deserialize-colon-separated.rs

// date: 2022-04-24
// license: GPLv3 https://www.gnu.org/licenses/gpl-3.0.txt
// author: nanpuyue <nanpuyue@gmail.com> https://blog.nanpuyue.com

use std::fmt::{self, Debug, Display, Formatter};
use std::iter::Peekable;
use std::str::Lines;
use std::{error, result};

use serde::de::{self, Visitor};
use serde::forward_to_deserialize_any;
use serde::Deserialize;

const DELIM: char = ':';

type Result<T> = result::Result<T, Error>;

pub fn from_str<'a, T: Deserialize<'a>>(s: &'a str) -> Result<T> {
    T::deserialize(&mut Deserializer::from_str(s))
}

pub struct Deserializer<'de> {
    lines: Peekable<Lines<'de>>,
}

#[derive(Debug)]
pub enum Error {
    Message(String),
    KeyNotFound,
    ValueNotFound,
    InvalidValue,
    Unimplemented,
}

impl Display for Error {
    fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
        Debug::fmt(&self, f)
    }
}

impl error::Error for Error {}

impl de::Error for Error {
    fn custom<T: fmt::Display>(msg: T) -> Self {
        Self::Message(msg.to_string())
    }
}

struct MapAccess<'a, 'de>(&'a mut Deserializer<'de>);

impl<'a, 'de> de::MapAccess<'de> for MapAccess<'a, 'de> {
    type Error = Error;

    fn next_key_seed<K: de::DeserializeSeed<'de>>(&mut self, seed: K) -> Result<Option<K::Value>> {
        Ok(seed.deserialize(&mut *self.0).ok())
    }

    fn next_value_seed<V: de::DeserializeSeed<'de>>(&mut self, seed: V) -> Result<V::Value> {
        seed.deserialize(&mut *self.0)
    }
}

impl<'de> Deserializer<'de> {
    fn from_str(s: &'de str) -> Self {
        Self {
            lines: s.lines().peekable(),
        }
    }

    fn next_key(&mut self) -> Option<&'de str> {
        self.lines.peek().and_then(|&x| x.split(DELIM).next())
    }

    fn next_value(&mut self) -> Option<&'de str> {
        self.lines
            .next()
            .and_then(|x| x.split(DELIM).nth(1).map(str::trim))
    }
}

impl<'de, 'a> de::Deserializer<'de> for &'a mut Deserializer<'de> {
    type Error = Error;

    fn deserialize_any<V: Visitor<'de>>(self, _visitor: V) -> Result<V::Value> {
        Err(Error::Unimplemented)
    }

    fn deserialize_string<V: Visitor<'de>>(self, visitor: V) -> Result<V::Value> {
        self.deserialize_str(visitor)
    }

    fn deserialize_option<V: Visitor<'de>>(self, visitor: V) -> Result<V::Value> {
        visitor.visit_some(self)
    }

    fn deserialize_str<V: Visitor<'de>>(self, visitor: V) -> Result<V::Value> {
        let value = self.next_value().ok_or(Error::ValueNotFound)?;
        visitor.visit_borrowed_str(value)
    }

    fn deserialize_f64<V: Visitor<'de>>(self, visitor: V) -> Result<V::Value> {
        let value = self
            .next_value()
            .ok_or(Error::ValueNotFound)
            .and_then(|x| x.parse().map_err(|_| Error::InvalidValue))?;
        visitor.visit_f64(value)
    }

    fn deserialize_identifier<V: Visitor<'de>>(self, visitor: V) -> Result<V::Value> {
        let ident = self.next_key().ok_or(Error::KeyNotFound)?;
        visitor.visit_borrowed_str(ident)
    }

    fn deserialize_map<V: Visitor<'de>>(self, visitor: V) -> Result<V::Value> {
        visitor.visit_map(MapAccess(self))
    }

    fn deserialize_struct<V: Visitor<'de>>(
        self,
        _name: &'static str,
        _fields: &'static [&'static str],
        visitor: V,
    ) -> Result<V::Value> {
        self.deserialize_map(visitor)
    }

    forward_to_deserialize_any! {
        bool i8 i16 i32 i64 i128 u8 u16 u32 u64 u128 f32 char
        bytes byte_buf unit unit_struct newtype_struct seq tuple
        tuple_struct enum ignored_any
    }
}

#[derive(Debug, PartialEq, Deserialize)]
#[serde(rename_all = "PascalCase")]
pub struct SomeStruct<'a> {
    pub field_a: &'a str,
    pub field_b: Option<f64>,
    pub field_c: Option<f64>,
}

fn main() {
    let input = "FieldA:helloworld\nFieldB:1.234567\nFieldC:7.654321\n";
    let data = from_str::<SomeStruct>(input).unwrap();
    assert_eq!(
        dbg!(data),
        SomeStruct {
            field_a: "helloworld",
            field_b: Some(1.234567),
            field_c: Some(7.654321)
        }
    );

    let input = "FieldA:helloworld\nFieldC:7.654321\n";
    let data = from_str::<SomeStruct>(input).unwrap();
    assert_eq!(
        dbg!(data),
        SomeStruct {
            field_a: "helloworld",
            field_b: None,
            field_c: Some(7.654321)
        }
    );

    let input = "FieldB:1.234567\nFieldC:7.654321\n";
    from_str::<SomeStruct>(input).expect_err("should be error");
}