refactor: cleanup parser
This commit is contained in:
PoiScript
parent
1bb5286dd3
commit
8bb7ae41d3
6 changed files with 128 additions and 99 deletions
|
|
@ -13,19 +13,19 @@ pub fn parse(src: &str) -> Option<(&str, Option<&str>, usize, usize, usize)> {
|
|||
let name = memchr2(b' ', b'\n', src.as_bytes())
|
||||
.filter(|&i| src.as_bytes()[8..i].iter().all(u8::is_ascii_alphabetic))?;
|
||||
let mut lines = Lines::new(src);
|
||||
let (pre_cont_end, cont_beg, _) = lines.next()?;
|
||||
let args = if pre_cont_end == name {
|
||||
let (pre_limit, begin, _) = lines.next()?;
|
||||
let args = if pre_limit == name {
|
||||
None
|
||||
} else {
|
||||
Some(&src[name..pre_cont_end])
|
||||
Some(&src[name..pre_limit])
|
||||
};
|
||||
let name = &src[8..name];
|
||||
let end_line = format!(r"#+END_{}", name.to_uppercase());
|
||||
let mut pre_end = cont_beg;
|
||||
let mut pre_end = begin;
|
||||
|
||||
for (_, end, line) in lines {
|
||||
if line.trim() == end_line {
|
||||
return Some((name, args, cont_beg, pre_end, end));
|
||||
return Some((name, args, begin, pre_end, end));
|
||||
} else {
|
||||
pre_end = end;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -11,9 +11,9 @@ pub fn parse(src: &str) -> Option<(&str, Option<&str>, usize, usize, usize)> {
|
|||
}
|
||||
|
||||
let mut lines = Lines::new(src);
|
||||
let (mut pre_cont_end, _, _) = lines.next()?;
|
||||
let (mut pre_limit, _, _) = lines.next()?;
|
||||
|
||||
for (cont_end, end, line) in lines {
|
||||
for (limit, end, line) in lines {
|
||||
if line.trim().eq_ignore_ascii_case("#+END:") {
|
||||
let bytes = src.as_bytes();
|
||||
|
||||
|
|
@ -23,21 +23,15 @@ pub fn parse(src: &str) -> Option<(&str, Option<&str>, usize, usize, usize)> {
|
|||
let name = &src[8..i].trim();
|
||||
|
||||
return Some(if bytes[i] == b'\n' {
|
||||
(name, None, i, pre_cont_end, end)
|
||||
(name, None, i, pre_limit, end)
|
||||
} else {
|
||||
let cont_beg = memchr(b'\n', bytes)
|
||||
let begin = memchr(b'\n', bytes)
|
||||
.map(|i| i + 1)
|
||||
.unwrap_or_else(|| src.len());
|
||||
(
|
||||
name,
|
||||
Some(&src[i..cont_beg].trim()),
|
||||
cont_beg,
|
||||
pre_cont_end,
|
||||
end,
|
||||
)
|
||||
(name, Some(&src[i..begin].trim()), begin, pre_limit, end)
|
||||
});
|
||||
}
|
||||
pre_cont_end = cont_end;
|
||||
pre_limit = limit;
|
||||
}
|
||||
|
||||
None
|
||||
|
|
|
|||
|
|
@ -1,39 +1,46 @@
|
|||
use crate::lines::Lines;
|
||||
|
||||
#[inline]
|
||||
pub fn is_item(src: &str) -> (bool, bool) {
|
||||
if src.is_empty() {
|
||||
return (false, false);
|
||||
pub fn is_item(text: &str) -> Option<bool> {
|
||||
if text.is_empty() {
|
||||
return None;
|
||||
}
|
||||
let bytes = src.as_bytes();
|
||||
let (i, ordered) = match bytes[0] {
|
||||
b'*' | b'-' | b'+' => (1, false),
|
||||
|
||||
let bytes = text.as_bytes();
|
||||
match bytes[0] {
|
||||
b'*' | b'-' | b'+' => {
|
||||
if text.len() > 1 && (bytes[1] == b' ' || bytes[1] == b'\n') {
|
||||
Some(false)
|
||||
} else {
|
||||
None
|
||||
}
|
||||
}
|
||||
b'0'...b'9' => {
|
||||
let i = bytes
|
||||
.iter()
|
||||
.position(|&c| !c.is_ascii_digit())
|
||||
.unwrap_or_else(|| src.len() - 1);
|
||||
let c = bytes[i];
|
||||
if !(c == b'.' || c == b')') {
|
||||
return (false, false);
|
||||
.unwrap_or_else(|| text.len() - 1);
|
||||
if (bytes[i] == b'.' || bytes[i] == b')')
|
||||
&& i + 1 < text.len()
|
||||
&& (bytes[i + 1] == b' ' || bytes[i + 1] == b'\n')
|
||||
{
|
||||
Some(true)
|
||||
} else {
|
||||
None
|
||||
}
|
||||
(i + 1, true)
|
||||
}
|
||||
_ => return (false, false),
|
||||
};
|
||||
|
||||
if i < src.len() {
|
||||
// bullet is follwed by a space or line ending
|
||||
(bytes[i] == b' ' || bytes[i] == b'\n', ordered)
|
||||
} else {
|
||||
(false, false)
|
||||
_ => None,
|
||||
}
|
||||
}
|
||||
|
||||
// returns (bullets, contents begin, contents end, end, has more)
|
||||
#[inline]
|
||||
pub fn parse(src: &str, ident: usize) -> (&str, usize, usize, usize, bool) {
|
||||
debug_assert!(is_item(&src[ident..]).0);
|
||||
