I'm writing some functions in Delphi using Assembly. So I want to put it in a .pas file called Strings.pas. To use in uses of a new Delphi software. What do I need to write, to make it a valid library?
My function is like this:
function Strlen(texto : string) : integer;
begin
asm
mov esi, texto
xor ecx,ecx
cld
#here:
inc ecx
lodsb
cmp al,0
jne #here
dec ecx
mov Result,ecx
end;
end;
That counts the numbers of chars in the string. How can I make it in a lib Strings.pas to call with uses Strings; in my form?
A .pas file is a unit, not a library. A .pas file needs to have unit, interface, and implementation statements, eg:
Strings.pas:
unit Strings;
interface
function Strlen(texto : string) : integer;
implementation
function Strlen(texto : string) : integer;
asm
// your assembly code...
// See Note below...
end;
end.
Then you can add the .pas file to your other projects and use the Strings unit as needed. It will be compiled directly into each executable. You don't need to make a separate library out of it. But if you want to, you can. Create a separate Library (DLL) or Package (BPL) project, add your .pas file to it, and compile it into an executable file that you can then reference in your other projects.
In the case of a DLL library, you will not be able to use the Strings unit directly. You will have to export your function(s) from the library (and string is not a safe data type to pass over a DLL boundary between modules), eg:
Mylib.dpr:
library Mylib;
uses
Strings;
exports
Strings.Strlen;
begin
end.
And then you can have your other projects declare the function(s) using external clause(s) that reference the DLL file, eg:
function Strlen(texto : PChar) : integer; external 'Mylib.dll';
In this case, you can make a wrapper .pas file that declares the functions to import, add that unit to your other projects and use it as needed, eg:
StringsLib.pas:
unit StringsLib;
interface
function Strlen(texto : PChar) : integer;
implementation
function Strlen; external 'Mylib.dll';
end.
In the case of a Package, you can use the Strings units directly. Simply add a reference to the package's .bpi in your other project's Requires list in the Project Manager, and then use the unit as needed. In this case, string is safe to pass around.
Note: in the assembly code you showed, for the function to not cause an access violation, you need to save and restore the ESI register. See the section on Register saving conventions in the Delphi documentation.
The correct asm version may be:
unit MyStrings; // do not overlap Strings.pas unit
interface
function StringLen(const texto : string) : integer;
implementation
function StringLen(const texto : string) : integer;
asm
test eax,eax
jz #done
mov eax,dword ptr [eax-4]
#done:
end;
end.
Note that:
I used MyStrings as unit name, since it is a very bad idea to overlap the official RTL unit names, like Strings.pas;
I wrote (const texto: string) instead of (texto: string), to avoid a reference count change at calling;
Delphi string type already has its length stored as integer just before the character memory buffer;
In Delphi asm calling conventions, the input parameters are set in eax edx ecx registers, and the integer result of a function is the eax register - see this reference article - for Win32 only;
I tested for texto to be nil (eax=0), which stands for a void '' string;
This would work only under Win32 - asm code under Win64 would be diverse;
Built-in length() function would be faster than an asm sub-function, since it is inlined in new versions of Delphi;
Be aware of potential name collisions: there is already a well known StrLen() function, which expects a PChar as input parameter - so I renamed your function as StringLen().
Since you want to learn asm, here are some reference implementation of this function.
