Correctness issue: pattern doesn't match when it should

[khyperia]

What version of ripgrep are you using?

% rg --version
ripgrep 0.10.0
-SIMD -AVX (compiled)
+SIMD +AVX (runtime)

What operating system are you using ripgrep on?

Arch Linux, using the Arch-official ripgrep package.

Bug description/reproduction steps

This prints out "abcd".

echo 'abcd' | rg 'a(.*(d|ce))'

Adding a "c" makes it not match (i.e. it doesn't print out anything).

echo 'abcd' | rg 'a(.*(cd|ce))'

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

I believe that both regexes should match the input, and print out a match in both cases. Testing using regex101 seems to imply this is true.

Real-world discovery background (non-minimal case that made me discover the minimal repro above):

I was trying to run git diff --name-only master... | rg 'js/src/(.*(cpp|\.h))' --replace '$1'.

git diff --name-only master... printed lines such as:

js/src/frontend/BinSource.yaml
js/src/frontend/BytecodeEmitter.cpp
js/src/frontend/ParseNode.h

and I wanted to extract:

frontend/BytecodeEmitter.cpp
frontend/ParseNode.h

(i.e. trim js/src/ off the front, and only match cpp and h files)

However, adding a \. in front of the cpp made nothing match.

git diff --name-only master... | rg 'js/src/(.*(\.cpp|\.h))' --replace '$1' -- prints nothing???

[BurntSushi]

This is a duplicate of #1064 and #1081. It's fixed on master.

From my digging, I thought that this bug wasn't a regression, but had always been there in previous versions of ripgrep. However, we've had this bug filed three times since the release of ripgrep 0.10, but never before. So something seems fishy.

[AntCrow]

FWIW, I don't think it's a regression. I found the same thing on 0.8.1, confirmed it on 0.10.0, and was going to report it until I found all the duplicates.