Write an IBAN validator

Write a program (the body of the function is enough) that accepts a string of alphanumeric characters and validates it according to ISO 13616:2007. The algorithm for validation is (source: wikipedia article on IBAN http://en.wikipedia.org/wiki/International_Bank_Account_Number):

Validating the IBAN An IBAN is validated by converting it into an integer and performing a basic mod-97 operation (as described in ISO 7064) on it. If the IBAN is valid, the remainder equals 1.The algorithm of IBAN validation is as follows:

• Check that the total IBAN length is correct as per the country. If not, the IBAN is invalid. The correct IBAN lengths can be found here: http://pastebin.com/kp4eECVk (can also be found below), ordered by the length of the IBAN number. the first 2 characters of each number is the country code. All other characters (lowercase letters on the pastebin, but can be any case in the actual IBAN) behind the first 2 can be any alphanumeric character.
• Move the four initial characters to the end of the string.
• Replace each letter in the string with two digits,thereby expanding the string, where A = 10, B = 11, ..., Z = 35.
• Interpret the string as a decimal integer and compute the remainder of that number on division by 97

If the remainder is 1, the check digit test is passed and the IBAN might be valid.

Example (fictitious United Kingdom bank, sort code 12-34-56, account number 98765432):

- IBAN:               GB82 WEST 1234 5698 7654 32
- Rearrange:          W E S T12345698765432 G B82
- Convert to integer: 3214282912345698765432161182
- Compute remainder:  3214282912345698765432161182    mod 97 = 1


The algorithm moest return true (or a truthy value) if the number is valid, and false (or a falsy value) if the number is invalid according to the algorithm. You do not need to check if the number actually exists, only if it's valid. The algorithm has to work for each of the different approved IBAN numbers mentioned in the wikipedia article above. The algorithm should be compatible with numbers both with or without separator characters between 2 alphanumeric characters. The separator character can either be periods, spaces or dashes and one number can contain different types of separators.

Usual loopholes apply: no external resources, no built-in functions or methods.

Puzzle type is code golf. Shortest code in bytecount wins. Standard boilerplate needed to execute the program (e.g. namespace, class, function declaration in OOP) is not included in the bytecount.

Bonus: if you can return the properly formatted IBAN number (according to the national format on wikipedia) instead of true in the case of a valid number, you get a 25% reduction in your score. If the number is invalid, you return a literal string with the value "Invalid".

Copy of the IBAN lengths in case the pastebin ever is removed:

