ripgrep doesn't match arbitrary bytes within a file

[sersorrel]

What version of ripgrep are you using?

ripgrep 11.0.1 (rev 1f1cd9b)
-SIMD -AVX (compiled)
+SIMD +AVX (runtime)

How did you install ripgrep?

GitHub deb, I think

What operating system are you using ripgrep on?

Ubuntu 18.04

Describe your question, feature request, or bug.

grep finds arbitrary bytes within a binary file, but ripgrep does not.

If this is a bug, what are the steps to reproduce the behavior?

$ grep $'\xa7' <(printf '\xa7')
Binary file /dev/fd/63 matches
$ echo $?
0
$ rg -uuu --text --binary '\xa7' <(printf '\xa7')
$ echo $?
1

If this is a bug, what is the actual behavior?

$ rg -uuu --text --binary --debug '\xa7' <(printf '\xa7')
DEBUG|grep_regex::literal|grep-regex/src/literal.rs:59: literal prefixes detected: Literals { lits: [Complete(§)], limit_size: 250, limit_class: 10 }
DEBUG|globset|globset/src/lib.rs:435: built glob set; 0 literals, 0 basenames, 11 extensions, 0 prefixes, 0 suffixes, 0 required extensions, 0 regexes
DEBUG|globset|globset/src/lib.rs:435: built glob set; 0 literals, 0 basenames, 11 extensions, 0 prefixes, 0 suffixes, 0 required extensions, 0 regexes
DEBUG|globset|globset/src/lib.rs:435: built glob set; 0 literals, 0 basenames, 11 extensions, 0 prefixes, 0 suffixes, 0 required extensions, 0 regexes

If this is a bug, what is the expected behavior?

Either ripgrep should be able to search for arbitrary bytes, or it should not print a message implying that it can do so:

$ rg $'\xa7'
found invalid UTF-8 in pattern at byte offset 0 (use hex escape sequences to match arbitrary bytes in a pattern, e.g., \xFF): '\xA7'

[BurntSushi]

By default, patterns are Unicode aware, so all escape sequences identify the codepoint to search for, which is not the same as the byte to search for. To search for a raw byte, you need to disable Unicode. For example, (?-u:\xa7) should do what you want. This is documented in the syntax docs linked from the man page.

The error message is indeed wrong, or at least, incomplete.