A fast PChar oriented version may be :
function StrLen(S: PAnsiChar): integer;
asm
test eax,eax
mov edx,eax
jz #0
xor eax,eax
#s: cmp byte ptr [eax+edx+0],0; je #0
cmp byte ptr [eax+edx+1],0; je #1
cmp byte ptr [eax+edx+2],0; je #2
cmp byte ptr [eax+edx+3],0; je #3
add eax,4
jmp #s
#1: inc eax
#0: ret
#2: add eax,2; ret
#3: add eax,3
end;
A more optimized version:
function StrLen(S: PAnsiChar): integer;
// pure x86 function (if SSE2 not available) - faster than SysUtils' version
asm
test eax,eax
jz ##z
cmp byte ptr [eax+0],0; je ##0
cmp byte ptr [eax+1],0; je ##1
cmp byte ptr [eax+2],0; je ##2
cmp byte ptr [eax+3],0; je ##3
push eax
and eax,-4 { DWORD Align Reads }
##Loop:
add eax,4
mov edx,[eax] { 4 Chars per Loop }
lea ecx,[edx-$01010101]
not edx
and edx,ecx
and edx,$80808080 { Set Byte to $80 at each #0 Position }
jz ##Loop { Loop until any #0 Found }
##SetResult:
pop ecx
bsf edx,edx { Find First #0 Position }
shr edx,3 { Byte Offset of First #0 }
add eax,edx { Address of First #0 }
sub eax,ecx { Returns Length }
##z: ret
##0: xor eax,eax; ret
##1: mov eax,1; ret
##2: mov eax,2; ret
##3: mov eax,3
end;
An SSE2 optimized version:
function StrLen(S: PAnsiChar): integer;
asm // from GPL strlen32.asm by Agner Fog - www.agner.org/optimize
or eax,eax
mov ecx,eax // copy pointer
jz #null // returns 0 if S=nil
push eax // save start address
pxor xmm0,xmm0 // set to zero
and ecx,0FH // lower 4 bits indicate misalignment
and eax,-10H // align pointer by 16
movdqa xmm1,[eax] // read from nearest preceding boundary
pcmpeqb xmm1,xmm0 // compare 16 bytes with zero
pmovmskb edx,xmm1 // get one bit for each byte result
shr edx,cl // shift out false bits
shl edx,cl // shift back again
bsf edx,edx // find first 1-bit
jnz #A200 // found
// Main loop, search 16 bytes at a time
#A100: add eax,10H // increment pointer by 16
movdqa xmm1,[eax] // read 16 bytes aligned
pcmpeqb xmm1,xmm0 // compare 16 bytes with zero
pmovmskb edx,xmm1 // get one bit for each byte result
bsf edx,edx // find first 1-bit
// (moving the bsf out of the loop and using test here would be faster
// for long strings on old processors, but we are assuming that most
// strings are short, and newer processors have higher priority)
jz #A100 // loop if not found
#A200: // Zero-byte found. Compute string length
pop ecx // restore start address
sub eax,ecx // subtract start address
add eax,edx // add byte index
#null:
end;
Or even a SSE4.2 optimized version:
function StrLen(S: PAnsiChar): integer;
asm // warning: may read up to 15 bytes beyond the string itself
or eax,eax
mov edx,eax // copy pointer
jz #null // returns 0 if S=nil
xor eax,eax
pxor xmm0,xmm0
{$ifdef HASAESNI}
pcmpistri xmm0,dqword [edx],EQUAL_EACH // comparison result in ecx
{$else}
db $66,$0F,$3A,$63,$02,EQUAL_EACH
{$endif}
jnz #loop
mov eax,ecx
#null: ret
#loop: add eax,16
{$ifdef HASAESNI}
pcmpistri xmm0,dqword [edx+eax],EQUAL_EACH // comparison result in ecx
{$else}
db $66,$0F,$3A,$63,$04,$10,EQUAL_EACH
{$endif}
jnz #loop
#ok: add eax,ecx
end;
You will find all those functions, including Win64 versions, in our very optimized SynCommons.pas unit, which is shared by almost all our Open Source projects.
The my two solutions to get the length of two types of string,
as for says Peter Cordes are not both useful.
Only the "PAnsiCharLen()" could be an alternative solution,
but not as fast as it is StrLen() (optimized) of Amaud Bouchez,
that it is about 3 times faster than mine.
10/14/2017 (mm/dd/yyy): Added one new function (Clean_Str).
However, for now, I propose three small corrections to both
of them (two suggested by Peter Cordes: 1) use MovZX instead of Mov && And;
2) Use SetZ/SetE instead LAHF/ShL, use XOr EAX,EAX instead XOr AL,AL);
in the future I could define the functions in assembly (now they are defined in Pascal):
unit MyStr;
{ Some strings' function }
interface
Function PAnsiCharLen(S:PAnsiChar):Integer;
{ Get the length of the PAnsiChar^ string. }
Function ShortStrLen(S:ShortString):Integer;
{ Get the length of the ShortString^ string. }
Procedure Clean_Str(Str:ShortString;Max_Len:Integer);
{ This function can be used to clear the unused space of a short string
without modifying his useful content (for example, if you save a
short-string field in a file, at parity of content the file may be
different, because the unused space is not initialized).