Country;Chars;IBAN Fields
Norway;15;NOkk bbbb cccc ccx
Belgium;16;BEkk bbbc cccc ccxx
Burundi;16;BIkk nnnn nnnn nnnn
Denmark;18;DKkk bbbb cccc cccc cc
Faroe Islands;18;FOkk bbbb cccc cccc cx
Finland;18;FIkk bbbb bbcc cccc cx
Greenland;18;GLkk bbbb cccc cccc cc
Netherlands;18;NLkk bbbb cccc cccc cc
Macedonia;19;MKkk bbbc cccc cccc cxx
Slovenia;19;SIkk bbss sccc cccc cxx
Austria;20;ATkk bbbb bccc cccc cccc
Bosnia and Herzegovina;20;BAkk bbbs sscc cccc ccxx
Estonia;20;EEkk bbss cccc cccc cccx
Kazakhstan;20;KZkk bbbc cccc cccc cccc
Lithuania;20;LTkk bbbb bccc cccc cccc
Luxembourg;20;LUkk bbbc cccc cccc cccc
Costa Rica;21;CRkk bbbc cccc cccc cccc c
Croatia;21;HRkk bbbb bbbc cccc cccc c
Latvia;21;LVkk bbbb cccc cccc cccc c
Liechtenstein;21;LIkk bbbb bccc cccc cccc c
Switzerland;21;CHkk bbbb bccc cccc cccc c
Bahrain;22;BHkk bbbb cccc cccc cccc cc
Bulgaria;22;BGkk bbbb ssss ddcc cccc cc
Georgia;22;GEkk bbcc cccc cccc cccc cc
Germany;22;DEkk bbbb bbbb cccc cccc cc
Ireland;22;IEkk aaaa bbbb bbcc cccc cc
Montenegro;22;MEkk bbbc cccc cccc cccc xx
Serbia;22;RSkk bbbc cccc cccc cccc xx
United Kingdom;22;GBkk bbbb ssss sscc cccc cc
Gibraltar;23;GIkk bbbb cccc cccc cccc ccc
Israel;23;ILkk bbbn nncc cccc cccc ccc
United Arab Emirates;23;AEkk bbbc cccc cccc cccc ccc
Andorra;24;ADkk bbbb ssss cccc cccc cccc
Czech Republic;24;CZkk bbbb ssss sscc cccc cccc
Moldova;24;MDkk bbcc cccc cccc cccc cccc
Pakistan;24;PKkk bbbb cccc cccc cccc cccc
Romania;24;ROkk bbbb cccc cccc cccc cccc
Saudi Arabia;24;SAkk bbcc cccc cccc cccc cccc
Slovakia;24;SKkk bbbb ssss sscc cccc cccc
Spain;24;ESkk bbbb gggg xxcc cccc cccc
Sweden;24;SEkk bbbc cccc cccc cccc cccx
Tunisia;24;TNkk bbss sccc cccc cccc cccc
Virgin Islands;24;VGkk bbbb cccc cccc cccc cccc
Algeria;24;DZkk nnnn nnnn nnnn nnnn nnnn
Portugal;25;PTkk bbbb ssss cccc cccc cccx x
Angola;25;AOkk nnnn nnnn nnnn nnnn nnnn n
Cape Verde;25;CVkk nnnn nnnn nnnn nnnn nnnn n
Mozambique;25;MZkk nnnn nnnn nnnn nnnn nnnn n
Iceland;26;ISkk bbbb sscc cccc iiii iiii ii
Turkey;26;TRkk bbbb bxcc cccc cccc cccc cc
Iran;26;IRkk nnnn nnnn nnnn nnnn nnnn nn
France;27;FRkk bbbb bggg ggcc cccc cccc cxx
Greece;27;GRkk bbbs sssc cccc cccc cccc ccc
Italy;27;ITkk xaaa aabb bbbc cccc cccc ccc
Mauritania;27;MRkk bbbb bsss sscc cccc cccc cxx
Monaco;27;MCkk bbbb bsss sscc cccc cccc cxx
San Marino;27;SMkk xaaa aabb bbbc cccc cccc ccc
Burkina Faso;27;BFkk nnnn nnnn nnnn nnnn nnnn nnn
Cameroon;27;CMkk nnnn nnnn nnnn nnnn nnnn nnn
Madagascar;27;MGkk nnnn nnnn nnnn nnnn nnnn nnn
Albania;28;ALkk bbbs sssx cccc cccc cccc cccc
Azerbaijan;28;AZkk bbbb cccc cccc cccc cccc cccc
Cyprus;28;CYkk bbbs ssss cccc cccc cccc cccc
Dominican Republic;28;DOkk bbbb cccc cccc cccc cccc cccc
Guatemala;28;GTkk bbbb cccc cccc cccc cccc cccc
Hungary;28;HUkk bbbs sssk cccc cccc cccc cccx
Lebanon;28;LBkk bbbb cccc cccc cccc cccc cccc
Poland;28;PLkk bbbs sssx cccc cccc cccc cccc
Benin;28;BJkk annn nnnn nnnn nnnn nnnn nnnn
Ivory Coast;28;CIkk annn nnnn nnnn nnnn nnnn nnnn
Mali;28;MLkk annn nnnn nnnn nnnn nnnn nnnn
Senegal;28;SNkk annn nnnn nnnn nnnn nnnn nnnn
Brazil;29;BRkk bbbb bbbb ssss sccc cccc ccct n
Palestinian;29;PSkk bbbb xxxx xxxx xccc cccc cccc c
Qatar;29;QAkk bbbb cccc cccc cccc cccc cccc c
Ukraine;29;UAkk bbbb bbcc cccc cccc cccc cccc c
Jordan;30;JOkk bbbb nnnn cccc cccc cccc cccc cc
Kuwait;30;KWkk bbbb cccc cccc cccc cccc cccc cc
Mauritius;30;MUkk bbbb bbss cccc cccc cccc cccc cc
Malta;31;MTkk bbbb ssss sccc cccc cccc cccc ccc

• To whoever suggested to add perl and python tags: language tags in challenges should be used to confine the challenge to a certain language (or at least, I think that's the intent of them). This challenge is open to all languages. As such, I rejected the edit. Commented Aug 1, 2014 at 20:03

J (294 - 73.5 = 220.5)

