organic/src/parser/subscript_and_superscript.rs

use nom::branch::alt;
use nom::bytes::complete::tag;
use nom::character::complete::anychar;
use nom::character::complete::one_of;
use nom::character::complete::space0;
use nom::combinator::map;
use nom::combinator::not;
use nom::combinator::opt;
use nom::combinator::peek;
use nom::combinator::recognize;
use nom::combinator::verify;
use nom::multi::many_till;

use super::org_source::OrgSource;
use super::Context;
use super::Object;
use crate::error::CustomError;
use crate::error::MyError;
use crate::error::Res;
use crate::parser::exiting::ExitClass;
use crate::parser::object_parser::standard_set_object;
use crate::parser::parser_context::ContextElement;
use crate::parser::parser_context::ExitMatcherNode;
use crate::parser::parser_context::SubscriptSuperscriptBrace;
use crate::parser::parser_with_context::parser_with_context;
use crate::parser::util::exit_matcher_parser;
use crate::parser::util::get_consumed;
use crate::parser::Subscript;
use crate::parser::Superscript;

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
pub fn subscript<'r, 's>(
    context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, Subscript<'s>> {
    // We check for the underscore first before checking the pre-character as a minor optimization to avoid walking up the context tree to find the document root unnecessarily.
    let (remaining, _) = tag("_")(input)?;
    pre(context, input)?;
    let (remaining, _body) = script_body(context, remaining)?;
    let (remaining, _) = space0(remaining)?;
    let source = get_consumed(input, remaining);
    Ok((
        remaining,
        Subscript {
            source: source.into(),
        },
    ))
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
pub fn superscript<'r, 's>(
    context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, Superscript<'s>> {
    // We check for the circumflex first before checking the pre-character as a minor optimization to avoid walking up the context tree to find the document root unnecessarily.
    let (remaining, _) = tag("^")(input)?;
    pre(context, input)?;
    let (remaining, _body) = script_body(context, remaining)?;
    let (remaining, _) = space0(remaining)?;
    let source = get_consumed(input, remaining);
    Ok((
        remaining,
        Superscript {
            source: source.into(),
        },
    ))
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn pre<'r, 's>(_context: Context<'r, 's>, input: OrgSource<'s>) -> Res<OrgSource<'s>, ()> {
    let preceding_character = input.get_preceding_character();
    match preceding_character {
        Some(c) if !c.is_whitespace() => {}
        _ => {
            return Err(nom::Err::Error(CustomError::MyError(MyError(
                "Must be preceded by a non-whitespace character.".into(),
            ))));
        }
    };
    Ok((input, ()))
}

#[derive(Debug)]
enum ScriptBody<'s> {
    Braceless(&'s str),
    WithBraces(Vec<Object<'s>>),
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn script_body<'r, 's>(
    context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, ScriptBody<'s>> {
    alt((
        map(parser_with_context!(script_asterisk)(context), |body| {
            ScriptBody::Braceless(body.into())
        }),
        map(parser_with_context!(script_alphanum)(context), |body| {
            ScriptBody::Braceless(body.into())
        }),
        map(parser_with_context!(script_with_braces)(context), |body| {
            ScriptBody::WithBraces(body.into())
        }),
    ))(input)
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn script_asterisk<'r, 's>(
    _context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, OrgSource<'s>> {
    tag("*")(input)
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn script_alphanum<'r, 's>(
    context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, OrgSource<'s>> {
    let (remaining, _sign) = opt(recognize(one_of("+-")))(input)?;
    let (remaining, _script) = many_till(
        parser_with_context!(script_alphanum_character)(context),
        parser_with_context!(end_script_alphanum_character)(context),
    )(remaining)?;
    let source = get_consumed(input, remaining);
    Ok((remaining, source))
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn script_alphanum_character<'r, 's>(
    _context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, OrgSource<'s>> {
    recognize(verify(anychar, |c| {
        c.is_alphanumeric() || r#",.\"#.contains(*c)
    }))(input)
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn end_script_alphanum_character<'r, 's>(
    context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, OrgSource<'s>> {
    let (remaining, final_char) = recognize(verify(anychar, |c| c.is_alphanumeric()))(input)?;
    peek(not(parser_with_context!(script_alphanum_character)(
        context,
    )))(remaining)?;
    Ok((remaining, final_char))
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn script_with_braces<'r, 's>(
    context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, Vec<Object<'s>>> {
    let (remaining, _) = tag("{")(input)?;
    let parser_context = context
        .with_additional_node(ContextElement::SubscriptSuperscriptBrace(
            SubscriptSuperscriptBrace {
                position: remaining.into(),
                depth: 0,
            },
        ))
        .with_additional_node(ContextElement::ExitMatcherNode(ExitMatcherNode {
            class: ExitClass::Gamma,
            exit_matcher: &script_with_braces_end,
        }));

    let (remaining, (children, _exit_contents)) = many_till(
        parser_with_context!(standard_set_object)(&parser_context),
        parser_with_context!(exit_matcher_parser)(&parser_context),
    )(remaining)?;

    let (remaining, _) = tag("}")(remaining)?;
    Ok((remaining, children))
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn script_with_braces_end<'r, 's>(
    context: Context<'r, 's>,
    input: OrgSource<'s>,
) -> Res<OrgSource<'s>, OrgSource<'s>> {
    let context_depth = get_bracket_depth(context)
        .expect("This function should only be called from inside a subscript or superscript.");
    let text_since_context_entry = get_consumed(context_depth.position, input);
    let mut current_depth = context_depth.depth;
    for c in Into::<&str>::into(text_since_context_entry).chars() {
        match c {
            '{' => {
                current_depth += 1;
            }
            '}' if current_depth == 0 => {
                panic!("Exceeded subscript or superscript brace depth.")
            }
            '}' if current_depth > 0 => {
                current_depth -= 1;
            }
            _ => {}
        }
    }
    if current_depth == 0 {
        let close_bracket = tag::<&str, OrgSource<'_>, CustomError<OrgSource<'_>>>("}")(input);
        if close_bracket.is_ok() {
            return close_bracket;
        }
    }
    return Err(nom::Err::Error(CustomError::MyError(MyError(
        "Not a valid end for subscript or superscript.".into(),
    ))));
}

#[cfg_attr(feature = "tracing", tracing::instrument(ret, level = "debug"))]
fn get_bracket_depth<'r, 's>(
    context: Context<'r, 's>,
) -> Option<&'r SubscriptSuperscriptBrace<'s>> {
    for node in context.iter() {
        match node.get_data() {
            ContextElement::SubscriptSuperscriptBrace(depth) => return Some(depth),
            _ => {}
        }
    }
    None
}