actions/typos
1
0
Fork 0
mirror of https://github.com/crate-ci/typos.git synced 2024-11-22 09:01:04 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge pull request #291 from epage/parse

Browse source 
refactor(parser): Consolidate utf8/ascii logic
This commit is contained in:
Ed Page 2021-06-29 03:27:39 -07:00 committed by GitHub
commit 21231bfc4d
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 54 additions and 43 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

97
crates/typos/src/tokens.rs
View file

@ -195,11 +195,63 @@ fn is_hex_digit(chr: u8) -> bool {
chr.is_ascii_hexdigit()
}
mod unicode_parser {
mod parser {
use nom::bytes::complete::*;
use nom::sequence::*;
use nom::IResult;
pub(crate) trait AsChar: nom::AsChar {
#[allow(clippy::wrong_self_convention)]
fn is_xid_continue(self) -> bool;
}
impl AsChar for u8 {
fn is_xid_continue(self) -> bool {
(b'a'..=b'z').contains(&self)
|| (b'A'..=b'Z').contains(&self)
|| (b'0'..=b'9').contains(&self)
|| self == b'_'
}
}
impl AsChar for char {
fn is_xid_continue(self) -> bool {
unicode_xid::UnicodeXID::is_xid_continue(self)
}
}
pub(crate) fn next_literal<T>(input: T) -> IResult<T, T>
where
T: nom::InputTakeAtPosition,
<T as nom::InputTakeAtPosition>::Item: AsChar,
{
preceded(literal_sep, identifier)(input)
}
fn literal_sep<T>(input: T) -> IResult<T, T>
where
T: nom::InputTakeAtPosition,
<T as nom::InputTakeAtPosition>::Item: AsChar,
{
take_till(AsChar::is_xid_continue)(input)
}
fn identifier<T>(input: T) -> IResult<T, T>
where
T: nom::InputTakeAtPosition,
<T as nom::InputTakeAtPosition>::Item: AsChar,
{
// Generally a language would be `{XID_Start}{XID_Continue}*` but going with only
// `{XID_Continue}+` because XID_Continue is a superset of XID_Start and rather catch odd
// or unexpected cases than strip off start characters to a word since we aren't doing a
// proper word boundary parse
take_while1(AsChar::is_xid_continue)(input)
}
}
mod unicode_parser {
use super::parser::next_literal;
pub(crate) fn iter_literals(mut input: &str) -> impl Iterator<Item = &str> {
std::iter::from_fn(move || match next_literal(input) {
Ok((i, o)) => {
@ -210,28 +262,10 @@ mod unicode_parser {
_ => None,
})
}
fn next_literal(input: &str) -> IResult<&str, &str> {
preceded(literal_sep, identifier)(input)
}
fn literal_sep(input: &str) -> IResult<&str, &str> {
take_till(unicode_xid::UnicodeXID::is_xid_continue)(input)
}
fn identifier(input: &str) -> IResult<&str, &str> {
// Generally a language would be `{XID_Start}{XID_Continue}*` but going with only
// `{XID_Continue}+` because XID_Continue is a superset of XID_Start and rather catch odd
// or unexpected cases than strip off start characters to a word since we aren't doing a
// proper word boundary parse
take_while1(unicode_xid::UnicodeXID::is_xid_continue)(input)
}
}
mod ascii_parser {
use nom::bytes::complete::*;
use nom::sequence::*;
use nom::IResult;
use super::parser::next_literal;
pub(crate) fn iter_literals(mut input: &[u8]) -> impl Iterator<Item = &str> {
std::iter::from_fn(move || match next_literal(input) {
@ -246,29 +280,6 @@ mod ascii_parser {
_ => None,
})
}
fn next_literal(input: &[u8]) -> IResult<&[u8], &[u8]> {
preceded(literal_sep, identifier)(input)
}
fn literal_sep(input: &[u8]) -> IResult<&[u8], &[u8]> {
take_till(is_continue)(input)
}
fn identifier(input: &[u8]) -> IResult<&[u8], &[u8]> {
// Generally a language would be `{XID_Start}{XID_Continue}*` but going with only
// `{XID_Continue}+` because XID_Continue is a superset of XID_Start and rather catch odd
// or unexpected cases than strip off start characters to a word since we aren't doing a
// proper word boundary parse
take_while1(is_continue)(input)
}
fn is_continue(c: u8) -> bool {
(b'a'..=b'z').contains(&c)
|| (b'A'..=b'Z').contains(&c)
|| (b'0'..=b'9').contains(&c)
|| c == b'_'
}
}
/// A term composed of Words.