I've not counted the function definition (f=:3 :0...)) because it can be considered boilerplate, counting the whole block gives a score of 304 - 76 = 228.

f=:3 :0
)


Tests:

   NB. invalid numbers
f ''
Invalid
f 'GB82 WEST 1234 5698 7654 31'
Invalid
f 'NL82 WEST 1234 5698 7654 32'
Invalid

NB. different separators and formatting
f 'GB82.WEST.1234.5698.7654.32'
GB82 WEST 1234 5698 7654 32
f 'GB82-WEST-1234-5698-7654-32'
GB82 WEST 1234 5698 7654 32
f 'GB82WEST12345698765432'
GB82 WEST 1234 5698 7654 32
f 'GB82 WEST 1234 5698 7654 32'
GB82 WEST 1234 5698 7654 32
f 'GB.82-WE ST-12.345698.76-5432'
GB82 WEST 1234 5698 7654 32

NB. wikipedia examples
f 'GR16 0110 1250 0000 0001 2300 695'
GR16 0110 1250 0000 0001 2300 695
f 'CH93 0076 2011 6238 5295 7'
CH93 0076 2011 6238 5295 7
f 'SA03 8000 0000 6080 1016 7519'
SA03 8000 0000 6080 1016 7519
f 'GB29 NWBK 6016 1331 9268 19'
GB29 NWBK 6016 1331 9268 19


Explanation:

• b=.y-.' -.': remove any separators from the argument and store the result in b.
• 1}.;' '&,&.>_4<\b: split b up in groups of four, add a space in front of every group, join the groups and remove the leading space. If y contained a valid IBAN number, this is its canonical representation (i.e. groups of four, separated by spaces, with the last group maybe having less than four elements).
• (...){'Invalid';: create an array with the string Invalid as element 0 and the formatted IBAN number as element 1. Select the right one based on whether the IBAN number is valid:

• Check the length:
• 'NO.BEBI.---.JOKWU.MT.': a list of all the country codes for each length, separated by dots
• +/"1(2{.b)&E.;.2: group the string by the dots, and see which one contains the given country code (the first 2 elements of b).
• 15+1 i.~: find the index of the matching one and add 15 to find the length.
• (#b)=: check it against the actual length of b.
• Check the number:
• 4|.b: rotate b to the left by 4 (rearrange)
• k=.3&u:: find the ASCII value for each number
• 48-~k-7*64<k: subtract 7 from each letter, then subtract 48 from all, giving the values
• 1".'x',~' '-.~":: format, remove spaces, add an 'x' at the end (for high-precision mode, which is necessary for large numbers), and turn it back into a number
• 1=97|: check if the number mod 97 equals 1.
• >: unbox the resulting string

