# Are these identifiers equivalent?

In the language Nim, the rules for differentiating identifiers are slightly more relaxed than most other languages. Two identifiers are equivalent or address the same variable if they follow these rules:

• the first character of both are the same (case sensitive)
• both strings are the same (case insensitive) after removing all instances of the characters - and _

## Challenge

Write a program/function that takes two strings that represent Nim identifiers and output a truthy or falsey value based on whether or not they are equivalent by the rules above.

### Specifications

• Standard I/O rules apply.
• Standard loopholes are forbidden.
• The strings will only contain ASCII printables. You do not need to check if it's a valid identifier.
• The strings may be taken as two separate inputs, list of strings, etc. (you know the drill)
• Empty strings need not be handled.
• The output must be consistent for both truthy and falsey values.
• This challenge is not about finding the shortest approach in all languages, rather, it is about finding the shortest approach in each language.
• Your code will be scored in bytes, usually in the encoding UTF-8, unless specified otherwise.
• Built-in functions that perform this task are allowed but including a solution that doesn't rely on a built-in is encouraged.
• Explanations, even for "practical" languages, are encouraged.

## Test cases

Input                                    Output

count, Count                             falsey
_test, test                              falsey
test, tset                               falsey
aVariableName, a_variable_name           truthy
numbers_are_cool123, numbersAreCool123   truthy
symbolsAre_too>_>, symbols_areTOO>>      truthy


## Ungolfed reference implementation

This is written in Nim, itself.

import strutils, re

proc sameIdentifier(a, b: string): bool =
a[0] == b[0] and
a.replace(re"_|–", "").toLower == b.replace(re"_|–", "").toLower

• I suggest a test case of f("_test", "test"). – Doorknob Jul 28 '17 at 2:51
• @Doorknob Added. – totallyhuman Jul 28 '17 at 3:14
• I suggest adding f("test", "tset"), as I think one answer gives an unexpected result for it. – Ørjan Johansen Jul 29 '17 at 0:11
• @ØrjanJohansen Done. – totallyhuman Jul 29 '17 at 18:36
• Wait, so input is strings that "represent Nim identifiers", and we "do not need to check if it's a valid identifier", but then one of the examples contains >? – aschepler Jul 29 '17 at 21:54

## JavaScript (ES6), 62 61 bytes

Saved 1 byte thanks to @JohanKarlsson

Takes input in currying syntax (a)(b). Returns a boolean.

a=>b=>(r=s=>s[0]+s.replace(/-|_/g,'').toUpperCase())(b)==r(a)


### Test cases

let f =

a=>b=>(r=s=>s[0]+s.replace(/-|_/g,'').toUpperCase())(b)==r(a)

console.log(f("count")("Count"))                           // falsey
console.log(f("_test")("test"))                            // falsey
console.log(f("aVariableName")("a_variable_name"))         // truthy
console.log(f("numbers_are_cool123")("numbersAreCool123")) // truthy

• /-|_/g saves a byte – Johan Karlsson Jul 28 '17 at 13:06

# Python 3, 76 bytes

lambda a,b:f(*a)==f(*b)
f=lambda f,*r:[f+k.lower()for k in r if~-(k in'-_')]


Try it online!

-1 byte thanks to notjagan
-3 bytes thanks to Wheat Wizard

• -1 byte. – notjagan Jul 28 '17 at 2:23
• @notjagan Neat trick; thanks! – HyperNeutrino Jul 28 '17 at 2:26
• @notjagan I did not know you could do that, but I guess it makes sense, cool – Stephen Jul 28 '17 at 2:27
• -14 bytes! – notjagan Jul 28 '17 at 3:31
• even shorter – Sriotchilism O'Zaic Jul 28 '17 at 5:57

# Actually, 15 bytes

⌠p"-_"(-Σùo⌡M═Y


Try it online!

Fun fact: this works with any number of inputs (it always returns truthy for less than 2 inputs).

Explanation:

⌠p"-_"(-Σùo⌡M═Y
⌠p"-_"(-Σùo⌡M    for each input:
p                 separate the first character
"-_"(-           remove all dashes and underscores from the rest of the string
Σù         concatenate the list from the last operation and lowercase the string
o        append it to the first character
═Y  are none of the elements unique?


# Pyth, 13 bytes

qFm[hd-r0d"-_


Try it online!

# Explanation

qFm[hd-r0d"-_
m              For each value in the input (which is a list of two strings):
[             Create a list consisting of
hd               the first character of each value
-r0d"-_        and the lowercase version of the value without "-" or "_"
qF               Fold over equivalence; checks to see if both lists are the same


# 05AB1E, 12 bytes

εćs„-_SKl«}Ë


Try it online!

Theoretically, εćs„-_-«}Ë should've worked for 10 bytes, but unfortunately this behavior is deprecated for now.

