A Rust library for using the Serde serialization framework with data in IEML document format.

IEML is a data description format whose key differences are the presence of tags and unordered anchors, and the ability to describe one object in multiple documents (files). Anchors are a primitive similar to references and pointers, which allows describing self-referential data structures and avoiding unnecessary copying of document contents.

- < profession/programmer
	name: > John Doe
	language: Rust
	experience: 2

- < profession/programmer
	name: > Juan Pérez
	language: Rust
	experience: 0.5

profession/programmer.ieml

profession: Programmer
name: @name
experience: @experience
additional:
	language: @language

There are three common ways that you might find yourself needing to work with IEML data in Rust.

• As text data. An unprocessed string of IEML data that you read from a file.
• As an untyped or loosely typed representation. Maybe you want to check that some IEML data is valid before passing it on, but without knowing the structure of what it contains. Or you want to do very basic manipulations like insert a key in a particular spot.
• As a strongly typed Rust data structure. When you expect all or most of your data to conform to a particular structure and want to get real work done without IEML's structure tripping you up.

• Data structures and APIs for working with them
• Parsing from raw input to an intermediate data structure
• Implementation of Serde traits for deserialization
• Generating a set of documents from an intermediate data structure
• Implementation of Serde traits for serialization

[dependencies]
serde = "1"
serde_ieml = "0.3"

IEML data is stored in a serde_ieml::Data structure, the details of its implementation are hidden, but the library provides several structures for viewing this data. The main one is serde_ieml::View from it you can get structures for each of the supported primitives.

The document reader can be passed to the function serde_ieml::de::parse::parse_with_reader to get serde_ieml::Data, the simplest reader is &str containing the top document directly. And using the view method to get serde_ieml::View.

use serde_ieml::{Data, View, de::parse::parse_with_reader};

// Some IEML input data as a &str. Maybe this comes
// from the user.
let input = r#"
name: > John Doe
age: 43
phones:
	- +44 1234567
	- +44 2345678
"#;

// Parse the string of data into serde_ieml::Data.
let data = parse_with_reader(input).unwrap();

// Getting reference structure allowing to read
// IEML data.
let view = data.view();

// Getting an object to view the map.
let map = view.map().unwrap();

// Getting an object to view the list contained
// in the map.
let list = map.get("phones").unwrap().list().unwrap();

// The first `string` call returns a structure
// allowing to read string primitive data
assert_eq!("John Doe", map.get("name").unwrap().string().unwrap().string());
assert_eq!("+44 1234567", list.get(0).unwrap().raw().unwrap().raw());

Serde provides a powerful way of mapping IEML data into Rust data structures largely automatically.

use serde::Deserialize;

#[derive(Deserialize)]
struct Person {
    name: String,
    age: u8,
    phones: Vec<String>,
}

// Some IEML input data as a &str. Maybe this comes from the user.
let input = r#"
name: > John Doe
age: 43
phones:
	- > +44 1234567
	- > +44 2345678
"#;

// Parse the string of data into a Person object.
let p: Person = serde_ieml::from_source(input).unwrap();

// Do things just like with any other Rust data structure.
assert_eq!("John Doe", p.name);
assert_eq!("+44 1234567", p.phones[0]);

Serde will automatically interpret the input data as a Person and produce informative error messages if the layout does not conform to what a Person is expected to look like.

Any type that implements Serde's serde::Deserialize trait can be deserialized this way. This includes built-in Rust standard library types like Vec<T> and HashMap<K, V>, as well as any structs or enums annotated with #[derive(Deserialize)].

Once we have p of type Person, our IDE and the Rust compiler can help us use it correctly like they do for any other Rust code. The IDE can autocomplete field names to prevent typos, which was impossible in the [Data] representation. And the Rust compiler can check that when we write p.phones[0], then p.phones is guaranteed to be a Vec<String> so indexing into it makes sense and produces a String.

The necessary setup for using Serde's derive macros is explained on the Using derive page of the Serde site.

Serde IEML provides a serde_ieml::data::make module to build serde_ieml::Data objects with a guarantee of a valid structure.

use serde_ieml::data::{make, name::Name};
use std::convert::Infallible;

// The same position is used for all nodes to simplify the example.
let mark = Default::default();

// The type of `john` is `serde_ieml::Data`
let (john, _) = make::make::<_, Infallible, _>(
    mark, 
    make::map(mark, |token| {
        let (token, _) = token.add(
            mark, 
            Name::new("name").unwrap(), 
            make::string(mark, (), "John Doe")
        )?;
        let (token, _) = token.add(
            mark, 
            Name::new("age").unwrap(), 
            make::raw(mark, (), "43")
        )?;
        let (token, _) = token.add(
            mark, 
            Name::new("phones").unwrap(), 
            make::list(
                mark, 
                |token| {
                    let (token, _) = token.add(
                        make::raw(mark, (), "+44 1234567")
                    )?;
                    let (token, _) = token.add(
                        make::raw(mark, (), "+44 2345678")
                    )?;
                    Ok((token, ()))
                }
            )
        )?;
        Ok((token, ()))
    })
).unwrap();

// Getting reference structure allowing to read
// IEML data.
let view = john.view();

// Getting an object to view the map.
let map = view.map().unwrap();

assert_eq!("John Doe", map.get("name").unwrap().string().unwrap().string());

All serde_ieml::data::make module functions except make and make_document return a lambda accepting a value of type Token, all token types describe the possibility or impossibility (if it is a used or error token) to add exactly one or more nodes to a particular IEML location.

The main purpose of this system is to allow you to write your own deserializers of IEML from other formats. This system avoids a number of errors, such as cyclic nodes or closed sections of a structure to which access is lost. The need for this system arises because IEML differs from other formats by its non-tree structure.