CJam, 151.5 141.75 points

0000000: 6c22 202e 2d22 2d3a 512c 220f a86e 5136  l" .-"-:Q,"..nQ6
0000010: 1bff 75f6 e8e4 0b35 5dab f190 0d85 59c4  ..u....5].....Y.
0000020: 1938 4366 3d29 5eaa e879 024a 77d9 8baf  .8Cf=)^..y.Jw...
0000030: 5c16 3258 a4d2 4e6c 1a60 429f affa b8f4  \.2X..Nl.B.....
0000040: 435d e706 457b 89a9 16b8 1d4b 08f7 9970  C]..E{.....K...p
0000050: eeb9 7467 f8e9 c935 33be 2467 3dd4 1afb  ..tg...53.$g=... 0000060: e2ec 20cc 99e4 2783 cb96 512d f9f8 7e75 .. ...'...Q-..~u 0000070: 7066 4307 2232 3536 6232 3762 2740 662b pfC."256b27b'@f+ 0000080: 2740 2f7b 5132 3c23 3126 217d 235f 573e '@/{Q2<#1&!}#_W> 0000090: 462a 2b3d 5134 6d3c 412c 7327 5b2c 3635 F*+=Q4m<A,s'[,65 00000a0: 3e2b 6623 737e 3937 2531 3d5d 312d 2249 >+f#s~97%1=]1-"I 00000b0: 6e76 616c 6964 2251 342f 532a 3f nvalid"Q4/S*?  The above program is 189 bytes long and qualifies for the bonus. At the cost of 26 more bytes – for a total score of 161.25 – we can avoid unprintable characters: l" .-"-:Q,",YER,moTd$V6nH\-Mh/I-z(]k!uw()=j9_[C3n&As0(F;TAn$eB-r%:p+^b,1Y.j;thavi@!d,Dt7M_x-5V:#o/m_CKj-c*Imy~IjXPBCo?aM#lrN:[email protected]"33f-94b27b'@f+'@/{Q2<#1&!}#_W>F*+=Q4m<A,s'[,65>+f#s~97%1=]1-"Invalid"Q4/S*?  You can test this version in the CJam interpreter. Example run $ cat gen.cjam
"l\" .-\"-:Q,"[32,15>{[
"NO"15"BE"16"BI"16"DK"18"FO"18"FI"18"GL"18"NL"18"MK"19"SI"19
"AT"20"BA"20"EE"20"KZ"20"LT"20"LU"20"CR"21"HR"21"LV"21"LI"21
"CH"21"BH"22"BG"22"GE"22"DE"22"IE"22"ME"22"RS"22"GB"22"GI"23
"SE"24"TN"24"VG"24"DZ"24"PT"25"AO"25"CV"25"MZ"25"IS"26"TR"26
"IR"26"FR"27"GR"27"IT"27"MR"27"MC"27"SM"27"BF"27"CM"27"MG"27
"AL"28"AZ"28"CY"28"DO"28"GT"28"HU"28"LB"28"PL"28"BJ"28"CI"28
"ML"28"SN"28"BR"29"PS"29"QA"29"UA"29"JO"30"KW"30"MU"30"MT"31
]2/{1=L=},0f=_!!{:+}*}fL]"@"*'@f-27b256b:c"\\\\"/"\\"*
"256b27b'@f+'@/{Q2<#1&!}#_W>F*+=Q4m<A,s'[,65>+f#s~97%1=]1-\"Invalid\"Q4/S*?"
$LANG=en_US cjam gen.cjam | tee >(cksum) > iban.cjam 770150303 189$ LANG=en_US cjam iban.cjam <<< GB82WEST12345698765432; echo
GB82 WEST 1234 5698 7654 32


How it works

"…"256b27b'@f+"


converts the string "…" into an integer by considering it a base-256 number, then into an array of integers by considering it a base 27-number, adds the character code of @ to each digit and casts to Character in the process.

As a result, the following code gets executed:

" Read one line from STDIN, remove allowed separators, store the result in variable
“Q” and push its length (“,”).                                                      ";

l" .-"-:Q,

" Push the string from above.
The correct number of characters for country code “NO” is 15.
The correct number of characters for country codes “BE” and “BI” is 16.
The number of characters should never be 17.
⋮                                                                                  ";

"NO@BEBI@@DKFOFIGLNL@MKSI@ATBAEEKZLTLU@CRHRLVLICH@BHBGGEDEIEMERSGB@"
"ALAZCYDOGTHULBPLBJCIMLSN@BRPSQAUA@JOKWMU@MT"++

" Split the above string at the at signs (“@”).                                       ";

'@/

" Find the index of the (first) substring such that the index of the country code
(“Q2<”) in the substring is even.                                                   ";

{Q2<#1&!}#

" If the country code is invalid, the index will be -1. If the index is greater than
-1 (“W>”), add 15 to it; if it isn't, leave -1 on the stack.                        ";

_W>F*+

" Compare the result to the length of the IBAN. Push 1 if they match and 0 otherwise. ";

=

" Push the IBAN rotated to the left by four characters.                               ";

Q4m<

" “A,s'[,65>+” pushes “0123456789ABCDEFGHIJKLMNOPQRSTUVXYZ”. For each character of
the IBAN, push its index in that string (-1 if not present).                       ";

A,s'[,65>+f#

" Stringify (“s”) the result, interpret the string (“~”) and compute the residue of
the (topmost) resulting integer modulo 97.                                          ";

s97%

" Push 1 if the residue is 1; otherwise, push 0                                       ";

1=

" Wrap the entire stack in an array and remove the integer 1 from it. If the IBAN is
valid, the array will be empty.

" If the result is falsy, the IBAN is valid; split it into substrings of length 4
and join them using spaces (“S”). Otherwise, the IBAN is invalid; say so.           ";

"Invalid"Q4/S*?

