Quinean Regex Tester

Question

This challenge is pretty simple. As input, you take in a regular expression.

Then, you output a truthy/falsey of whether or not your source code matches the regular expression. It's that simple! Just two more things:

• No quine builtins; you may, however, access the code's source code by file IO, etc.
• This is code-golf, so shortest code in bytes wins!

Example

If your source code was say, abc, an input of a\wc would return true and an input of a\dc would return false.

• Mama Fun Roll: Example please?
• Maltysen: @CᴏɴᴏʀO'Bʀɪᴇɴ no, that's fine.
• Dennis: 1. I don't think just a BRE with simple character classes is specific enough. What BRE features have to be supported? 2. \d is not special in BRE; it matches the character d. 3. Choosing a specific regex flavor restricts your challenge to languages that support it, and few languages support BRE. Is that intentional?
• Dennis: I'd recommend leaving it up to the answerer. If language x uses regex flavor y by default, let it use that flavor in this challenge.
• feersum: @Maltysen Why don't you add a substring example to the question body?

Answer 1

Z shell, 12 bytes

grep "$@"<0ドル

Zsh conditionals understand only exit codes, and the scripts exits with 0 or 1 accordingly.

In addition, this prints a non-empty string (the source code) for a match and an empty one for a mismatch, which could be as truthy/falsy values in combination with test/[.

The program reads its own file, but according to this comment by the OP, this is allowed.

• Conor O'Brien: Aaand Dennis won. ¯\_(ツ)_/¯
• feersum: This doesn't work. It breaks on patterns with spaces in them.
• Dennis: @feersum Whoops! Thanks for pointing that out. I have edited my answer.
• Ben Millwood: Now it breaks if it is written to a file with spaces in the name. Or a file called -v. Or...
• Dennis: @BenMillwood I'd normally say don't save it with such a file name, but switching to zsh makes it bullet proof without incrementing the byte count.

Answer 2

JavaScript (ES6), 39

(f=_=>!!`(f=${f})()`.match(prompt()))()

• Mama Fun Roll: Just about to post this, but you beat me to it. Great job!
• Nico: The beginning of your code looks like me when trying to understand this : (f=_=)

Answer 3

Python 3, 119 bytes

This just looks cooler, IMO (and it doesn't read the file).

(lambda i:print(bool(__import__('re').search(input(),i))))("(lambda i:print(bool(__import__('re').search(input(),i))))")

---

Python 3, 67 bytes

print(bool(__import__('re').search(input(),open(__file__).read())))

Added after reading this comment.

• cat: int is shorter than bool.

Answer 4

Julia, (削除) 64 (削除ここまで) 54 bytes

r=readline;show(ismatch(Regex(r()),open(r,@__FILE__)))

Julia regular expressions use PCRE. While reading the source code of the file is a standard loophole for quines, in this case it has been explicitly allowed. Takes input with no trailing newline.

Answer 5

Japt, 22 bytes

"+Q 3sAJ fU"+Q 3sAJ fU

Standard quine framework with a few bytes added to fit this challenge. Truthy = match(es), falsy = null. Try it online!

 // Implicit: U = input string, A = 10, J = -1, Q = quotation mark
"..."+Q // Take this string and concatenate a quotation mark.
3 // Repeat three times.
sAJ // Slice off the first 10 and last 1 chars.
fU // Match U to the result.

Answer 6

Perl, 21 bytes

open 0;$_=<0>=~$_

17 bytes plus 4 bytes for -pl0. Run like this:

echo open | perl -pl0 quinean

The source file must contain only the code above (no shebang, no trailing newline). Outputs 1 if the regex matches and the empty string if it doesn't (the empty string is falsey in Perl).

---

Four bytes can be saved if the input is guaranteed not to end in a newline:

open 0;say<0>=~<>

Run like this:

echo -n open | perl -M5.010 quinean

say requires Perl 5.10+ and must be enabled with -M5.010. According to Meta, "the -M5.010, when needed, is free," giving a score of 17 bytes.

How it works

This is a simple variation on the standard "cheating" quine:

open 0;print<0>

This opens the file named in 0ドル and reads the contents with <0>.

$_=<0>=~$_ reads one line from the source file, does a regex match against the contents of $_ (which were read by the -p flag), and assigns the result to $_. -p prints $_ automatically at the end.

Answer 7

Mathematica, 63 bytes

StringMatchQ[ToString[#0, InputForm], RegularExpression[#1]] & 

Note the trailing space. Uses the standard Mma quine mechanism, and tests if it matches the regex.

Answer 8

Jolf, (削除) 18 (削除ここまで) 15 bytes

Supports the JS flavour of RegEx, I hope that's okay. Try it here!.

 h$code.value#i

Commented:

 $code.value# the document's element "code" (the program container)
_h i and output if it has (matches) the input string (i.e. regex)

• Dennis: In which browser does this work? Both Chrome and Firefox complain that x.step is not a function.
• Conor O'Brien: @Dennis Huh. I must have broken the interpreter last night. What else is wrong? I currently am unable to debug, am at school.
• user48538: Good. Now add a shortcut to the document's element "code" so we can make it shorter.
• Dennis: @CᴏɴᴏʀO'Bʀɪᴇɴ It also gives a reference error for math.
• Conor O'Brien: @Dennis Ah, that's why. I forgot to update the HTML, I added math.js. I will revise when I arrive home, if that's not too late. (In about 4 ish hours)

Answer 9

ESMin, 14 chars / 26 bytes (non-competitive)

⟮!!(CX222+ᶈ0)đï

Try it here (Firefox only).

Using a version with bug fixes written after the challenge.

Explanation

⟮!!(CX222+ᶈ0)đï // implicit: ï=input
⟮ // copy block: copy following code for later use
 (CX222+ᶈ0) // take convert 10222 to char, add stuff inside copy block
!! đï // check if input matches resulting string
 // implicit output

NOTE: Copy blocks are NOT quine operators. They are meant to be more versatile alternatives to variable declarations.

• lirtosiast: I think you can save a byte by changing X to 10.