• Oh yeah lol, I still need to fix that. You can save a byte by using „-_SK instead of '-K'_K. – Adnan Jul 28 '17 at 11:19
• @Adnan And I knew there was a way. Thanks! – Erik the Outgolfer Jul 28 '17 at 12:59
• 11 bytes if you change SK to м. – Kevin Cruijssen May 11 '18 at 9:46
• @KevinCruijssen Hm, I'll update this answer. I don't think м existed back then. :P – Erik the Outgolfer May 11 '18 at 11:03

# Jelly, 11 bytes

ḟ⁾-_Œl,Ḣµ€E


Try it online!

-2 bytes thanks to Erik the Outgolfer
-1 byte thanks to Jonathan Allan

• Take a list of two strings like ["symbolsAre_too>_>", "symbols_areTOO>>"] and use Ḣ;ḟ⁾-_Œl$µ€E instead for -2. Beat that Pyth! – Erik the Outgolfer Jul 28 '17 at 9:33 • ...or even just ḟ⁾-_Œl,Ḣµ€E for 11 bytes. – Jonathan Allan Jul 28 '17 at 14:29 • No problem, maybe credit Erik for 2 and me for 1 :) – Jonathan Allan Jul 28 '17 at 22:37 • @JonathanAllan Ah. Good idea; I'll do that :) – HyperNeutrino Jul 28 '17 at 22:39 # Ruby, 86646361 51 bytes f=->x{x[0]+x.upcase.delete("-_")} ->x,y{f[x]==f[y]}  Try it online! This feels really long still feels a bit long. I would appreciate the help of any Ruby gurus out there in making this at least a bit shorter. • Not a guru, but I got inspired to check out the String method list. .delete("_-") is shorter. – Ørjan Johansen Jul 28 '17 at 6:50 • f[x] is always a valid substitution for f.call(x) when stabby lambdas are involved. – Value Ink Jul 28 '17 at 23:41 • @ValueInk Thanks! I've been trying to figure out how to golf in Ruby based on Stack Overflow answers, so I didn't know that was an option. – Sriotchilism O'Zaic Jul 29 '17 at 0:58 ## C++, 288 248 bytes -5 bytes thanks to Zacharý #include<string> #include<algorithm> #define E(a,v)a.erase(std::remove(a.begin(),a.end(),v),a.end()); #define F(a)for(auto&c:a)c=toupper(c); int e(std::string&a,std::string&b){if(a[0]!=b[0])return 0;E(a,45)E(b,45)E(a,95)E(b,95)F(a)F(b)return a==b;}  Thanks you, Preprocessor. Also, this code takes advantage of the fact that in C++ the rule to cast int to bool is int_var!=0 • Add a ; after the definition of F. Then, change the first return statement to return 0;E(a,45)E(b,45)E(a,95)E(b,95)F(a)F(b). – Zacharý Jul 29 '17 at 15:05 # CJam, 20 bytes {_"-_"f-:el:=\:c:=*}  Takes input in the form of ["string1","string2"]. Try it Online (testing version) { _ e# make copy of input "-_"f- e# remove all "-" and "_" from both words in copy :el e# convert words in copy to lowercase := e# 1 if both words in copy are equal, 0 if not \ e# move original version of input to top of stack :c e# convert each word in original input to only 1st character := e# 1 if both characters from original input are equal, 0 if not * e# multply the two numbers we obtained. If and only if both are 1 (true) we return 1 (true) }  # Haskell, 85787671 68 bytes 2 bytes saved thanks to Ørjan Johansen import Data.Char s(a:x)=a:[toLower a|a<-x,all(/=a)"-_"] x!y=s x==s y  Try it online! Errors on the empty string. • all(/=a)"-_". Also after your latest edit, f needs to become an operator. – Ørjan Johansen Jul 28 '17 at 5:41 • @ØrjanJohansen Ah thanks. I thought there was a shorter way to do notElem but I couldn't remember it for the life of me. – Sriotchilism O'Zaic Jul 28 '17 at 5:43 # Python 2, 72 bytes lambda x,y:r(x)==r(y) r=lambda x:x[0]+x[1:].lower().translate(None,"-_")  Try it online! Won't work with Python 3 because of the new translate syntax. # Excel, 105 bytes =AND(CODE(A1)=CODE(B1),SUBSTITUTE(SUBSTITUTE(A1,"_",""),"-","")=SUBSTITUTE(SUBSTITUTE(B1,"_",""),"-",""))  CODE() returns numeric code of first character. String comparison in Excel is case insensitive. # Husk, 13 bytes ¤=§,←(m_ω-"-_  Try it online! Builds for each string a pair consisting of the first character of the string and the whole string lowercased and with all occurrences of -/_ removed. Then checks if the two pairs are equal. A particularity is that - in Husk is set difference (i.e. it removes only the first occurrence found): in order to remove all occurrences, the fixed point of -"-_ is found with ω-"-_. # Japt, 14 25 bytes g ¥Vg ©Uu k"_-" ¥Vu k"_-"  Checks case-insensitive string equality by removing all characters in word 2 from word 1, and removing the -_ characters; that results in an empty string ("") if the words are equal. Thanks Ørjan Johansen for pointing out the problem with this. Checks first-char equality and if the uppercased inputs are equal after removing _-. Try it online! ## Explanation Implicit input: U and V are input strings g ¥Vg  Check if first letter of U (implicit) equals (¥) the first char of V. ©Uu k"_-" ¥Vu k"_-"  And (©) check if U, uppercased (u) and with _- removed (k), equals (¥) the same for V. Implicitly return the boolean result. • I cannot get the link to work, but that explanation sounds like it does the wrong thing. What does it give for test vs. tset? – Ørjan Johansen Jul 29 '17 at 0:10 • @ØrjanJohansen Good point... it'll fail for that case. As for the link, I've tested it and it works fine. – Justin Mariner Jul 29 '17 at 0:26 • Yeah the link is my own fault - one of these days I need to change to a modern browser. I had better luck with Try it online! – Ørjan Johansen Jul 29 '17 at 0:27 • @ØrjanJohansen Could I ask what browser you're using? I'm in the process of improving that CodePen and would like to make it as compatible as TIO. – Justin Mariner Jul 29 '17 at 0:39 • cough still using Internet Explorer. – Ørjan Johansen Jul 29 '17 at 0:47 # Python 2, 79 73 bytes -6 bytes thanks to @notjagan: check the length of set of all reduced names is 1 or not. lambda*l:len({x[0]+re.sub('-|_','',x[1:].lower())for x in l})<2 import re  Try it online! • -6 bytes. – notjagan Jul 28 '17 at 3:28 • Of course, set, I forgot its existence!! Thanks a lot!! – officialaimm Jul 28 '17 at 4:01 # Perl 5, 67 bytes s/.//,push@a,$&,y/_-//dr for<>;say($a[0]eq$a[2]&&lc$a[3]eq lc$a[1])