• LANG=en_US java -jar cjam-0.6.2.jar iban.cjam<<<GB82WEST12345698765432 outputs Invalid, what am I doing wrong (running on Windows8.1 and Cygwin)?
– user16402
Commented Aug 2, 2014 at 8:50
• @professorfish: I don't have a Windows computer, so I can't check from this end, but let's start by verifying the following: 1. Did iban.cjam get generated successfully? md5sum iban.cjam should print 1960c33e31ae5646cd0400826757b3bc. 2. Is the locale installed properly? You can check by executing locale -a | grep en_US. Commented Aug 2, 2014 at 14:00
• It hasn't generated properly (fb620d509887f1a7298c3e5ff312401a). locale -a|grep en_US outputs en_US and en_US.utf8
– user16402
Commented Aug 2, 2014 at 14:57
• @professorfish: Interesting. You could try this to avoid the iban.cjam file: LANG=en_US cjam <(LANG=en_US cjam gen.cjam) <<< GB82WEST12345698765432. I've included an ASCII-only version in my answer that should be easier to verify. Commented Aug 2, 2014 at 15:12

Bash, 738 519 444 434 427

Here's something to start us off, I wrote it (and saved 317 characters, mostly in the country code storage bit) while the question was in the sandbox. Let me know if there are any issues.

The function reads from stdin, which is quite common in Bash (the question says "accepts a string of alphanumeric characters", it doesn't require this to be via arguments).

It returns 0 if the IBAN is valid and a non-zero value if it is invalid.

An IBAN containing any characters other than the delimiters . - and A-Z0-9 is invalid

t=return
v=tr -d .\ -
[ "tr -dc A-Z0-9<<<$v" =$v ]||$t ((${#v}>9))||$t r=grep -oE "[0-9]+[A-Z]{2}*${v:0:2}"<<<15NO16BEBI18DKFIFOGLNL19MKSI20ATBAEEKZLTLU21CHCRHRLILV22BGBHDEGBGEIEMERS23AEGIIL24ADCZDZESMDPKROSASESKTNVG25AOCVMZPT26IRISTR27BFCMFRGRITMCMGMRSM28BJCICYDOGTHULBMLPLSNAZ29BRPSQAUA30JOKWMU31MT
[ "${r:0:2}" =${#v} ]||$t v=${v:4:99}${v:0:4} d=({A..Z}) for a in {10..35};{ v=${v//${d[a-10]}/$a}
}
$t bc<<<$v%97-1


Explanation

t=return # Alias return to $t v=tr -d .\ - # Remove delimiters from STDIN and save to a variable$v.
# If you want the function to take the IBAN as an argument, insert <<<$1 just before the last  in the line above # Check that the number now contains no characters other than A-Z0-9 (i.e. if all other characters are removed, the string should remain the same because there aren't any of them) # If not, return. By default, return uses the exit status of the previous command, which will be 1 [ "tr -dc A-Z0-9<<<$v" = $v ]||$t
# Check that $v is long enough to have a country code at the start, return if not # I could have put 2 instead of 9, but the character count is the same ((${#v}>9))||$t # give grep the country code data string, which is of the format <length1><country><country><length2><country><country><country>... # for example, this tells you that EE has an IBAN length of 20 # grep searches for a number ([0-9]+), followed by any number including zero of country codes ([A-Z]{2}*), followed by the first two characters of the input IBAN, i.e. its country code (${v:0:2})
# -o makes grep only output the match, not the line containing it; -E enables extended regexes
# The result is saved to the variable $r r=grep -oE "[0-9]+[A-Z]{2}*${v:0:2}"<<<15NO16BEBI18DKFIFOGLNL19MKSI20ATBAEEKZLTLU21CHCRHRLILV22BGBHDEGBGEIEMERS23AEGIIL24ADCZDZESMDPKROSASESKTNVG25AOCVMZPT26IRISTR27BFCMFRGRITMCMGMRSM28BJCICYDOGTHULBMLPLSNAZ29BRPSQAUA30JOKWMU31MT
# Check that the length specified by the country code, the first two chars of $r, is equal to the actual IBAN's length${#v}. return 1 if not.
[ "${r:0:2}" =${#v} ]||$t v=${v:4:99}${v:0:4} # Put the first 4 chars to the back d=({A..Z}) # Make an array of letters A to Z # Loop for each number from 10 to 35 for a in {10..35};{ # in the IBAN, replace letters A to Z by the corresponding number from 10 to 35 v=${v//${d[a-10]}/$a}
}
# $v is now an integer # note: I used bc here because Bash's built-in arithmetic couldn't handle big numbers # find the remainder of dividing$v by 97, subtract 1, and return the value
# if the remainder is 1, then the IBAN is valid and 1-1=0 is returned.
$t bc<<<$v%97-1


