#![forbid(unsafe_code)]#![deny(rust_2018_idioms, missing_docs)]//! [Parse][parse()] .mailmap files as used in git repositories and remap names and emails//! using an [accelerated data-structure][Snapshot]. use bstr::BStr; ///pub mod parse; /// Parse the given `buf` of bytes line by line into mapping [Entries][Entry].////// Errors may occur per line, but it's up to the caller to stop iteration when/// one is encountered.pub fn parse(buf: &[u8]) -> parse::Lines<'_> {    parse::Lines::new(buf)} /// Similar to [parse()], but will skip all lines that didn't parse correctly, silently squelching all errors.pub fn parse_ignore_errors(buf: &[u8]) -> impl Iterator<Item = Entry<'_>> {    parse(buf).filter_map(Result::ok)} mod entry; ///pub mod snapshot; /// A data-structure to efficiently store a list of entries for optimal, case-insensitive lookup by email and/// optionally name to find mappings to new names and/or emails.////// The memory layout is efficient, even though lots of small allocations are performed to store strings of emails and names.#[derive(Default, Clone)]pub struct Snapshot {    /// Sorted by `old_email`    entries_by_old_email: Vec<snapshot::EmailEntry>,} /// An typical entry of a mailmap, which always contains an `old_email` by which/// the mapping is performed to replace the given `new_name` and `new_email`.////// Optionally, `old_name` is also used for lookup.////// Typically created by [parse()].#[derive(PartialEq, Eq, Debug, Hash, Ord, PartialOrd, Clone, Copy, Default)]#[cfg_attr(feature = "serde1", derive(serde::Serialize, serde::Deserialize))]pub struct Entry<'a> {    #[cfg_attr(feature = "serde1", serde(borrow))]    /// The name to map to.    pub(crate) new_name: Option<&'a BStr>,    /// The email map to.    pub(crate) new_email: Option<&'a BStr>,    /// The name to look for and replace.    pub(crate) old_name: Option<&'a BStr>,    /// The email to look for and replace.    pub(crate) old_email: &'a BStr,}