Implement parser for inline source blocks.

This commit is contained in:
Tom Alexander 2023-07-21 22:51:19 -04:00
parent e0d2bb8213
commit e8979513aa
Signed by: talexander
GPG Key ID: D3A179C9A53C0EDE
3 changed files with 187 additions and 5 deletions

View File

@ -80,7 +80,7 @@ fn header<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s
#[tracing::instrument(ret, level = "debug")]
fn header_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let context_depth = get_bracket_depth(context)
.expect("This function should only be called from inside a inline babel call header.");
.expect("This function should only be called from inside an inline babel call header.");
let text_since_context_entry = get_consumed(context_depth.position, input);
let mut current_depth = context_depth.depth;
for c in text_since_context_entry.chars() {
@ -125,7 +125,7 @@ fn argument<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'
#[tracing::instrument(ret, level = "debug")]
fn argument_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let context_depth = get_bracket_depth(context)
.expect("This function should only be called from inside a inline babel call argument.");
.expect("This function should only be called from inside an inline babel call argument.");
let text_since_context_entry = get_consumed(context_depth.position, input);
let mut current_depth = context_depth.depth;
for c in text_since_context_entry.chars() {

View File

@ -1,6 +1,24 @@
use nom::bytes::complete::tag;
use nom::bytes::complete::tag_no_case;
use nom::character::complete::anychar;
use nom::character::complete::line_ending;
use nom::character::complete::one_of;
use nom::character::complete::space0;
use nom::combinator::opt;
use nom::combinator::recognize;
use nom::combinator::verify;
use nom::multi::many_till;
use super::Context;
use crate::error::CustomError;
use crate::error::Res;
use crate::parser::util::not_yet_implemented;
use crate::parser::exiting::ExitClass;
use crate::parser::parser_context::ContextElement;
use crate::parser::parser_context::ExitMatcherNode;
use crate::parser::parser_context::InlineSourceBlockBracket;
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::InlineSourceBlock;
#[tracing::instrument(ret, level = "debug")]
@ -8,6 +26,151 @@ pub fn inline_source_block<'r, 's>(
context: Context<'r, 's>,
input: &'s str,
) -> Res<&'s str, InlineSourceBlock<'s>> {
not_yet_implemented()?;
todo!()
let (remaining, _) = tag_no_case("src_")(input)?;
let (remaining, _) = lang(context, remaining)?;
let (remaining, _header1) = opt(parser_with_context!(header)(context))(remaining)?;
let (remaining, _body) = body(context, remaining)?;
let (remaining, _) = space0(remaining)?;
let source = get_consumed(input, remaining);
Ok((remaining, InlineSourceBlock { source }))
}
#[tracing::instrument(ret, level = "debug")]
fn lang<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let parser_context =
context.with_additional_node(ContextElement::ExitMatcherNode(ExitMatcherNode {
class: ExitClass::Beta,
exit_matcher: &lang_end,
}));
let (remaining, lang) = recognize(many_till(
verify(anychar, |c| !(c.is_whitespace() || "[{".contains(*c))),
parser_with_context!(exit_matcher_parser)(&parser_context),
))(input)?;
Ok((remaining, lang))
}
#[tracing::instrument(ret, level = "debug")]
fn lang_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
recognize(one_of("[{"))(input)
}
#[tracing::instrument(ret, level = "debug")]
fn header<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let (remaining, _) = tag("[")(input)?;
let parser_context = context
.with_additional_node(ContextElement::InlineSourceBlockBracket(
InlineSourceBlockBracket {
position: input,
depth: 0,
},
))
.with_additional_node(ContextElement::ExitMatcherNode(ExitMatcherNode {
class: ExitClass::Beta,
exit_matcher: &header_end,
}));