Try it online!

Takes the identifiers as input on separate lines.

Explanation:

s/.//,             # remove the first character
push@a,            # storage space, even positions are first character
# odd positions are remainder
$&, # implicit variable holding last matched pattern (first char) y/_-//dr # Remove _ and - from remainder of input for<>; # iterate over all input lines say # output ($a[0]eq$a[2]&& # check that first character is identical and lc$a[3]eq lc$a[1]) # so is the lowercase version of the rest  # Retina, 28 bytes TL\_-l_(?<=.).+ ^(.*)¶\1$


Try it online!

# Charcoal, 29 bytes

∧⁼§θ⁰§η⁰⁼↧⪫⪪⪫⪪θ_ω-ω↧⪫⪪⪫⪪η_ω-ω


Try it online!

This prints a - for truthy and nothing for falsey.

Link to the verbose version. It first compares the first character of both input strings (⁼§θ⁰§η⁰) and then compares the rest of both strings after removing the underscores and the hyphens (⪫⪪⪫⪪θ_ω-ω) and converting to lowercase (↧).

# C#, 101 89 bytes

string g(string s)=>string.Concat(s.ToUpper().Split('-','_'));f=>s=>f[0]==s[0]&g(f)==g(s)


Saved 12 bytes thanks to @kusi581.

• if you use a local function for string.Concat(...) you can save 2 bytes ;) – kusi581 Jul 28 '17 at 10:49
• @kusi581 Thanks, saved 12 bytes. – TheLethalCoder Jul 28 '17 at 10:53

# Java (OpenJDK 8), 95 bytes

a->b->a.charAt(0)==b.charAt(0)&&a.replaceAll("_|-","").equalsIgnoreCase(b.replaceAll("_|-",""))


Try it online! Pretty straight forward.

# Pyke, 13 bytes

F'hl1"-_"-)Xq


Try it online!

F         )   -  for i in input():
'hl1         -   i[0], i.lower()
"-_"-    -         ^.remove("-_")
Xq - equal(^)


# C (gcc), 126 114 bytes

#define p(s)do++s;while(*s==45||*s==95);*s>95?*s-=32:0;
f(char*a,char*b){while(*a&&*a==*b){p(a)p(b)}return*a==*b;}


Try it online!

