Implement a basic paragraph parser.

This commit is contained in:
Tom Alexander 2023-03-25 11:22:59 -04:00
parent b88365e7eb
commit d582c8603a
Signed by: talexander
GPG Key ID: D3A179C9A53C0EDE
3 changed files with 51 additions and 3 deletions

View File

@ -155,7 +155,7 @@ fn headline<'r, 's>(
} }
fn headline_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> { fn headline_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
line_ending(input) alt((line_ending, eof))(input)
} }
/// Check that we are at the start of a line /// Check that we are at the start of a line
@ -199,7 +199,7 @@ fn is_slice_of(parent: &str, child: &str) -> bool {
} }
/// Get a slice of the string that was consumed in a parser using the original input to the parser and the remaining input after the parser. /// Get a slice of the string that was consumed in a parser using the original input to the parser and the remaining input after the parser.
fn get_consumed<'s>(input: &'s str, remaining: &'s str) -> &'s str { pub fn get_consumed<'s>(input: &'s str, remaining: &'s str) -> &'s str {
assert!(is_slice_of(input, remaining)); assert!(is_slice_of(input, remaining));
let source = { let source = {
let offset = remaining.as_ptr() as usize - input.as_ptr() as usize; let offset = remaining.as_ptr() as usize - input.as_ptr() as usize;

View File

@ -1,4 +1,19 @@
use nom::branch::alt;
use nom::character::complete::line_ending;
use nom::character::complete::space0;
use nom::combinator::eof;
use nom::combinator::map;
use nom::combinator::not; use nom::combinator::not;
use nom::combinator::recognize;
use nom::multi::many1;
use nom::sequence::tuple;
use crate::parser::document::get_consumed;
use crate::parser::object::standard_set_object;
use crate::parser::parser_context::ChainBehavior;
use crate::parser::parser_context::ContextElement;
use crate::parser::parser_context::ExitMatcherNode;
use crate::parser::parser_with_context::parser_with_context;
use super::error::Res; use super::error::Res;
use super::greater_element::PlainList; use super::greater_element::PlainList;
@ -23,5 +38,35 @@ impl<'s> Source<'s> for Element<'s> {
pub fn element<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, Element<'s>> { pub fn element<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, Element<'s>> {
not(|i| context.check_exit_matcher(i))(input)?; not(|i| context.check_exit_matcher(i))(input)?;
todo!()
let paragraph_matcher = parser_with_context!(paragraph)(context);
map(paragraph_matcher, Element::Paragraph)(input)
}
fn paragraph<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, Paragraph<'s>> {
let parser_context =
context.with_additional_node(ContextElement::ExitMatcherNode(ExitMatcherNode {
exit_matcher: ChainBehavior::AndParent(Some(&paragraph_end)),
}));
let standard_set_object_matcher = parser_with_context!(standard_set_object)(&parser_context);
let (remaining, children) = many1(standard_set_object_matcher)(input)?;
let source = get_consumed(input, remaining);
Ok((remaining, Paragraph { source, children }))
}
fn paragraph_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
// TODO: Other elements should also end paragraphs
alt((recognize(tuple((line_ending, many1(blank_line)))), eof))(input)
}
/// A line containing only whitespace and then a line break
///
/// It is up to the caller to ensure this is called at the start of a line.
fn blank_line(input: &str) -> Res<&str, &str> {
not(eof)(input)?;
recognize(tuple((space0, alt((line_ending, eof)))))(input)
} }

View File

@ -1,4 +1,7 @@
use super::object::Object;
#[derive(Debug)] #[derive(Debug)]
pub struct Paragraph<'s> { pub struct Paragraph<'s> {
pub source: &'s str, pub source: &'s str,
pub children: Vec<Object<'s>>,
} }