let (remaining, header_contents) = recognize(many_till(
anychar,
parser_with_context!(exit_matcher_parser)(&parser_context),
))(remaining)?;
let (remaining, _) = tag("]")(remaining)?;
Ok((remaining, header_contents))
}
#[tracing::instrument(ret, level = "debug")]
fn header_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let context_depth = get_bracket_depth(context)
.expect("This function should only be called from inside an inline source block header.");
let text_since_context_entry = get_consumed(context_depth.position, input);
let mut current_depth = context_depth.depth;
for c in text_since_context_entry.chars() {
match c {
'[' => {
current_depth += 1;
}
']' if current_depth == 0 => {
panic!("Exceeded inline source block header bracket depth.")
}
']' if current_depth > 0 => {
current_depth -= 1;
}
_ => {}
}
}
if current_depth == 0 {
let close_bracket = tag::<&str, &str, CustomError<&str>>("]")(input);
if close_bracket.is_ok() {
return close_bracket;
}
}
line_ending(input)
}
#[tracing::instrument(ret, level = "debug")]
fn body<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let (remaining, _) = tag("{")(input)?;
let parser_context = context
.with_additional_node(ContextElement::InlineSourceBlockBracket(
InlineSourceBlockBracket {
position: input,
depth: 0,
},
))
.with_additional_node(ContextElement::ExitMatcherNode(ExitMatcherNode {
class: ExitClass::Beta,
exit_matcher: &body_end,
}));
let (remaining, body_contents) = recognize(many_till(
anychar,
parser_with_context!(exit_matcher_parser)(&parser_context),
))(remaining)?;
let (remaining, _) = tag("}")(remaining)?;
Ok((remaining, body_contents))
}
#[tracing::instrument(ret, level = "debug")]
fn body_end<'r, 's>(context: Context<'r, 's>, input: &'s str) -> Res<&'s str, &'s str> {
let context_depth = get_bracket_depth(context)
.expect("This function should only be called from inside an inline source block body.");
let text_since_context_entry = get_consumed(context_depth.position, input);
let mut current_depth = context_depth.depth;
for c in text_since_context_entry.chars() {
match c {
'{' => {
current_depth += 1;
}
'}' if current_depth == 0 => {
panic!("Exceeded inline source block body bracket depth.")
}
'}' if current_depth > 0 => {
current_depth -= 1;
}
_ => {}
}
}
if current_depth == 0 {
let close_bracket = tag::<&str, &str, CustomError<&str>>("}")(input);
if close_bracket.is_ok() {
return close_bracket;
}
}
line_ending(input)
}
#[tracing::instrument(ret, level = "debug")]
pub fn get_bracket_depth<'r, 's>(
context: Context<'r, 's>,
) -> Option<&'r InlineSourceBlockBracket<'s>> {
for node in context.iter() {
match node.get_data() {
ContextElement::InlineSourceBlockBracket(depth) => return Some(depth),
_ => {}
}
}
None
}

View File

@ -181,6 +181,19 @@ pub enum ContextElement<'r, 's> {
/// unbalanced brackets can be detected in the middle of an
/// object.
BabelHeaderBracket(BabelHeaderBracket<'s>),
/// Stores the current bracket or parenthesis depth inside an inline babel call.
///
/// Inside an inline babel call the headers must have balanced
/// parentheses () and the arguments must have balanced brackets
/// [], so this stores the amount of opening brackets subtracted
/// by the amount of closing brackets within the definition must
/// equal zero.
///
/// A reference to the position in the string is also included so
/// unbalanced brackets can be detected in the middle of an
/// object.
InlineSourceBlockBracket(InlineSourceBlockBracket<'s>),
}
pub struct ExitMatcherNode<'r> {
@ -206,6 +219,12 @@ pub struct BabelHeaderBracket<'s> {
pub depth: usize,
}
#[derive(Debug)]
pub struct InlineSourceBlockBracket<'s> {
pub position: &'s str,
pub depth: usize,
}
impl<'r> std::fmt::Debug for ExitMatcherNode<'r> {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
let mut formatter = f.debug_struct("ExitMatcherNode");