talexander/organic
1
0
Fork 0
Code Issues Pull Requests Releases Activity

Handle tabs for plain list descriptions.

Browse Source 
This bug probably exists in hundreds of places across the code base. I am going to have to write a "fuzzer" that replaces random whitespace with tabs to find them all.
This commit is contained in:
Tom Alexander 2023-09-08 20:02:49 -04:00
parent 344ef04453
commit a8fbf01124
Signed by: talexander
GPG Key ID: D3A179C9A53C0EDE
2 changed files with 8 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/parser/plain_list.rs
View File

@ -332,7 +332,7 @@ fn item_tag<'b, 'g, 'r, 's>(
),
|(children, _exit_contents)| !children.is_empty(),
)(input)?;
let (remaining, _) = tag(" ::")(remaining)?;
let (remaining, _) = tuple((one_of(" \t"), tag("::")))(remaining)?;
Ok((remaining, children))
}
@ -341,9 +341,10 @@ fn item_tag_end<'b, 'g, 'r, 's>(
_context: RefContext<'b, 'g, 'r, 's>,
input: OrgSource<'s>,
) -> Res<OrgSource<'s>, OrgSource<'s>> {
recognize(alt((
tag(" :: "),
recognize(tuple((tag(" ::"), alt((line_ending, eof))))),
recognize(tuple((
one_of(" \t"),
tag("::"),
alt((recognize(one_of(" \t")), line_ending, eof)),
)))(input)
}

6
src/parser/util.rs
View File

@ -2,13 +2,13 @@ use nom::branch::alt;
use nom::character::complete::anychar;
use nom::character::complete::line_ending;
use nom::character::complete::none_of;
use nom::character::complete::one_of;
use nom::character::complete::space0;
use nom::combinator::eof;
use nom::combinator::not;
use nom::combinator::opt;
use nom::combinator::peek;
use nom::combinator::recognize;
use nom::combinator::verify;
use nom::multi::many0;
use nom::multi::many_till;
use nom::sequence::tuple;
@ -94,9 +94,9 @@ pub fn maybe_consume_object_trailing_whitespace_if_not_exiting<'b, 'g, 'r, 's>(
) -> Res<OrgSource<'s>, Option<OrgSource<'s>>> {
// We have to check exit matcher after each character because description list tags need to end with a space unconsumed (" ::").
let (remaining, _) = many_till(
verify(anychar, |c| *c == ' '),
one_of(" \t"),
alt((
peek(recognize(verify(anychar, |c| *c != ' '))),
peek(recognize(none_of(" \t"))),
parser_with_context!(exit_matcher_parser)(context),
)),
)(input)?;