Code Golf Stack Exchange is a question and answer site for programming puzzle enthusiasts and code golfers. It only takes a minute to sign up.
Sign up to join this community
Apparently, we really like printing the alphabet in different languages.
How about the Russian Cyrillic alphabet?
Print/output/return this exact text (trailing whitespace is permitted)
АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя
This is the Russian Cyrillic alphabet with upper and lower case interleaved (eg AaBbCc...)
This is code-golf so crunch those bytes folks!
ВЫВОД: "АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя"
Just for fun, given the challenge. It may not be golfed; I'm not really familiar with the language.
Alphabet is a built-in function.
-1 byte from Martin.
ToUpperCase@#<>#&/@Alphabet@"Russian"<>""&
This may not work on some older versions of Mathematica, but it does work on Wolfram sandbox and Try it online!.
Another version without builtin Alphabet or ToUpperCase:
Insert[FromCharacterCode@{#,#+32}&/@(1039+Range@32),"Ёё",6]<>""&
or
Insert[FromCharacterCode@{#,#+32}&/@1040~Range~1071,"Ёё",6]<>""&
Unfortunately Martin's trick (use a<>b instead of {a,b}) doesn't work here.
Alphabet is a built-in. Freaking goats is a built-in.
\$\endgroup\$
Jun 21, 2017 at 12:18
40r71s6j25Q+ȷỌżŒl$
A full program. Pretty similar in nature to an existing answer by Erik the Outgolfer, but I think it's different enough.
40r71s6j25Q+ȷỌżŒl$ Main link: no arguments
40r71 inclusive range from 40 to 71: [40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71]
s6 split into sixes: [[40,41,42,43,44,45],[46,47,48,49,50,51],[52,53,54,55,56,57],[58,59,60,61,62,63],[64,65,66,67,68,69],[70,71]]
j25 join with 25s: [40,41,42,43,44,45,25,46,47,48,49,50,51,25,52,53,54,55,56,57,25,58,59,60,61,62,63,25,64,65,66,67,68,69,25,70,71]
Q de-duplicate: [40,41,42,43,44,45,25,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71]
ȷ literal 1000
+ addition: [1040,1041,1042,1043,1044,1045,1025,1046,1047,1048,1049,1050,1051,1052,1053,1054,1055,1056,1057,1058,1059,1060,1061,1062,1063,1064,1065,1066,1067,1068,1069,1070,1071]
Ọ cast to characters: "АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ"
$ last two links as a monad:
Œl lowercase: "абвгдеёжзийклмнопрстуфхцчшщъыьэюя"
ż zip: ["Аа","Бб","Вв","Гг","Дд","Ее","Ёё","Жж","Зз","Ии","Йй","Кк","Лл","Мм","Нн","Оо","Пп","Рр","Сс","Тт","Уу","Фф","Хх","Цц","Чч","Шш","Щщ","Ъъ","Ыы","Ьь","Ээ","Юю","Яя"]
implicit print: АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя
+ȷ but because of the s6(...)Q approach.
\$\endgroup\$
Jun 21, 2017 at 12:23
Note: Java is only valid if you replace => with ->
s=>"АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя"
Ahhhh!!! Found a shorter way for
a=>string.Concat(new string[32].Select((x,i)=>(x=(char)(i+1040)+"")+x.ToLower())).Insert(12,"Ёё")
Shortened it by using a different approach
() with a single character to save a byte.
\$\endgroup\$
;Bi";<=>?@," c+981 ®+v
;B :The uppercase alphabet
i :Prepend ...
";<=>?@," : this string
c :Map over the charcodes of each character in the string...
+981 : and increment them by 981
® :Map over each character in the string, replacing them with themselves...
+Zv : and their lowercase selfs
:Implicit output of resulting string
Ё & ё are missing. Also, you can save a couple of bytes with multiline and small tweak: ethproductions.github.io/japt/… EDIT: See you got those improvements already.