Clears a String Str_Ptr ^: String [], which has
Max_Len = SizeOf (String []) - 1 characters, placing # 0
all characters beyond the position of Str_Ptr ^ [Str_Ptr ^ [0]] }
implementation
Function PAnsiCharLen(S:PAnsiChar):Integer;
{ EAX EDX ECX are 1°, 2° AND 3° PARAMETERs.
Can freely modify the EAX, ECX, AND EDX REGISTERs. }
Asm
ClD {Clear string direction flag}
Push EDI {Save EDI's reg. into the STACK}
Mov EDI,S {Load S into EDI's reg.}
XOr EAX,EAX {Set AL's reg. with null terminator}
Mov ECX,-1 {Set ECX's reg. with maximum length of the string}
RepNE ScaSB {Search null and decrease ECX's reg.}
SetE AL {AL is set with FZero}
Add EAX,ECX {EAX= maximum_length_of_the_string - real_length_of_the_string}
Not EAX {EAX= real_length_of_the_string}
Pop EDI {Restore EDI's reg. from the STACK}
End;
Function ShortStrLen(S:ShortString):Integer; Assembler;
{ EAX EDX ECX are 1°, 2° AND 3° PARAMETERs.
Can freely modify the EAX, ECX, AND EDX REGISTERs. }
Asm
MovZX EAX,Byte Ptr [EAX] {Load the length of S^ into EAX's reg. (function's result)}
End;
Procedure Clean_Str(Str:ShortString;Max_Len:Integer); Assembler;
(* EAX EDX ECX are 1°, 2° AND 3° PARAMETERs.
Can freely modify the EAX, ECX, AND EDX REGISTERs. *)
Asm
ClD {Clear string direction flag}
Push EDI {Save EDI's reg. into the STACK}
Mov EDI,Str {Load input string pointer into EDI's reg.}
Mov ECX,Max_Len {Load allocated string length into ECX's reg.}
MovZX EDX,Byte Ptr [EDI] {Load real string length into EDX's reg.}
StC {Process the address of unused space of Str; ...}
AdC EDI,EDX {... skip first byte and useful Str space}
Cmp EDX,ECX {If EDX>ECX ...}
CMovGE EDX,ECX {... set EDX with ECX}
Sub ECX,EDX {ECX contains the size of unused space of Str}
XOr EAX,EAX {Clear accumulator}
Rep StoSB {Fill with 0 the unused space of Str}
Pop EDI {Restore EDI's reg. from the STACK}
End;
end.
Old (incomplete) answer:
"Some new string's functions, not presents in Delphi library, could be these:"
Type Whole=Set Of Char;
Procedure AsmKeepField (PStrIn,PStrOut:Pointer;FieldPos:Byte;
All:Boolean);
{ Given "field" as a sequence of characters that does not contain spaces
or tabs (# 32, # 9), it takes FieldPos (1..N) field
to PStrIn ^ (STRING) and copies it to PStrOut ^ (STRING).
If All = TRUE, it also takes all subsequent fields }
Function AsmUpCComp (PStr1,PStr2:Pointer):Boolean;
{ Compare a string PStr1 ^ (STRING) with a string PStr2 ^ (STRING),
considering the PStr1 alphabetic characters ^ always SHIFT }
Function UpCaseStrComp (Str1,Str2:String;Mode:Boolean):ShortInt;
{ Returns: -1 if Str1 < Str2.
0 is Str1 = Str2.
1 is Str1 > Str2.
MODE = FALSE means "case sensitive comparison" (the letters are
consider them as they are).
MODE = TRUE means that the comparison is done by considering
both strings as if they were all uppercase }
Function KeepLS (Str:String;CntX:Byte):String;
{ RETURN THE PART OF STR THAT INCLUDES THE FIRST CHARACTER
OF STR AND ALL THE FOLLOW UP TO THE POSITION CntX (0 to N-1) INCLUDED }
Function KeepRS (Str:String;CntX,CsMode:Byte):String;
{ RETURN THE PART OF STR STARTING TO POSITION CntX + 1 (0 to N-1)
UP TO END OF STR.
