Spurious matches with substring matching and non-ASCII about nucleo HOT 8 CLOSED

good find, I did have cases for checking that unicode substring search works for matches but not rejection test cases for cases where patterns did not match. The bug was a small oversight and has been fixed in 34553f0.

I included this in #33 just in time for the new release

from nucleo.

I'm actually getting the opposite problem now where now it's not matching even though it should be.

from nucleo.

hmm yeah that is an orthogonal bug, fixed in c7893db

from nucleo.

Now I'm getting a crash:

        let needle = Utf32String::from("adi");
        let haystack =
            Utf32String::from("At the Road’s End - Seeming - SOL: A Self-Banishment Ritual");
        let mut matcher = Matcher::new(Config::DEFAULT);
        assert_ne!(
            matcher.substring_match(haystack.slice(..), needle.slice(..)),
            None
        )

range end index 60 out of range for slice of length 59
stack backtrace:
   0: rust_begin_unwind
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/std/src/panicking.rs:597:5
   1: core::panicking::panic_fmt
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/panicking.rs:72:14
   2: core::slice::index::slice_end_index_len_fail_rt
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:76:5
   3: core::slice::index::slice_end_index_len_fail
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:68:9
   4: <core::ops::range::Range<usize> as core::slice::index::SliceIndex<[T]>>::index
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:411:13
   5: core::slice::index::<impl core::ops::index::Index<I> for [T]>::index
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:18:9
   6: nucleo_matcher::exact::<impl nucleo_matcher::Matcher>::substring_match_non_ascii
             at /home/vector/.cargo/git/checkouts/nucleo-fe29e1ee969779b0/c7893db/matcher/src/exact.rs:248:28
   7: nucleo_matcher::Matcher::substring_match_impl
             at /home/vector/.cargo/git/checkouts/nucleo-fe29e1ee969779b0/c7893db/matcher/src/lib.rs:485:17
   8: nucleo_matcher::Matcher::substring_match
             at /home/vector/.cargo/git/checkouts/nucleo-fe29e1ee969779b0/c7893db/matcher/src/lib.rs:422:9

from nucleo.

Seems like the index generation is also really glitchy:

from nucleo.

Now I'm getting a crash:

        let needle = Utf32String::from("adi");
        let haystack =
            Utf32String::from("At the Road’s End - Seeming - SOL: A Self-Banishment Ritual");
        let mut matcher = Matcher::new(Config::DEFAULT);
        assert_ne!(
            matcher.substring_match(haystack.slice(..), needle.slice(..)),
            None
        )

range end index 60 out of range for slice of length 59
stack backtrace:
   0: rust_begin_unwind
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/std/src/panicking.rs:597:5
   1: core::panicking::panic_fmt
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/panicking.rs:72:14
   2: core::slice::index::slice_end_index_len_fail_rt
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:76:5
   3: core::slice::index::slice_end_index_len_fail
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:68:9
   4: <core::ops::range::Range<usize> as core::slice::index::SliceIndex<[T]>>::index
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:411:13
   5: core::slice::index::<impl core::ops::index::Index<I> for [T]>::index
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/slice/index.rs:18:9
   6: nucleo_matcher::exact::<impl nucleo_matcher::Matcher>::substring_match_non_ascii
             at /home/vector/.cargo/git/checkouts/nucleo-fe29e1ee969779b0/c7893db/matcher/src/exact.rs:248:28
   7: nucleo_matcher::Matcher::substring_match_impl
             at /home/vector/.cargo/git/checkouts/nucleo-fe29e1ee969779b0/c7893db/matcher/src/lib.rs:485:17
   8: nucleo_matcher::Matcher::substring_match
             at /home/vector/.cargo/git/checkouts/nucleo-fe29e1ee969779b0/c7893db/matcher/src/lib.rs:422:9

For this crash it looks like above the code in the commit that Pascal refrenced where it is enumerating to the end of the haystack it should be enumerating to the end minus the length of the needle.

from nucleo.

Yeah I have a fix for that but I dont want to rush another fix (altough it is yet another existing bug, I guess I just need to weite moee unicode-unicode suabteing tests. I never type unicode so I rarely run into these).

from nucleo.

Yeah, I don't type Unicode usually but I have a music collection with a lot of foreign stuff or weird smart quotes and I'm writing a music player so I've been hitting the ascii-needle/unicode haystack codepath a lot.

from nucleo.