Examples

ibanfn<<<'-GB82-WEST-1234 5698.7654.32    ' #returns 0, valid
ibanfn<<<'-GB82-WEST-1234 5698.7654.33    ' #returns 27, invalid, fails remainder test
ibanfn<<<'GB82 WEST 1234 5698 7654 32a'     #returns 1, invalid, contains invalid character a
ibanfn<<<'YY82 WEST 1234 5698 7654 32'      #returns 1, invalid, country code does not exist
ibanfn<<<'GB82 WEST 1234 5698 7654 3210'    #returns 1, invalid, wrong length


Python 3 - (483 - 25%) 362

import re
def c(i):
for m in b" !!#####%%%%%%&&&&&''''''''((())))))))))))****+++,,,,,,,,,------------....///0":
if(i[:2],len(i)+17)==(q[:2],m):
i=i[4:]+i[:4]
for z in range(26):i=re.sub(chr(65+z),str(z+10),i)
return(int(i)%97==1)and re.findall('.{,4}',s)or 0
q=q[2:]
return 0

• WT?? - that's some golfing! Nice work. It looks a bit like line noise from a distance ... j/k besides it's a complement ;-) Commented Aug 1, 2014 at 20:42

Perl (356+2 * 75% = 268.5)

the code is so confusing that even SE's syntax highlighting gets lost in it :)

#!perl -ln
use bignum;s/\W//g;$_=$T=uc;$q=15;/\w/?$H{$_}=$q:$q++for'NO,BEBI,,NLGLDKFOFI,SIMK,KZEEATLTLUBA,HRCHLVCRLI,RSIEGEDEBGBHGBME,GIAEIL,SASESKESCZMDTNPKADDZVGRO,PTMZCVAO,TRISIR,CMMGMRSMITFRGRBFMC,ALAZPLCIGTHUSNCYDOBJLBML,UABRQAPS,KWMUJO,MT'=~/,|../g;/../;$q=$H{$&}==y///c;/..../;$_=$'.$&;s/[A-Z]/ord($&)-55/ge;print+(Invalid,join' ',($T=~/.{1,4}/g))[1==$_%97*$q]  Explanation #!perl -nl  -n means read stdin line by line; -l adds newlines to print use bignum;  required for modulo operation later to return correct value. s/\W//g;  remove everything that's not \w from IBAN. $_=$T=uc;  convert the iban number to upper case, also save it to$T - will be used for pretty-printing later.

$q=15;  set the temporary variable to 15, it will be used to build a hash table with country code to iban length mapping. /\w/ ?$H{$_}=$q : $q++ for 'NO,BEBI,,NLGLDKFOFI,SIMK,KZEEATLTLUBA,HRCHLVCRLI,RSIEGEDEBGBHGBME,GIAEIL,SASESKESCZMDTNPKADDZVGRO,PTMZCVAO,TRISIR,CMMGMRSMITFRGRBFMC,ALAZPLCIGTHUSNCYDOBJLBML,UABRQAPS,KWMUJO,MT' =~ /,|../g;  split the large string into an array of commas or two-letter country codes, iterate over it. if the element starts with a letter, it's a country code - save it in the hash with a value of$q; otherwise, a comma means to increment $q. so, NO will get the value of 15, BE and BI will be 16, and so on. /../;  match first two characters from IBAN (the country code) $q=$H{$&}==y///c;


check if IBAN has correct length for the country code, save result in $q /..../;$_=$'.$&;


move first four characters to the end of IBAN

s/[A-Z]/ord($&)-55/ge;  replace all letters with corresponding numbers, starting with A = 10 print+(Invalid,join' ',($T=~/.{1,4}/g))[1==$_%97*$q]


print either "Invalid" or the pretty-printed IBAN. 1==$_%97*$q will only equal 1 for an IBAN with a correct length and correct remainder.

• nice! In the "perl only" subcontest you win ;-) Even though it's confusing could you try explaining it a bit as in the bash example? Pretty please?! If you explain yours I'll explain mine :-) (though I might steal one of your tricks real the IBAN lengths country first). Cheers. Commented Jul 31, 2014 at 3:24

Perl 6 - 354 chars - (not sure of bonus)

my $r = slurp 'ir'; my @cs =$r.comb(/(\D)+/);my @ns = $r.comb(/(\d)+/); my %h; for @ns Z @cs ->$n, $c {%h{$n}=[$c.comb(/\w**2/)]}; for my @ =lines() { my$o=$_; my$y=$_.substr(0,2);$_=s:g/\s|\-|\.//.uc.comb.rotate(4).join;
my $l=$_.chars ;
$_.=trans(['A'..'Z'] => ['10'..'35']); next if$y !~~ %h{$l}.any; say$_%97==1??"$o ok"!!"Invalid"; }  Reading input from IBANS.txt on STDIN and slurping the rules from the file ir (I left the rules out of the total in case they were boilerplate - the rules file is 191 characters so the total would be 545.  perl6 ibanvalidate.p6 IBANS.txt  IBANS.txt is as follows: GB82 WEST 1234 5698 7654 32 GB82-WEST-1234-5698-7654-32 gb82-west-1234-5698-7654-32 GB82WEST12345698765432 GB82.WEST.1234.5698.7654.32 GR16 0110 1250 0000 0001 2300 695 GB82 WEST 1234 5698 7654 31 NL82 WEST 1234 5698 7654 32 GB29 NWBK 6016 1331 9268 19 Whee perl6 CANADA 000 Banks 007 911  Notes • validates country code and length and outputs as below • Character count from: wc -m ibanvalidate.p6 • Currently not much error checking validation of input is done. • Whitespace is signficant in Perl6 (or more significant than Perl 5) so is counted. Typical output: GB82 WEST 1234 5698 7654 32 ok GB82-WEST-1234-5698-7654-32 ok GB82WEST12345698765432 ok GB82.WEST.1234.5698.7654.32 ok GR16 0110 1250 0000 0001 2300 695 ok Invalid GB29 NWBK 6016 1331 9268 19 ok  This is not typical Perl6 code (especially the comb - lines a friendly and influential Perl developper mentioned this in passing): I'm a beginner. Once the contest is done I will add/change and make any modifications suggested by Perl6-ers. Thanks for playing and being nice :-) • Where's the country code checking? – user16402 Commented Jul 29, 2014 at 13:09 • @professorfish ... I did that in the wrong version of perl ;-) it will add 200+ more characters I'm sure. Commented Jul 29, 2014 at 14:00 Perl 6, 311 bytes, score 233.25 {/:i^(..)<[A..Z0..9]>+$/&&'NobeBIdkfofiglNlmkSiatbaeekzltLucrhrlvliChbhbggedeiemersGbgiilAeadczmdpkrosaskessetnvgDzptaocvMzistrIrfrgritmrmcsmbfcmMgalazcydogthulbplbjcimlSnbrpsqaUajokwMumt'~~/^:i(..)*$($0)/&&.comb==$0.comb(/<:Lu>/)+15&&[.comb].rotate(4)>>.&{:36($_)}.join%97==1&&~.comb(4)||'Invalid'}o{TR/\ .-//}


Try it online!

I'm not sure

• where separator characters are allowed exactly. I simply remove all of them.
• whether the final formatted output should be uppercase.

