Just another traffic jam!

Question

One lane is closed, the vehicles have to thread their way in according to the zip-lock system: One car from one lane, the next car from the other lane, always taking turns.

Input: Two vehicle queues, represented e.g. as strings, in which short cars are represented by individual symbols, lorries by several identical symbols. The queues may be of different lengths and may even be empty. Symbols may appear more than once. If your language does not support strings, you can also use two lists (see second set of test data) or other forms of input as long as the purpose of the task is not lost.

Output: A vehicle queue in the same representation. The traffic jam is long enough, so at least the source code should be as short as possible!

Test data:

Nzzll,iaaee                   --> Nizzaallee
aabcddddeeeee,1112345555611   --> aa111b2c3dddd4eeeee5555611
foo+bar ooooooooooooooooooooo --> fooooooooooooooooooooooo+bar
feeeuve553iiiiiiiiiii,        --> feeeuve553iiiiiiiiiii
UvtMCaegIYuiet,poeyhllnefoLkI --> UpvoteMyChallengeIfYouLikeIt

Test data for languages that do not support strings:

[1,3,5,7,9,11],[2,4,6,7,7,7,7,7] --> [1,2,3,4,5,6,7,7,7,7,7,7,9,11]
[4,9,9,13,13],[41,41,41,1,11,44] --> [4,41,41,41,9,9,1,13,13,11,44]
[] [2,6,4,3,3,56,8,5,4,3]        --> [2,6,4,3,3,56,8,5,4,3]

Answer 1

Jelly, 3 bytes

ŒgZ

A full program that accepts a pair of the two strings as the first argument and prints the resulting single lane.

Try it online!

How?

ŒgZ - Main Link: pair of lists of characters
Œg  - group runs of equal elements of each
  Z - transpose
    - implicit, smashing print

Answer 2

Ruby, 33 bytes

f=->a,b{a[/(.)\1*/]?$&+f[b,$']:b}

Try it online!

How it works:

If the first queue is not empty, extract the first vehicle, then exchange the queues, and repeat.

If the first queue is empty, return the whole second queue.

Answer 3

05AB1E, 5 bytes

€γõζS

Input as a pair of strings; output as a list of characters.

Try it online or verify all test cases.

Explanation:

€      # Map over both strings in the (implicit) input-pair:
 γ     #  Split them into equal adjacent groups of substrings
   ζ   # Zip/transpose this list of lists,
  õ    # using an empty string ("") as filler for unequal length lists
    S  # Convert this list of pairs of strings to a flattened list of characters
       # (which is output implicitly as result)

Answer 4

Uiua, 16 bytes

/◇⊂♭⍉⬚∘⊟□[]∩⊜□∩.

Try it!

Takes lists of postive integers as input (because this saves a few bytes over strings).

• ∩. duplicate both inputs

• ∩⊜□ split both into contiguous equal elements

• ⬚∘⊟□[] couple lists to matrix, filling with boxed empty list

• ⍉ transpose

• ♭ deshape

• /◇⊂ join all

Answer 5

Wolfram Language (Mathematica), 26 bytes

Flatten[Split/@#,{2,1,3}]&

Try it online!

Input and output lists of characters.

Answer 6

Perl 5 + -p0, 28 bytes

1 byte saved thanks to @Xcali!

s/^(.)\1*/!print$&/gem&&redo

Try it online!

Answer 7

Charcoal, 35 bytes

Ｆ²⊞υ⪪⮌Ｓ¹Ｗ∧⌊υ⊟§υ⁰«ι¿⌕⮌§υ⁰ι≔⮌υυ»Ｗ⌈υ⊟ι

Try it online! Link is to verbose version of code. Explanation: Seems to be a similar idea to @l4m2's answer, but it just took me that long to think of it after writing my Retina answer that he got there first.

Ｆ²⊞υ⪪⮌Ｓ¹

Input the two strings, reversing them and splitting them into characters so that they can be processed more easily.

Ｗ∧⌊υ⊟§υ⁰«

Until one queue is empty, remove the next character from the first queue, ...

ι

... print it, and...

¿⌕⮌§υ⁰ι

... if the next character in this queue is from a different vehicle, then...

≔⮌υυ

... switch the queues.

»Ｗ⌈υ⊟ι

Output the remaining vehicles from the other queue.

Answer 8

JavaScript (Node.js), 52 48 bytes

f=([c,...a],b)=>c?[c,...c==a[0]?f(a,b):f(b,a)]:b

Attempt This Online!

