Struct Membership

pub struct Membership {
    pub relation_id: Option<String>,
    pub role_id: Option<String>,
    pub direction: Option<String>,
    pub mbid: String,
    pub name: String,
    pub credited_as: Option<String>,
    pub side: Side,
    pub kind: String,
    pub attributes: Vec<String>,
    pub began: Option<String>,
    pub ended: Option<String>,
    pub over: Option<bool>,
}

One spell in a band, as a source dates it.

A person can appear twice, and the second row is not a duplicate: Ozzy Osbourne was in Black Sabbath from 1968 to 1979 and again from 1997, and a list that folded the two into one span would claim he was there in 1985. MusicBrainz states them as two relations and they are kept as two rows.

Fields

relation_id: Option<String>

MusicBrainz relationship row identifier, when exposed.

role_id: Option<String>

Stable MusicBrainz relationship type identifier.

direction: Option<String>

Direction exactly as returned by MusicBrainz.

mbid: String

The other artist’s MusicBrainz identifier, which is what makes a later fetch about them an update rather than a second opinion.

name: String

The other artist’s name, as the source spells it.

credited_as: Option<String>

Exact spelling used for the related artist in this relationship.

side: Side

Which end of the relation that name is — see Side.

kind: String

How the source names the relationship, in its own words: member of band, instrumental supporting musician.

Kept verbatim and shown verbatim, the same rule KnownRelease::stated_type follows. The two are not the same fact — a founding member and a guitarist hired for one tour are both on a record and only one of them was in the band — and paraphrasing them into one word would throw away the distinction MusicBrainz took the trouble to draw.

attributes: Vec<String>

What the source lists against the relationship.

Not “instruments”, although most of them are. Measured on a real answer: Judas Priest’s line-up carries ["guitar family", "original"] on one row, where original marks an original member and is not an instrument at all. Calling this field instruments would put a lie in a column header, so it carries the source’s own word and its own values.

began: Option<String>

When it started: a year, or a fuller date when the source has one.

ended: Option<String>

When it stopped.

over: Option<bool>

Whether the source says it is over.

The same distinction ArtistFacts::active draws, and it matters more here: a membership with no end date may be a current one or one nobody has filled in, and printing “1968–” for both tells the reader something the source never said. Some(false) is “still in the band”.

Implementations

impl Membership

pub fn covers(&self, year: u32) -> Option<bool>

Whether this membership was running in a given year: Some(true), Some(false), or None where the source’s dates cannot say.

The three-way answer is the point. A membership with no start date places nobody, and one the source says has ended without saying when places nobody either — the end is somewhere, and “somewhere” cannot be compared with a year. Both come back None, so a caller has to decide what to do about them rather than being handed a false that looks like knowledge.

pub fn years(&self) -> String

The years as a reader recognises them: 1968–1979, 1968– for a current member, 1968 where the source only knows a start and does not say whether it ended.

Never invents an end. A dash trailing into nothing is a claim that the membership is still going, and it is one MusicBrainz makes with ended: false rather than with a missing date.

Trait Implementations

impl Clone for Membership

fn clone(&self) -> Membership

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Membership

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

Formats the value using the given formatter.

impl PartialEq for Membership

fn eq(&self, other: &Membership) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Eq for Membership

impl StructuralPartialEq for Membership