Explanation

{
/                # Match against regex
:i             #   ignore case
^              #   start of string
(..)           #   country code (stored in $0) <[A..Z0..9]>+ # alphanumeric characters$              #   end of string
/
&&
'NobeBIdk...'    # Country code string with uppercase letter
# for each length increase
~~               # match against
/                # regex
:i             #   ignore case
^              #   start of string
(..)*          #   series of two characters (stored in $0)$($0) # country code / && .comb== # IBAN length equals$0.comb(/<:Lu>/) # number of uppercase characters
+15              # plus 15
&&
[.comb]          # Split IBAN into characters
.rotate(4)       # Rotate by four
>>.&{:36($_)} # Convert each from base-36 .join # Join %97==1 # Modulo 97 check && ~.comb(4) # Add space after every four characters || 'Invalid' # Return 'Invalid' if any of the checks failed } o # Function composition { TR/\ .-// # Remove space, dot, dash }  Python3.x (539 chars - 25% = 404.25) (To be said first: I'm a bit confused what HAS to be counted by your rules so I just counted my whole function) 539 characters - Including import and def IBAN(i): (tabs at the beginning of the line are counted anyway in python so they don't matter) Bonus applied as stated in the rules. def IBAN(i): import re i=re.sub(r'[^\w]','',i).upper() k=len(i) z=i[:2]in re.search(r'%s([A-Z ]*)'%k,'15NO16BE BI18DK FO FI GL NL19MK SI20AT BA ES KZ LT LU21CR HR LV LI CH22BH BG GE DE IE MK RS GB23GI IL AE24MD PK RO SA SK ES SE TN VG DZ25CV MZ26IS TR IR FR GR IT MR MC SM BF CM MG28AL AZ CY DO GT HU LB PL BJ CI ML SN29BR PS QA UA30JO KW MU31MT').group(1)or None if int(''.join([c if c.isdigit()else str(ord(c)-55)for c in i[4:]+i[:4]]))%97!=1 or z is None:return('Invalid') return(' '.join([i[0+j:4+j]for j in range(0,k,4)]))  Code with comments (not completed yet... I'm way too tired to write the remaining part of the comments... I'll do that tomorrow) def IBAN(i): import re # Remove all non word characters (including space) and change to uppercase # as0098-7*46.76A -> AS009874676A i=re.sub(r'[^\w]','',i).upper() # Get the length of it because I need it later multiple times k=len(i) # r'%s([A-Z ]*)'%k = Replace %s with the length of i # re.search( ,'15NO16BE BI18DK FO FI GL NL19MK SI20AT BA ES KZ LT LU21CR HR LV LI CH22BH BG GE DE IE MK RS GB23GI IL AE24MD PK RO SA SK ES SE TN VG DZ25CV MZ26IS TR IR FR GR IT MR MC SM BF CM MG28AL AZ CY DO GT HU LB PL BJ CI ML SN29BR PS QA UA30JO KW MU31MT') # = search (with regexp) in this string (which are the lengths and countrycodes seperated by space) # Return the first found match (scroll to the right to see) .group(1) # i[:2]in or None # = The first two characters in the returned string (in the same order and same case) z=i[:2]in re.search(r'%s([A-Z ]*)'%k,'15NO16BE BI18DK FO FI GL NL19MK SI20AT BA ES KZ LT LU21CR HR LV LI CH22BH BG GE DE IE MK RS GB23GI IL AE24MD PK RO SA SK ES SE TN VG DZ25CV MZ26IS TR IR FR GR IT MR MC SM BF CM MG28AL AZ CY DO GT HU LB PL BJ CI ML SN29BR PS QA UA30JO KW MU31MT').group(1)or None if int(''.join([c if c.isdigit()else str(ord(c)-55)for c in i[4:]+i[:4]]))%97!=1 or z is None:return('Invalid') return(' '.join([i[0+j:4+j]for j in range(0,k,4)]))  Perl (535) Not really golfed yet - no idea if I qualify for the bonus ;-) I will add some @professorfish type explanations. #!perl -ln$r="15NO16BEBI18DKFIFOGLNL19MKSI20ATBAEEKZLTLU21CHCRHRLILV22BGBHDEGBGEIEMERS23AEGIIL24ADCZDZESMDPKROSASESKTNVG25AOCVMZPT26IRISTR27BFCMFRGRITMCMGMRSM28BJCICYDOGTHULBMLPLSNAZ29BRPSQAUA30JOKWMU31MT";
$c{$1} = [unpack("(A2)*", $2)] while($r) =~/(\d{2})(\D{2,})/g;
@h{("A".."Z")}=(10..35);
chomp;$o=$_;$_=uc$_;
s/[ -.]//g;
$l=y///c; ($m)=($_=~m/^.{2}/g); ($n=$_) =~s/^(....)(.*)/$2$1/dg;$n=~s/([A-Z])/$h{$1}/g;
print $m ~~ @{$c{$l}} ? "length valid for$m":"invalid length";
use bignum;
print $n%97==1 ? "$o ok":"$o Invalid";  JavaScript (Node.js), 372 bytes * .75 = 279 s=>(s=s[R="replace"](/[^A-Z0-9]/g,"")).length-14-"NO|BEBI||DKFOFIGLNL|MKSI|BATEEKZLTLU|CRCHRLVLI|GBHBGEDEIEMERS|GILAE|SADCZMDZPKROSASKESETNVG|PTAOCVMZ|ISTRIR|BFRGRITSMRMCMG|ALBJAZCYDOGTHUPLCIMLSN|BRPSQAUA|JOKWMU|MT".match(eval(/.*${s.slice(0,2)}/i))[0].split|.length||BigInt((s.slice(4)+s.slice(0,4))[R](/\w/g,x=>parseInt(x,36)))%97n-1n?"Invalid":s[R](/(.{4})/g,"\$1 ")
`

Try it online!

Test cases borrowed from @marinus 's J answer.