IF CsMode = 0 (INSERT MODE), IF CsMode = 1 (OVERWRITE-MODE):
IN THIS CASE, THE CHARACTER TO CntX + 1 POSITION IS NOT INCLUDED }
Function GetSubStr (Str:String;
Pos,Qnt:Byte;CH:Char):String;
{ RETURN Qnt STR CHARACTERS FROM POSITION Pos (1 to N) OF STR;
IF EFFECTIVE LENGTH IS LESS THAN Qnt, WILL ADDED CHARACTER = CH }
Function Keep_Right_Path_Str_W(PathName:String;FieldWidth:Byte;
FormatWithSpaces:Boolean):String;
{ RESIZE A STRING OF A FILE PATH, FROM PathName;
THE NEW STRING WILL HAVE A MAXIMUM LENGTH OF FieldWidth CHARACTERS.
REPLACE EXCEDENT CHARACTERS WITH 3 POINTS,
INSERTED AFTER DRIVE AND ROOT.
REPLACE SOME DIRECTORY WITH 3 POINTS,
ONLY WHEN IT IS NECESSARY, POSSIBLE FROM SECOND.
FORMAT RETURN WITH SPACE ONLY IF FormatWithSpaces = TRUE }
Function KeepBarStr (Percentage,Qnt:Byte;
Ch1,Ch2,Ch3:Char):String;
{ THIS IS A FUNCTION WICH MAKES A STRING WICH CONTAINS A REPRESENTATION OF STATE
OF ADVANCEMENT OF A PROCESS; IT RETURNS A CHARACTERS' SEQUENCE, CONSTITUTED BY "<Ch1>"
(LENGTH = Percentage / 100 * Qnt), WITH AN APPROXIMATION OF THE LAST CHARACTER TO
"<Ch2>" (IF "Percentage / 100 * Qnt" HAS HIS FRACTIONAL'S PART GREATER THAN 0.5),
FOLLOWED BY AN OTHER CHARACTERS' SEQUENCE, CONSTITUTED BY "<Ch3>" (LENGTH = (100 -
Percentage) / 100 * Qnt). }
Function Str2ChWhole (Str:String;Var StrIndex:Byte;
Var ChSet:Whole;
Mode:Boolean):Boolean;
{ CONVERT A PART OF Str, POINTED BY StrIndex, IN A ChSet CHARACTER SET;
IF Mode = TRUE, "StrIn" SHOULD CONTAIN ASCII CODES
OF CORRESPONDING CHARACTERS EXPRESSED IN DECIMAL SIZE;
OTHERWISE IT SHOULD CONTAIN CORRESPONDING CHARACTER SYMBOLS }
Function ChWhole2Str (ChSet:Whole;Mode:Boolean):String;
{ CONVERT A SET OF CHARACTERS IN A CORRESPONDING STRING;
IF Mode = TRUE ELEMENTS OF ChSet WILL BE CONVERTED IN ASCII CODES
EXPRESSED IN DECIMAL SIZE; OTHERWISE THE CORRESPONDING SYMBOLS
WILL BE RETURNED }
Function ConverteFSize (FSize:LongInt;
Var SizeStr:TSizeStr):Integer;
{ MAKES THE CONVERSION OF THE DIMENSION OF A FILE IN A TEXT,
LARGE TO MAXIMUM 5 CHARACTERS, AND RETURN THE COLOR OF THIS STRING }
Function UpCasePos (SubStr,Str:String):Byte;
{ Like the Pos () system function, but not "case sensitive" }
Closed. This question needs to be more focused. It is not currently accepting answers.
Closed 8 years ago.
Locked. This question and its answers are locked because the question is off-topic but has historical significance. It is not currently accepting new answers or interactions.
The challenge
The shortest code by character count, that will input a string using only alphabetical characters (upper and lower case), numbers, commas, periods and question mark, and returns a representation of the string in Morse code.
The Morse code output should consist of a dash (-, ASCII 0x2D) for a long beep (AKA 'dah') and a dot (., ASCII 0x2E) for short beep (AKA 'dit').
Each letter should be separated by a space (' ', ASCII 0x20), and each word should be separated by a forward slash (/, ASCII 0x2F).
Morse code table:
alt text http://liranuna.com/junk/morse.gif
Test cases:
Input:
Hello world
Output:
.... . .-.. .-.. --- / .-- --- .-. .-.. -..
Input:
Hello, Stackoverflow.