debug_assert!(
|
||||
is_item(&src[ident..]).is_some(),
|
||||
"{:?} is not a list item",
|
||||
src
|
||||
);
|
||||
debug_assert!(
|
||||
src[..ident].chars().all(|c| c == ' ' || c == '\t'),
|
||||
"{:?} doesn't starts with indentation {}",
|
||||
|
|
@ -42,60 +49,60 @@ pub fn parse(src: &str, ident: usize) -> (&str, usize, usize, usize, bool) {
|
|||
);
|
||||
|
||||
let mut lines = Lines::new(src);
|
||||
let (mut pre_cont_end, mut pre_end, first_line) = lines.next().unwrap();
|
||||
let beg = match memchr::memchr(b' ', &first_line.as_bytes()[ident..]) {
|
||||
let (mut pre_limit, mut pre_end, first_line) = lines.next().unwrap();
|
||||
let begin = match memchr::memchr(b' ', &first_line.as_bytes()[ident..]) {
|
||||
Some(i) => i + ident + 1,
|
||||
None => {
|
||||
let len = first_line.len();
|
||||
return (
|
||||
&first_line,
|
||||
first_line,
|
||||
len,
|
||||
len,
|
||||
len,
|
||||
is_item(lines.next().unwrap().2).0,
|
||||
is_item(lines.next().unwrap().2).is_some(),
|
||||
);
|
||||
}
|
||||
};
|
||||
let bullet = &src[0..beg];
|
||||
let bullet = &src[0..begin];
|
||||
|
||||
while let Some((mut cont_end, mut end, mut line)) = lines.next() {
|
||||
while let Some((mut limit, mut end, mut line)) = lines.next() {
|
||||
// this line is emtpy
|
||||
if line.is_empty() {
|
||||
if let Some((next_cont_end, next_end, next_line)) = lines.next() {
|
||||
if let Some((next_limit, next_end, next_line)) = lines.next() {
|
||||
// next line is emtpy, too
|
||||
if next_line.is_empty() {
|
||||
return (bullet, beg, pre_cont_end, next_end, false);
|
||||
return (bullet, begin, pre_limit, next_end, false);
|
||||
} else {
|
||||
// move to next line
|
||||
pre_end = end;
|
||||
cont_end = next_cont_end;
|
||||
limit = next_limit;
|
||||
end = next_end;
|
||||
line = next_line;
|
||||
}
|
||||
} else {
|
||||
return (bullet, beg, pre_cont_end, end, false);
|
||||
return (bullet, begin, pre_limit, end, false);
|
||||
}
|
||||
}
|
||||
|
||||
let line_ident = count_ident(line);
|
||||
|
||||
if line_ident < ident {
|
||||
return (bullet, beg, pre_cont_end, pre_end, false);
|
||||
return (bullet, begin, pre_limit, pre_end, false);
|
||||
} else if line_ident == ident {
|
||||
return (
|
||||
bullet,
|
||||
beg,
|
||||
pre_cont_end,
|
||||
begin,
|
||||
pre_limit,
|
||||
pre_end,
|
||||
is_item(&line[ident..]).0,
|
||||
is_item(&line[ident..]).is_some(),
|
||||
);
|
||||
}
|
||||
|
||||
pre_end = end;
|
||||
pre_cont_end = cont_end;
|
||||
pre_limit = limit;
|
||||
}
|
||||
|
||||
(bullet, beg, src.len(), src.len(), false)
|
||||
(bullet, begin, src.len(), src.len(), false)
|
||||
}
|
||||
|
||||
#[inline]
|
||||
|
|
@ -112,18 +119,18 @@ mod tests {
|
|||
fn is_item() {
|
||||
use super::is_item;
|
||||
|
||||
assert_eq!(is_item("+ item"), (true, false));
|
||||
assert_eq!(is_item("- item"), (true, false));
|
||||
assert_eq!(is_item("10. item"), (true, true));
|
||||
assert_eq!(is_item("10) item"), (true, true));
|
||||
assert_eq!(is_item("1. item"), (true, true));
|
||||
assert_eq!(is_item("1) item"), (true, true));
|
||||
assert_eq!(is_item("10. "), (true, true));
|
||||
assert_eq!(is_item("10.\n"), (true, true));
|
||||
assert_eq!(is_item("10."), (false, false));
|
||||
assert_eq!(is_item("+"), (false, false));
|
||||
assert_eq!(is_item("-item"), (false, false));
|
||||
assert_eq!(is_item("+item"), (false, false));
|
||||
assert_eq!(is_item("+ item"), Some(false));
|
||||
assert_eq!(is_item("- item"), Some(false));
|
||||
assert_eq!(is_item("10. item"), Some(true));
|
||||
assert_eq!(is_item("10) item"), Some(true));
|
||||
assert_eq!(is_item("1. item"), Some(true));
|
||||
assert_eq!(is_item("1) item"), Some(true));
|
||||
assert_eq!(is_item("10. "), Some(true));
|
||||
assert_eq!(is_item("10.\n"), Some(true));
|
||||
assert_eq!(is_item("10."), None);
|
||||
assert_eq!(is_item("+"), None);
|
||||
assert_eq!(is_item("-item"), None);
|
||||
assert_eq!(is_item("+item"), None);
|
||||
}
|
||||
|
||||
#[test]
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue