3 releases
0.1.3 | Aug 11, 2024 |
---|---|
0.1.2 | Sep 6, 2023 |
0.1.1 |
|
0.1.0 | Jul 15, 2023 |
#1960 in Encoding
71 downloads per month
Used in 4 crates
(via wagon-lexer)
21KB
408 lines
Logos Display
A set of derive macros that automatically implement Display
and Debug
for an enum, based on the Logos token
and regex
attributes. Usable in no_std
contexts.
How To Use
Simply use logos_display::Display
and/or use logos_display::Debug
and add it to your derives, like so:
use logos_display::{Display, Debug}
#[derive(Display, Debug, Logos)]
enum A {
#[token("{")]
LCur,
#[regex("[a-z]")]
Lower
}
|
V
impl core::fmt::Display for A {
fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
use core::fmt::Write;
match &self {
A::LCur => write!(f, "{}", "{"),
A::Lower => write!(f, "{}", "[a-z]"),
}
}
}
Difference between Display
and Debug
If the enum variant is a unit type, there is no difference. But in the case where the variant is a tuple or struct variant, the Debug
version will also show the inner value held by the instance, whereas the Display
version will only output the name of the outer layer. Like so:
use logos_display::{Display, Debug}
#[derive(Display, Debug, Logos)]
enum A {
#[token("{")]
LCur,
#[regex("[a-z]", |lex| some_func(lex.slice()))]
Lower(TypeOne, TypeTwo)
}
|
V
impl core::fmt::Debug for A {
fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
use core::fmt::Write;
match &self {
A::LCur => write!(f, "{}", "{"),
A::Lower(_arg1, _arg2) => write!(f, "{}{:?}", "[a-z]", vec![_arg1, _arg2]),
}
}
}
This does of course require the inner types to implement Debug
in some form as well.
Dealing with non-tokens
In the case that a variant is not a token or regex, the name of the variant will be used for the Display method (so variant B
will have "B"
as it's string representation). If you want to override any of this functionality, you can add an display_override("string")
attribute to the variant as follows:
use logos_display::Display
#[derive(Display, Logos)]
enum A {
#[display_override("fancy curly thing")]
#[token("{")]
LCur,
#[regex("[a-z]")]
Lower
}
|
V
impl core::fmt::Display for A {
fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
use core::fmt::Write;
match &self {
A::LCur => write!(f, "{}", "fancy curly thing"),
A::Lower => write!(f, "{}", "[a-z]"),
}
}
}
Multiple tokens
When a variant accepts multiple tokens, by default, they will be concatenated using /
in the string representation, like so:
use logos_display::Display
#[derive(Display, Logos)]
enum A {
#[token("{")]
#[token("}")]
Cur
}
|
V
impl core::fmt::Display for A {
fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
use core::fmt::Write;
match &self {
A::LCur => write!(f, "{}", "{/}"),
}
}
}
This functionality can be overwritten using the display_concat("string")
attribute on the original enum:
use logos_display::Display
#[derive(Display, Logos)]
#[display_concat(" or ")]
enum A {
#[token("{")]
#[token("}")]
Cur
}
|
V
impl core::fmt::Display for A {
fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
use core::fmt::Write;
match &self {
A::LCur => write!(f, "{}", "{ or }"),
}
}
}
Additionally, you can pass None
to this attribute in order to disable concatonation. In this case, the token that is encountered last will be used:
use logos_display::Display
#[derive(Display, Logos)]
#[display_concat(None)]
enum A {
#[token("{")]
#[token("}")]
Cur
}
|
V
impl core::fmt::Display for A {
fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
use core::fmt::Write;
match &self {
A::LCur => write!(f, "{}", "}"),
}
}
}
Dependencies
~245–690KB
~17K SLoC