\$\endgroup\$
Ёё. You can do i"Ёё"B but that would cost too many bytes.
\$\endgroup\$
Ãi'ЁB to the end of the first line would be a byte shorter: ethproductions.github.io/japt/…
\$\endgroup\$
Ãi'Ё6, but yeah good idea.
\$\endgroup\$
71r40s26j25+ȷṚỌżŒl$
Thanks to Jonathan Allan for posting his answer (even though he outgolfs me) for -1 :)
for(;$i++<33;)echo$c=ڀ^"
".chr($i-7?16+$j++:1),mb_strtolower($c);
slight improvement over the previous attempt by xoring instead of appending.
ڀ^"\n" constructs the necessary "\0xd0\128" and xors in the lower 7 bit.
f=
_=>"АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ".replace(/./g,c=>c+c.toLowerCase())
console.log(f())
žv47ŸžB+çJ6т‚£'ЁýDløJ
Explanation
žv47Ÿ # push range [16 ... 47]
žB+ # add 1024 to each
çJ # convert to a string of characters with those code points
6т‚£ # split into 2 pieces, the first being 6 chars long
'Ёý # merge the 2 strings on "Ё"
Dlø # zip with a lower case copy
J # join to a string
l WORKS ON RUSSIAN ALPHABETS?! WHAT.
\$\endgroup\$
Jun 21, 2017 at 15:00
X12ssmCM+B+1040d32U32"Ёё
U32 generate the range [0, 1, 2, ..., 30, 31]
m d for each element in the range,
+1040 add 1040, giving the uppercase cyrillic letter
+B 32 bifurcate over adding 32, giving [upper, lower]
CM convert each into a character
ss join into a single string
X12 "Ёё insert this string at index 12
rB ... Z: X12ssmrBC+1040dZU32"Ёё
\$\endgroup\$
T„(GÇ«ŸçJ„ЁЖ6ǝDløJ
Credit to Emigna for the lowercase trick and for 3° isntead of 3Tm.
T # Push 10. | [10]
„(G # Push "(G" | [10, "(G"]
Ç # Convert to ASCII. | [10, [40, 71]]
« # Concatenate. | [[1040, 1071]]
Ÿ # Range. | [[1040, ..., 1071]
çJ # Convert to char, join. | ["А..Я"]
„ЁЖ # Push "ЁЖ" | ["А..Я", "ЁЖ"]
6ǝ # Replace "Ж" w/ "ЁЖ" | ["А.Ё.Я"]
Dl # Duplicate, lowercase. | ["А..Ё..Я", "а..ё..я"]
øJ # Zip together and join. | ["АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя"]
Alternate 20 byte version (no UTF-8): T„(GÇŸ„.Ç6ǝ˜«çJDløJ
Even more random 20 byte nonsense: T„(GÇŸ„.Ç6ǝ˜«vyçDlJ
(The key takeaway is that I don't think I'm getting below 20 here).
40 79 not awful.
\$\endgroup\$
Jun 21, 2017 at 15:26
"ABCDEF2GHIJKLMNOPQRSTUVWXYZ[\\]^_`"{#Ϗ+..#Ёe#P:# |+}>&@
Woo!
Adds Ϗ (975) to each character in ABCDEF2GHIJKLMNOPQRSTUVWXYZ[\]^_`
Then duplicates and adds space (32), unless the character is Ё, in which case it adds P (80)
Ϗ, since Ё is two bytes? I'm assuming UTF-8 - is that what's here?
\$\endgroup\$
Jun 21, 2017 at 15:05
Ϗ is Ё, and yes, Braingolf uses UTF-8
\$\endgroup\$
Ϗ, then that's a test against 2 (1 byte) rather than against Ё (2 bytes).
