out, for reusing a buffer in a loop.

pub fn decoded_len(&self) -> usize

Length of the transfer-decoded body. Base64 and quoted-printable bodies are measured without being decoded, except base64 with bytes outside the alphabet.

pub fn decoded_checked(&self) -> (Cow<'m, [u8]>, DecodeProblems)

MessagePart::decoded together with the problems found while decoding: malformed base64 or quoted-printable, or an unknown Content-Transfer-Encoding. The bytes are the ones decoded() returns; decoded() stays the faster call when the problems are not needed.

pub fn has_known_transfer_encoding(&self) -> bool

False when the part declares a Content-Transfer-Encoding other than 7bit, 8bit, binary, base64 or quoted-printable, the condition DecodeProblems::UNKNOWN_TRANSFER_ENCODING reports, without decoding the body. True when the field is absent.

pub fn text(&self) -> Option<Cow<'m, str>>

The text of a text or HTML part: transfer-decoded, then converted from its charset (UTF-8 when the charset is missing or unknown; byte sequences that are not valid in the charset become U+FFFD). Borrowed when the body needs no conversion. None for other parts. MessagePart::text_checked also reports decoding problems.

pub fn text_checked(&self) -> Option<(Cow<'m, str>, DecodeProblems)>

MessagePart::text together with the problems found while decoding it: those of MessagePart::decoded_checked, an unknown charset, and byte sequences that are not valid in the charset. The text is the one text() returns. None for parts that are not text.

For JMAP, isEncodingProblem is !problems.is_empty().

pub fn text_prefix(&self, max_chars: usize) -> Option<Cow<'m, str>>

The first max_chars characters of MessagePart::text, decoding only as much of the body as they need. None for other parts.

pub fn text_prefix_bytes(&self, max_bytes: usize) -> Option<Cow<'m, str>>

MessagePart::text cut at str::floor_char_boundary of max_bytes: the longest prefix that is at most max_bytes bytes long and ends on a character boundary. None for other parts.

pub fn text_into(&self, out: &mut String) -> bool

Appends the text of a text or HTML part to out; returns false for other parts.

pub fn html_utf8(&self) -> Option<Cow<'m, str>>

The HTML of an HTML part with its <meta> charset declarations rewritten to UTF-8 (issue #109), so that a browser reading the decoded text does not apply the original charset a second time. None for other parts.

Trait Implementations

impl<'m> Clone for MessagePart<'m>

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<'m> Copy for MessagePart<'m>

impl Debug for MessagePart<'_>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.