#define p(s)                   // Define helper macro p           \
do ++s;                    // Increment pointer at least once \
while (*s==45 | *s==95);   // and past any '-' or '_'         \
*s>95 ? *s -= 32 : 0;      // If lowercase letter, convert to upper

f(char* a, char* b) {          // Define main function f
while (*a && *a == *b) {   // Loop until end of either string
// or a difference found
p(a)                   // Transform pointer and one char
p(b)                   // via helper p above
}
return *a==*b;             // Test chars equal (on success, both '\0')
}

• The question specifies ASCII printables, so (1) The first while test can be shortened to *s%50==45. (2) However, the lowercasing is wrong, e.g. it fails on t~ vs. t^. – Ørjan Johansen Jul 29 '17 at 3:54
• @ØrjanJohansen I thought we could assume the inputs were both valid identifiers. But then that example with > was added, hmm. – aschepler Jul 29 '17 at 22:00
• Huh. I was going by that example too. Looking now in the Nim manual, even - isn't actually allowed, but the algorithm still includes it... – Ørjan Johansen Jul 29 '17 at 23:05
• @ØrjanJohansen Yeah, I noticed - isn't in the grammar description of identifier - but then other parts of that document imply it is allowed. – aschepler Jul 29 '17 at 23:24

# Dyalog APL, 4732282726 22 bytes

-4 bytes thanks to Kritixi Lithos

{(=/⊃¨⍵)∧≡/819⌶⍵~'-_'}


Takes input as a list of the strings.

Try it online!

### How?

{(=/⊃¨⍵)∧≡/819⌶⍵~'-_'}
⍵~'-_'   Remove '-' and '_'
819⌶         Lowercase
≡/             Equality between elements
∧               And
(=/⊃¨⍵)                The first element of each element is equal

• I think you can do ⊃⍺=⍵ instead instead of ⍺[1]=⍵[1] – Cows quack Jul 29 '17 at 14:33
• No, because the arguments could be of a different length! – Zacharý Jul 29 '17 at 14:38
• In that case, ⊃⍵ instead of ⍵[1] should work – Cows quack Jul 29 '17 at 14:38
• Maybe even ⊃⍺=⊃⍵ instead of ⍺[1]=⍵[1] – Cows quack Jul 29 '17 at 14:41

# Common Lisp, 98 bytes

(lambda(x y)(and(eql(elt x 0)(elt y 0))(string-equal(#1=remove #\-(#1##\_ y))(#1##\-(#1##\_ x)))))


Try it online!

Ungolfed (super straightforward!) version:

(defun f(x y)
(and (eql (elt x 0) (elt y 0))         ; check if initial characters are identical
(string-equal                     ; string comparison (case insensitive)
(remove #\- (remove #\_ y))     ; remove from both strings the unwanted chars
(remove #\- (remove #\_ x)))))


# R, 76 bytes

function(l)(g=substr(l,1,1))[1]==g[2]&(h=tolower(gsub('-|_','',l)))[1]==h[2]


Anonymous function that takes input as a list of two strings. Takes advantage of the fact that R's string operations, while quite long in # of characters, are vectorized. Additionally wrapping an assignment in parentheses will bind the variable, so (g=substr(l,1,1)) retains a variable to be reused later in the line and similarly for h.

R returns the last evaluated expression as function output.

Ungolfed:

function(l){
g <- substr(l,1,1)
h <- tolower(gsub("_|-","",l))
(g[1]==g[2])&(h[1]==h[2])
}


Try it online! (all test cases)

# Brachylog, 17 bytes

hᵛ&{{¬∈"_-"&ụ}ˢ}ᵛ


Try it online!

Outputs through predicate success/failure.

h                    The first element
ᵛ                   is the same for each element of the input,
&                  and
{           }ᵛ    for each element of the input the following are the same:
{      &ụ}ˢ      every element uppercased which satisfies the condition that
¬∈              it is not an element of
"_-"          the string "_-".


# Erlang 113 bytes

A=fun(L)->string:to_lower(lists:flatten(string:tokens(L,"-_")))end,fun([C|D],[C|E])->A(D)==A(E);(_,_)->a==b end.


a pair of anonymous functions that compare the two lists. meant to be pasted in the erlang shell.

A=fun(L) ->
string:to_lower( % case insensitive
lists:flatten( % squash all characters back into one list
string:tokens(L,"-_") % return a list of list of characters
)
)
end.
fun([C|D],[C|E]) -> % are the first characters exactly the same?
A(D)==A(E); % does the rest compare correctly?
(_,_) -> % first chars not the same
a==b % shorter than 'false'
end.


# Clip, 25 bytes

&=(x(y=AxAy[Aa--m.La'-'_


Explanation:

x, y and z may be referenced in a Clip program to implicitly take up to three inputs. Since this program only references x and y, it takes two inputs which are assigned to x and y.

 =(x(y                    First characters of x and y are equal
&                         And
=AxAy               A(x) == A(y)
[Aa            Function A, takes parameter a
m.La     Map all elements of a to lower case
--     '-'_ Remove all occurrences of '-' and '_'


Takes two strings from standard input, outputs 1 and 0` for true and false respectively.