First version of tokenizer construct

This commit is contained in:
Lukas Stancik
2026-04-04 07:44:43 +00:00
parent 5d3afabd24
commit d12ef37dd5
2 changed files with 132 additions and 0 deletions
+1
View File
@@ -194,5 +194,6 @@ pub mod partial_whitespace;
pub mod raw_flow;
pub mod raw_text;
pub mod string;
pub mod task;
pub mod text;
pub mod thematic_break;
+131
View File
@@ -0,0 +1,131 @@
use crate::construct::partial_space_or_tab::space_or_tab;
use crate::event::Name;
use crate::state::{Name as StateName, State};
use crate::tokenizer::Tokenizer;
pub fn start(tokenizer: &mut Tokenizer) -> State {
if matches!(tokenizer.current, Some(b':'))
{
log::debug!("---start--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
tokenizer.tokenize_state.marker = tokenizer.current.unwrap();
tokenizer.enter(Name::Task);
tokenizer.enter(Name::TaskFence);
tokenizer.enter(Name::TaskSequence);
State::Retry(StateName::TaskOpenSequence)
} else {
State::Nok
}
}
pub fn open_sequence(tokenizer: &mut Tokenizer) -> State {
log::debug!("---open_sequence--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
if tokenizer.current == Some(tokenizer.tokenize_state.marker) {
tokenizer.tokenize_state.size += 1;
tokenizer.consume();
State::Next(StateName::TaskOpenSequence)
} else if tokenizer.tokenize_state.size == 2 {
tokenizer.tokenize_state.size = 0;
tokenizer.exit(Name::TaskSequence);
tokenizer.exit(Name::TaskFence);
if matches!(tokenizer.current, Some(b'\t' | b' ')) {
tokenizer.attempt(State::Next(StateName::TaskContentStart), State::Nok);
State::Retry(space_or_tab(tokenizer))
} else {
State::Retry(StateName::TaskContentStart)
}
} else {
tokenizer.tokenize_state.marker = 0;
tokenizer.tokenize_state.size = 0;
State::Nok
}
}
pub fn content_start(tokenizer: &mut Tokenizer) -> State {
log::debug!("---content_start--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
match tokenizer.current {
None | Some(b':') => {
State::Retry(StateName::TaskContentEnd)
},
Some(_) => {
tokenizer.enter(Name::TaskContent);
State::Retry(StateName::TaskContentInside)
}
}
}
pub fn content_inside(tokenizer: &mut Tokenizer) -> State {
log::debug!("---content_inside--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
match tokenizer.current {
None => {
tokenizer.tokenize_state.marker = 0;
State::Nok
}
Some(b':') => {
tokenizer.exit(Name::TaskContent);
State::Retry(StateName::TaskContentEnd)
}
Some(_) => {
tokenizer.consume();
State::Next(StateName::TaskContentInside)
}
}
}
pub fn content_end(tokenizer: &mut Tokenizer) -> State {
log::debug!("---content_end--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
match tokenizer.current {
None => {
tokenizer.tokenize_state.marker = 0;
State::Nok
}
Some(b':') => {
State::Retry(StateName::TaskCloseStart)
}
Some(_) => {
unreachable!("expected eof/eol")
}
}
}
pub fn close_start(tokenizer: &mut Tokenizer) -> State {
log::debug!("---close_start--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
if tokenizer.current == Some(tokenizer.tokenize_state.marker) {
tokenizer.enter(Name::TaskFence);
tokenizer.enter(Name::TaskSequence);
State::Retry(StateName::TaskCloseSequence)
} else {
State::Nok
}
}
pub fn close_sequence(tokenizer: &mut Tokenizer) -> State {
log::debug!("---close_sequence--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
if tokenizer.current == Some(tokenizer.tokenize_state.marker) {
tokenizer.tokenize_state.size += 1;
tokenizer.consume();
State::Next(StateName::TaskCloseSequence)
} else if tokenizer.tokenize_state.size == 1 {
log::debug!("else if");
tokenizer.tokenize_state.size = 0;
tokenizer.exit(Name::TaskSequence);
State::Retry(StateName::TaskCloseAfter)
} else {
log::debug!("else");
tokenizer.tokenize_state.size = 0;
State::Nok
}
}
pub fn close_after(tokenizer: &mut Tokenizer) -> State {
log::debug!("---close_after--- {:?} {:?}", tokenizer.tokenize_state.size, tokenizer.current.unwrap() as char);
match tokenizer.current {
None => State::Nok,
_ => {
tokenizer.exit(Name::TaskFence);
tokenizer.exit(Name::Task);
State::Ok
}
}
}