Output:
.... . .-.. .-.. --- --..-- / ... - .- -.-. -.- --- ...- . .-. ..-. .-.. --- .-- .-.-.-
Code count includes input/output (that is, the full program).
C (131 characters)
Yes, 131!
main(c){for(;c=c?c:(c=toupper(getch())-32)?
"•ƒŒKa`^ZRBCEIQiw#S#nx(37+$6-2&#/4)'18=,*%.:0;?5"
[c-12]-34:-3;c/=2)putch(c/2?46-c%2:0);}
I eeked out a few more characters by combining the logic from the while and for loops into a single for loop, and by moving the declaration of the c variable into the main definition as an input parameter. This latter technique I borrowed from strager's answer to another challenge.
For those trying to verify the program with GCC or with ASCII-only editors, you may need the following, slightly longer version:
main(c){for(;c=c?c:(c=toupper(getchar())-32)?c<0?1:
"\x95#\x8CKa`^ZRBCEIQiw#S#nx(37+$6-2&#/4)'18=,*%.:0;?5"
[c-12]-34:-3;c/=2)putchar(c/2?46-c%2:32);}
This version is 17 characters longer (weighing in at a comparatively huge 148), due to the following changes:
+4: getchar() and putchar() instead of the non-portable getch() and putch()
+6: escape codes for two of the characters instead of non-ASCII characters
+1: 32 instead of 0 for space character
+6: added "c<0?1:" to suppress garbage from characters less than ASCII 32 (namely, from '\n'). You'll still get garbage from any of !"#$%&'()*+[\]^_`{|}~, or anything above ASCII 126.
This should make the code completely portable. Compile with:
gcc -std=c89 -funsigned-char morse.c
The -std=c89 is optional. The -funsigned-char is necessary, though, or you will get garbage for comma and full stop.
135 characters
c;main(){while(c=toupper(getch()))for(c=c-32?
"•ƒŒKa`^ZRBCEIQiw#S#nx(37+$6-2&#/4)'18=,*%.:0;?5"
[c-44]-34:-3;c;c/=2)putch(c/2?46-c%2:0);}
In my opinion, this latest version is much more visually appealing, too. And no, it's not portable, and it's no longer protected against out-of-bounds input. It also has a pretty bad UI, taking character-by-character input and converting it to Morse Code and having no exit condition (you have to hit Ctrl+Break). But portable, robust code with a nice UI wasn't a requirement.
A brief-as-possible explanation of the code follows:
main(c){
while(c = toupper(getch())) /* well, *sort of* an exit condition */
for(c =
c - 32 ? // effectively: "if not space character"
"•ƒŒKa`^ZRBCEIQiw#S#nx(37+$6-2&#/4)'18=,*%.:0;?5"[c - 44] - 34
/* This array contains a binary representation of the Morse Code
* for all characters between comma (ASCII 44) and capital Z.
* The values are offset by 34 to make them all representable
* without escape codes (as long as chars > 127 are allowed).
* See explanation after code for encoding format.
*/
: -3; /* if input char is space, c = -3
* this is chosen because -3 % 2 = -1 (and 46 - -1 = 47)
* and -3 / 2 / 2 = 0 (with integer truncation)
*/
c; /* continue loop while c != 0 */
c /= 2) /* shift down to the next bit */
putch(c / 2 ? /* this will be 0 if we're down to our guard bit */
46 - c % 2 /* We'll end up with 45 (-), 46 (.), or 47 (/).
* It's very convenient that the three characters
* we need for this exercise are all consecutive.
*/
: 0 /* we're at the guard bit, output blank space */
);
}
Each character in the long string in the code contains the encoded Morse Code for one text character. Each bit of the encoded character represents either a dash or a dot. A one represents a dash, and a zero represents a dot. The least significant bit represents the first dash or dot in the Morse Code. A final "guard" bit determines the length of the code. That is, the highest one bit in each encoded character represents end-of-code and is not printed. Without this guard bit, characters with trailing dots couldn't be printed correctly.
For instance, the letter 'L' is ".-.." in Morse Code. To represent this in binary, we need a 0, a 1, and two more 0s, starting with the least significant bit: 0010. Tack one more 1 on for a guard bit, and we have our encoded Morse Code: 10010, or decimal 18. Add the +34 offset to get 52, which is the ASCII value of the character '4'. So the encoded character array has a '4' as the 33rd character (index 32).