Array IO, though inputting string usually work

Answer 9

J, 32 bytes

[:;(,/:,&(#\))&(<;.1~2~:/\'_'&,)

Attempt This Online!

The 2nd part of this is the only interesting insight:

• [:;...&(<;.1~2~:/\'_'&,) This is the preprocessing step that breaks the lists down into the grouped sections. A standard J idiom which is unfortunately a bit verbose. I could save two bytes by using integers instead of strings as input.
• ,/:,&(#\) The main logic. Since we can't zip uneven lists in J using ,., we just cat the lists , and then sort them /: by 1..n catted with 1..m (where n and m are the two list lengths). This works because sort is stable.

Answer 10

Noulith, 97 bytes

f:=\a,b->if (a.len<1)b else (c,d:=a.first,a.tail;c$if (d.len>0 and c==d.first)f(d,b)else f(b,d))

Try it online!

Explanation

f:=\a,b ->                  # recursive function
if (a.len<1) b              # return early when first input is empty
else (                      # otherwise
c,d:=a.first,a.tail;        # get first c(h)ar and the rest
c $                         # concat the first c(h)ar with
if (d.len>0 and c==d.first) # if next is a truck
f(d,b)                      # the rest of the truck
else f(b,d))                # otherwise the next lane

Answer 11

C (clang), 61 60 bytes

f(a,b)char*a,*b;{*a?putchar(*a)-*++a?f(b,a):f(a,b):puts(b);}

Try it online!

• saved 1 Byte thanks to @l4m2

Answer 12

R, 155 bytes

\(r,s,`!`=length,`?`=rep,`/`=\(x,n)rle(el(strsplit(x,"")))[[n]],`*`=\(x,n)c(x/n,0?max(!r/1,!s/1)-!x/1))o=Map(\(a,b,d,e)cat(a?b,d?e,sep=""),r*2,r*1,s*2,s*1)

Attempt This Online!

The solution is based on rle function ("run length encoding") which transforms a character vector into two vectors by counting and removing the repeated adjacent characters.

The smaller vectors output by rle are matched with 0's in order to have the same length as the larger ones and to avoid the recycling of the characters. Then the elements of the both character vectors are restored back to their original lengthes with the rep and output one after one in the zip-lock fashion. Note that added extra elements rep("0",0) result in character(0) which are ignored by cat.

Answer 13

Vyxal s, 22 bits^v2, 2.75 bytes

vĠ∩

Try it Online!

Bitstring:

1000110100011111100101

The footer inserts a newline for test case purposes. Port of the jelly answer.

Explained

vĠ∩­⁡​‎‎⁡⁠⁡‏⁠‎⁡⁠⁢‏‏​⁡⁠⁡‌⁢​‎‎⁡⁠⁣‏‏​⁡⁠⁡‌­
vĠ   # ‎⁡Vectorise group-by
  ∩  # ‎⁢Transpose
💎

Created with the help of Luminespire.

Answer 14

JavaScript (ES11), 62 bytes

Expects ([str1, str2]).

a=>[...a+a].map((_,i)=>a[i&1].match(/(.)\1*/g)?.[i>>1]).join``

Attempt This Online!

Answer 15

Retina 0.8.2, 26 bytes

O$#`.(?<=(\2*(.))+)
$#1
¶

Try it online! Takes input on separate lines but link is to test suite that splits on commas for convenience. Explanation:

O$#`.(?<=(\2*(.))+)
$#1

Sort each character by its vehicle index in its queue. This is a stable sort so that the first vehicle of the first queue sorts first followed by the first vehicle of the second queue etc. The newline is not considered for the sort but it's used by the lookbehind to find the front of the second queue.

¶

The vehicles were sorted without regard for the placement of the newline which is just unnecessary at this point.

Answer 16

Go, 133 bytes

func f(a,b string)string{if len(a)<1{return b}
c,a:=a[0],a[1:]
if len(a)>0&&c==a[0]{return string(c)+f(a,b)}
return string(c)+f(b,a)}

Attempt This Online!

Recursive function.

Explanation

func f(a,b string)string{
if len(a)<1{return b}    // return early when first input is empty
c,a:=a[0],a[1:]          // get the current c(h)ar
if len(a)>0&&c==a[0]{    // if the next c(h)ar is the same...
return string(c)+f(a,b)} // recurse to get the truck
return string(c)+f(b,a)} // otherwise move on to the next lane

Answer 17

BQN, 12 bytes

∾∾○(+`»⊸≠)⊔∾

Try it online!

