organic/src/parser/document.rs

161 lines
5.0 KiB
Rust
Raw Normal View History

2023-03-23 23:35:32 +00:00
use nom::branch::alt;
use nom::bytes::complete::tag;
2023-03-23 23:53:20 +00:00
use nom::character::complete::line_ending;
use nom::character::complete::space1;
2023-03-23 23:35:32 +00:00
use nom::combinator::eof;
use nom::combinator::recognize;
use nom::multi::many1;
2023-03-23 23:35:32 +00:00
use nom::multi::many1_count;
use nom::sequence::tuple;
use crate::parser::error::CustomError;
use crate::parser::error::MyError;
2023-03-23 23:53:20 +00:00
use crate::parser::object::standard_set_object;
2023-03-23 23:35:32 +00:00
use crate::parser::parser_context::ChainBehavior;
2023-03-23 21:59:39 +00:00
use crate::parser::parser_context::ContextElement;
use crate::parser::parser_context::ContextTree;
2023-03-23 23:35:32 +00:00
use crate::parser::parser_context::ExitMatcherNode;
2023-03-23 21:59:39 +00:00
2023-03-23 21:51:49 +00:00
use super::element::Element;
2023-03-23 21:59:39 +00:00
use super::error::Res;
2023-03-23 23:35:32 +00:00
use super::parser_with_context::parser_with_context;
2023-03-23 21:51:49 +00:00
use super::source::Source;
2023-03-23 23:35:32 +00:00
use super::Context;
2022-10-15 00:17:48 +00:00
2023-03-23 21:51:49 +00:00
#[derive(Debug)]
pub struct Document<'s> {
pub source: &'s str,
pub zeroth_section: Option<Section<'s>>,
pub children: Vec<Heading<'s>>,
}
#[derive(Debug)]
pub struct Heading<'s> {
pub source: &'s str,
pub children: Vec<DocumentElement<'s>>,
}
#[derive(Debug)]
pub struct Section<'s> {
pub source: &'s str,
pub children: Vec<Element<'s>>,
}
#[derive(Debug)]
pub enum DocumentElement<'s> {
Heading(Heading<'s>),
Section(Section<'s>),
}
2023-03-23 21:51:49 +00:00
impl<'s> Source<'s> for Document<'s> {
fn get_source(&'s self) -> &'s str {
self.source
}
}
2022-12-18 09:22:28 +00:00
2023-03-23 21:51:49 +00:00
impl<'s> Source<'s> for DocumentElement<'s> {
fn get_source(&'s self) -> &'s str {
match self {
DocumentElement::Heading(obj) => obj.source,
DocumentElement::Section(obj) => obj.source,
}
}
}
2023-03-23 21:59:39 +00:00
pub fn document(input: &str) -> Res<&str, Document> {
let initial_context: ContextTree<'_, '_> = ContextTree::new();
let document_context =
initial_context.with_additional_node(ContextElement::DocumentRoot(input));
todo!()
}
2023-03-23 23:35:32 +00:00
fn section<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, Section<'s>> {
// TODO: The zeroth section is specialized so it probably needs its own parser
let parser_context = context
.with_additional_node(ContextElement::ExitMatcherNode(ExitMatcherNode {
exit_matcher: ChainBehavior::AndParent(Some(&section_end)),
}))
.with_additional_node(ContextElement::Context("section"));
todo!()
}
fn section_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let heading_matcher = parser_with_context!(heading)(context);
alt((recognize(heading_matcher), eof))(input)
}
fn heading<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, Heading<'s>> {
2023-03-23 23:53:20 +00:00
todo!()
}
fn headline<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
2023-03-23 23:35:32 +00:00
let document_root = context.get_document_root().unwrap();
let preceding_character = get_one_before(document_root, input)
.map(|slice| slice.chars().next())
.flatten();
match preceding_character {
Some('\n') => {}
2023-03-23 23:53:20 +00:00
Some(_) => {
2023-03-23 23:35:32 +00:00
// Not at start of line, cannot be a heading
return Err(nom::Err::Error(CustomError::MyError(MyError(
"Heading not at start of line",
))));
}
// If None, we are at the start of the file which allows for headings
None => {}
};
2023-03-23 23:53:20 +00:00
let parser_context =
context.with_additional_node(ContextElement::ExitMatcherNode(ExitMatcherNode {
exit_matcher: ChainBehavior::AndParent(Some(&headline_end)),
}));
let standard_set_object_matcher = parser_with_context!(standard_set_object)(&parser_context);
2023-03-23 23:53:20 +00:00
let foo = tuple((
2023-03-23 23:35:32 +00:00
many1_count(tag("*")),
2023-03-23 23:53:20 +00:00
space1,
many1(standard_set_object_matcher),
alt((line_ending, eof)),
2023-03-23 23:35:32 +00:00
))(input)?;
todo!()
}
2023-03-23 23:53:20 +00:00
fn headline_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
line_ending(input)
}
2023-03-23 23:35:32 +00:00
fn get_one_before<'s>(document: &'s str, current_position: &'s str) -> Option<&'s str> {
assert!(is_slice_of(document, current_position));
if document.as_ptr() as usize == current_position.as_ptr() as usize {
return None;
}
let offset = current_position.as_ptr() as usize - document.as_ptr() as usize;
let previous_character_offset = document.floor_char_boundary(offset - 1);
Some(&document[previous_character_offset..offset])
}
fn is_slice_of(parent: &str, child: &str) -> bool {
let parent_start = parent.as_ptr() as usize;
let parent_end = parent_start + parent.len();
let child_start = child.as_ptr() as usize;
let child_end = child_start + child.len();
child_start >= parent_start && child_end <= parent_end
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn get_one_before_unicode() {
let input = "🧡💛💚💙💜";
let (green_heart_index, _) = input.char_indices().skip(2).next().unwrap();
let starting_with_green_heart = &input[green_heart_index..];
let yellow_heart = get_one_before(input, starting_with_green_heart).unwrap();
assert!(is_slice_of(input, yellow_heart));
assert_eq!(yellow_heart, "💛");
}
}