\$\endgroup\$
Jun 21, 2017 at 15:31
P though, otherwise it'll cost us a byte to swap the values
\$\endgroup\$
s->"АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя"
Source file should be saved in Windows-1251 encoding so the source makes for 71 bytes, plus flag to compiler -encoding WINDOWS-1251 for 22 bytes more.
javac always uses the default charset by default). Just mention that and you're good with only 71 bytes. But my answer even beats that ;)
\$\endgroup\$
Jun 22, 2017 at 9:52
-3 bytes using the form (using '%c' to directly apply chr()) from totallyhuman's answer, do go give some credit!
v=32
while v:print('Ёё'*(v==26),end='%c%c'%(1072-v,1104-v));v-=1
A full program (two double-byte characters used, Ё and ё).
8 bytes saved thanks to @Zacharý
⎕UCS 66⍴⍉2 33⍴1025+(14+⍳6),0,(20+⍳32),80,52+⍳26
žB> instead of 1025
\$\endgroup\$
_=>{var r="";for(var i='А';i<='Я';)r+=""+i+(char)(i+32)+(i++=='Е'?"Ёё":"");return r;}
@Mayube suggested this to save 2 bytes, and succeeded to golf it down by 1 byte by declaring chr(i).
@Rod golfed 4 bytes by removing chr(i).
for i in[chr(ord(i)+975)for i in"ABCDEF2GHIJKLMNOPQRSTUVWXYZ[\\]^_`"]:print(end=i+i.lower())
Any suggestions are welcome.
for i in"АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ":print(end=i+i.lower())
map(chr,range(65,91)) you have "ABC..Z" in 21 byte instead of 29. // So we have: i in map(chr,range(65,91))+"[\\]^_'" (37byte) vs i in"ABCDEF2GHIJKLMNOPQRSTUVWXYZ[\\]^_"` (40byte).
\$\endgroup\$
Ё and hence the 2 placed between F and G, which mapping chr over range does not yield (see my Python answer for a different workaround).
\$\endgroup\$
Jun 21, 2017 at 12:30
1040ÐU5+ŸžB>X6+D25+Ÿ)˜çDl)ø˜J
Uses the lowercase / uppercase trick from the python answer
1040ÐU5+ŸžB>X6+D25+Ÿ)˜çDl)ø˜J
1040 # Push 1040
Ð # Triplicate top of stack
U # Saves top of stack in X
5+Ÿ # Push range [1040, ..., 1045]
žB> # Push 1025
X # Push X (1040)
6+D25+Ÿ # Push range [1046, ..., 1071]
) # Wrap stack to array
˜ # Flatten array
ç # Convert to character
Dl # Create a lowercase copy
)ø # Zip both together
˜J # Flatten and join
# Implicit output
for($t=IntlChar;$i++<33;)echo$c=$t::chr($i-7?1040+$j++:1025),$t::tolower($c);
for(;$i<mb_strlen($r=АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ);)echo$m=mb_substr($r,$i++,1),mb_strtolower($m);
<?=АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя;
printf %b\\n \\u04{51,{3,4}{{0..9},{a..f}}}|sort|sed 's/./\u&&/'|tr -d \\n
This requires a UTF-8 locale such as ru_RU.utf8 (English UTF-8 locales also sort correctly). I've not added any score for this, as per the consensus on meta.
u=?Џ
l=?Я
32.times{$><<u.next!+l.next!+(l==?е?"Ёё":"")}
(Just a demonstration of String#next!, the multibyte literals ruin the 55 characters score.)
Sample run:
bash-4.4$ ruby -e 'u=?Џ;l=?Я;32.times{$><<u.next!+l.next!+(l==?е?"Ёё":"")}'
АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя
32.times{|x|$><<((x==6?"Ёё":"")<<x+1040<<x+1072)}
Thanks to manatwork for at least 4 bytes.
22 bytes of code (encoded in UTF-8), +2 for the -P flag.
#h0o#k2_dÃi'Ё6
®+Zv
Similar to the below method, except instead of pairing two arrays, I join each capital letter with itself, lower-cased.
I had some help from @Shaggy, who discovered that Ё wasn't in the [1040...1071] char range.