Input 𝕨F𝕩. Assumes the leading character isn't a null character.

                            traffic jaaam
           ∾    join        trafficjaaam
          ⊔     group by
    +`»⊸≠        vehicle #s
 ∾○              (joined)   123445612223
                            tj raaa am ff i c
∾               join groups tjraaaamffic

Flattened group-by (∾⊔) is a little shorter than sort-by (⍋⊸⊏).

Answer 18

Japt -P, 6 bytes

ËòÀÃÕc

Try it

ËòÀÃÕc     :Implicit input of string array
Ë          :Map
 ò         :  Partition between characters
  À        :    That are not equal
   Ã       :End map
    Õ      :Transpose
     c     :Flatten
           :Implicitly join and output

Answer 19

x86-64 machine code, 21 bytes

AA 3A 06 74 03 48 87 D6 AC 84 C0 75 F3 52 5E 3A 06 A4 75 FB C3

Try it online!

Following the standard calling convention for Unix-like systems (from the System V AMD64 ABI), this takes in RDI an address at which to place the result, as a null-terminated byte string, and in RSI and RDX, addresses of the inputs, as null-terminated byte strings. The starting point is after the first 8 bytes.

In assembly, rearranged for easier reading:

f:  lodsb           # Load a byte from address RSI into AL, advancing the pointer.
    test al, al     # Check the value of that byte.
    jnz r           # Jump if it's nonzero.

(This section actually comes first.)
r:  stosb           # Write that byte to the output string, advancing the pointer.
    cmp al, [rsi]   # Compare that byte to the next byte in the string.
    je f            # Jump if they are equal.
    xchg rsi, rdx   # (Otherwise) Swap the pointers (in RSI and RDX).
(From here, continue at the beginning.)

(When the byte is zero, continue to here.)
    push rdx; pop rsi   # Transfer the other pointer from RDX to RSI.
r1: cmp al, [rsi]       # Compare the value at that address to AL, which is 0.
    movsb               # Write the byte from address RSI to the output string,
                        #  advancing both pointers.
    jne r1              # Jump back if the byte was not equal to 0.
    ret                 # Return.

Answer 20

Bash, 103 99 bytes

The function. Input is parameters 1 and 2. Output is variable o.

f(){
o=  
while [ "$1$2" ]
do  
c=`expr "$1" : "${1::1}*"`
o+=${1::$c}
set -- "$2" "${1:$c}"
done
}

Try it online!

Uses expr to determine the repeats of the first character, "outputs" the repeats, swaps the strings, and repeats until empty.

Due to limitations in expr, it can't handle characters [, \, or ^, or trailing single parentheses. All of these cause infinite loops, possibly with errors spewing. See the TIO for details.

I could save two characters by removing the quotes on the while statement, but this would make space not be a valid character.

Answer 21

Python, 160 159 bytes

from itertools import*
c,g,l=chain.from_iterable,groupby,list;f=lambda a,b:''.join(c(c(zip_longest([l(t)for r,t in g(a)],[l(t)for r,t in g(b)],fillvalue=''))))

Attempt This Online!

Edit 1: removed an extra space

Answer 22

GNU sed -E,  63  59 bytes

The first approach did build a third string behind the two strings to be merged, but it turned out it's cheaper to invest in introducing a mark _ in the first string marking the position where to insert the chars from the second string:

s/^/_/
:1
s/_((.?)\2*)(.*,)((.?)\5*)/\1\4_\3/
/_,/!b1
s///g

The magic mainly takes place in the third line: The ERE (.?)\2* (and the same with 5) match zero or more occurences of the same character (using a backreference, which is not required for ERE by POSIX, but supported by the GNU implementation. Thus, we move those matches in front of the _.

We can't use t for looping, because an empty substitution is always possible, but that's not too expensive, because the addess pattern _, can be reused for removing.

Try it online!

Answer 23

Desmos, 96 bytes

C=sgn(L[2...]-L)^2
g(L)=∑_{n=1}^{[1...C.count]}C[n]
f(A,B)=sort(join(A,B),join(0,g(A),0,g(B)))

Ok this can probably be improved somehow but for now this is good enough.

Try It On Desmos!

Try It On Desmos! - Prettified

Answer 24

Python 3.8 (pre-release), 59 bytes

f=lambda a,b:a+b and a[:a.rfind(s:=a.lstrip(a[:1]))]+f(b,s)

Try it online!