aboutsummaryrefslogtreecommitdiffstats
path: root/src/construct/partial_whitespace.rs
diff options
context:
space:
mode:
Diffstat (limited to 'src/construct/partial_whitespace.rs')
-rw-r--r--src/construct/partial_whitespace.rs66
1 files changed, 66 insertions, 0 deletions
diff --git a/src/construct/partial_whitespace.rs b/src/construct/partial_whitespace.rs
new file mode 100644
index 0000000..dd0d2b5
--- /dev/null
+++ b/src/construct/partial_whitespace.rs
@@ -0,0 +1,66 @@
+//! A little helper to parse `space_or_tab`
+//!
+//! They’re formed with the following BNF:
+//!
+//! ```bnf
+//! space_or_tab ::= 1*(' ' '\t')
+//! ```
+//!
+//! Depending on where whitespace can occur, it can be optional (or not),
+//! and present in the rendered result (or not).
+//!
+//! ## References
+//!
+//! * [`micromark-factory-space/index.js` in `micromark`](https://github.com/micromark/micromark/blob/main/packages/micromark-factory-space/dev/index.js)
+//!
+//! <!-- To do: link stuff -->
+
+use crate::tokenizer::{Code, State, StateFnResult, TokenType, Tokenizer};
+
+// To do: should `token_type` be a `Some`, with `None` defaulting to something?
+// To do: should `max: Some(usize)` be added?
+
+/// Before whitespace.
+///
+/// ```markdown
+/// alpha| bravo
+/// ```
+pub fn start(tokenizer: &mut Tokenizer, code: Code, token_type: TokenType) -> StateFnResult {
+ match code {
+ Code::VirtualSpace | Code::Char('\t' | ' ') => {
+ // To do: lifetimes.
+ let clone = token_type.clone();
+ tokenizer.enter(token_type);
+ tokenizer.consume(code);
+ (
+ State::Fn(Box::new(|tokenizer, code| inside(tokenizer, code, clone))),
+ None,
+ )
+ }
+ _ => (State::Nok, None),
+ }
+}
+
+/// In whitespace.
+///
+/// ```markdown
+/// alpha |bravo
+/// alpha | bravo
+/// ```
+fn inside(tokenizer: &mut Tokenizer, code: Code, token_type: TokenType) -> StateFnResult {
+ match code {
+ Code::VirtualSpace | Code::Char('\t' | ' ') => {
+ tokenizer.consume(code);
+ (
+ State::Fn(Box::new(|tokenizer, code| {
+ inside(tokenizer, code, token_type)
+ })),
+ None,
+ )
+ }
+ _ => {
+ tokenizer.exit(token_type);
+ (State::Ok, Some(vec![code]))
+ }
+ }
+}