#h0o#k2_dÃi'Ё6
íUmv)c
#h0o#k2_dÃi'Ё6
íUmv)c
// Implicit U =
o // Range
#h0 // [1040...
#k2 // 1071]
_ // Map; At each char,
d // Get the char-code
à // End mapping;
i'Ё6 // Insert "Ё" into index 6;
í // U paired with:
Um // U, where each char:
v // Converted to lowercase
c // Flatten
-P // Join into a string
ＦＥ⮌⪫⪪…⮌…γ⁷²¦³²Ｌαi℅⁺℅ιφ⁺↥ιι
Try it online! Link is to verbose version of code. Explanation:
ＦＥ⮌⪫⪪…⮌…γ⁷²¦³²Ｌαi
Take the first 72 printable ASCII characters, reverse them, then take the first 32 of those (this is actually shorter than slicing and reversing, although slicing didn't exist at original time of posting anyway), then insert the i after the N, then reverse again. The resulting characters happen to have ASCII codes that are exactly 1000 less than the desired Russian lower case letters.
℅⁺℅ιφ
Calculate the Russian lower case letter by adding the predefined variable φ (f) which has the value 1000.
⁺↥ιι
Print the upper and lower case letters. («↥ιι would probably also work.)
gr'v6R,' &w'ѷ~-.O' +'С'ёSOk@
g Read the first character of the source code. That's g itself, with
code point 103, so this pushes 103. The exact value is irrelevant
but we need something greater than 31.
r Range. Turns 103 into 0, 1, 2, ..., 103.
'v Push 118.
6R, Move it down the stack by 6 elements, so the top of the stack is
..., 96, 97, 118, 98, 99, 100, 101, 102, 103.
' &w Push the current IP address to the return address stack 32 times,
which will cause the following part to be executed 33 times.
'ѷ~- Subtract the value on top of the stack from 1143, which gives the
33 code points of the upper case letters.
.O Print a copy of this value as a character.
' + Add 32. This turns the upper case letters in the main range to their
lower case counterparts. However, this doesn't work for Ё/ё, where
we obtain C instead.
'C'ёS Replace the divisor 1057 with 1105. The only value among the 33 which
is divisible by 1057 is 1057 (C) itself, so that it simply gets turned into
1105 (ё).
O Output the lower case character.
k Jump back to the w as long as there's a copy on the return address stack.
@ Terminate the program.
()->{String s="";for(char i=0,u=1039,l=1071;i<33;)s=i++==6?s+"Ёё":s+ ++u+ ++l;return s;}
Encoded in Windows-1251. Must use a default Russian Windows installation.
i++ and change the code accordingly to save 2 bytes: ()->{String s="";for(char i=0,u=1040,l=1072;i<33;i++)s=i==5?s+"Ёё":s+ ++u+ ++l;return s;}
\$\endgroup\$
Jun 22, 2017 at 10:19
i is only a counter. I intended to use it otherwise at first.
\$\endgroup\$
Jun 22, 2017 at 10:19
/ZАе"е"ёяSr@
\"'Я"rЁа""o/
My other Alice solution was entirely in Cardinal mode, this one is entirely in Ordinal mode. It's conceptually much simpler, but unfortunately two byte longer.
Ignoring the Ordinal template, the program is entirely linear:
"АЯ"r"ая"rZ'е"еЁё"So@
Alice does have a built-in to convert between lower and upper case, but unfortunately in the Ruby interpreter it only works on ASCII letters, so we still have to get both ranges separately.
"АЯ"r Push "АБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ".
"ая"r Push "абвгдежзийклмнопрстуфхцчшщъыьэюя".
Z Zip. Interleave the two strings.
'е"еЁё"S Replace "е" with "еЁё".
o@ Print the result and terminate the program.
Ёё, may we output codepoints U+00CB and U+00EB instead of U+0401 and U+0451? \$\endgroup\$