This technique is similar to that used to encode characters in ACoolie's, strager's(2), Miles's, pingw33n's, Alec's, and Andrea's solutions, but is slightly simpler, requiring only one operation per bit (shifting/dividing), rather than two (shifting/dividing and decrementing).
EDIT:
Reading through the rest of the implementations, I see that Alec and Anon came up with this encoding scheme—using the guard bit—before I did. Anon's solution is particularly interesting, using Python's bin function and stripping off the "0b" prefix and the guard bit with [3:], rather than looping, anding, and shifting, as Alec and I did.
As a bonus, this version also handles hyphen (-....-), slash (-..-.), colon (---...), semicolon (-.-.-.), equals (-...-), and at sign (.--.-.). As long as 8-bit characters are allowed, these characters require no extra code bytes to support. No more characters can be supported with this version without adding length to the code (unless there's Morse Codes for greater/less than signs).
Because I find the old implementations still interesting, and the text has some caveats applicable to this version, I've left the previous content of this post below.
Okay, presumably, the user interface can suck, right? So, borrowing from strager, I've replaced gets(), which provides buffered, echoed line input, with getch(), which provides unbuffered, unechoed character input. This means that every character you type gets translated immediately into Morse Code on the screen. Maybe that's cool. It no longer works with either stdin or a command-line argument, but it's pretty damn small.
I've kept the old code below, though, for reference. Here's the new.
New code, with bounds checking, 171 characters:
W(i){i?W(--i/2),putch(46-i%2):0;}c;main(){while(c=toupper(getch())-13)
c=c-19?c>77|c<31?0:W("œ*~*hXPLJIYaeg*****u*.AC5+;79-#6=0/8?F31,2:4BDE"
[c-31]-42):putch(47),putch(0);}
Enter breaks the loop and exits the program.
New code, without bounds checking, 159 characters:
W(i){i?W(--i/2),putch(46-i%2):0;}c;main(){while(c=toupper(getch())-13)
c=c-19?W("œ*~*hXPLJIYaeg*****u*.AC5+;79-#6=0/8?F31,2:4BDE"[c-31]-42):
putch(47),putch(0);}
Below follows the old 196/177 code, with some explanation:
W(i){i?W(--i/2),putch(46-i%2):0;}main(){char*p,c,s[99];gets(s);
for(p=s;*p;)c=*p++,c=toupper(c),c=c-32?c>90|c<44?0:W(
"œ*~*hXPLJIYaeg*****u*.AC5+;79-#6=0/8?F31,2:4BDE"[c-44]-42):
putch(47),putch(0);}
This is based on Andrea's Python answer, using the same technique for generating the morse code as in that answer. But instead of storing the encodable characters one after another and finding their indexes, I stored the indexes one after another and look them up by character (similarly to my earlier answer). This prevents the long gaps near the end that caused problems for earlier implementors.
As before, I've used a character that's greater than 127. Converting it to ASCII-only adds 3 characters. The first character of the long string must be replaced with \x9C. The offset is necessary this time, otherwise a large number of characters are under 32, and must be represented with escape codes.
Also as before, processing a command-line argument instead of stdin adds 2 characters, and using a real space character between codes adds 1 character.
On the other hand, some of the other routines here don't deal with input outside the accepted range of [ ,.0-9\?A-Za-z]. If such handling were removed from this routine, then 19 characters could be removed, bringing the total down as low as 177 characters. But if this is done, and invalid input is fed to this program, it may crash and burn.
The code in this case could be:
W(i){i?W(--i/2),putch(46-i%2):0;}main(){char*p,s[99];gets(s);
for(p=s;*p;p++)*p=*p-32?W(
"œ*~*hXPLJIYaeg*****u*.AC5+;79-#6=0/8?F31,2:4BDE"
[toupper(*p)-44]-42):putch(47),putch(0);}
Using a Morse Code Font?
Console.Write(params[0]);
Perl, 170 characters (with a little help from accomplished golfer mauke). Wrapped for clarity; all newlines are removable.
$_=uc<>;y,. ,|/,;s/./$& /g;#m{A..Z,0..9,qw(| , ?)}=
".-NINNN..]IN-NII..AMN-AI---.M-ANMAA.I.-].AIAA-NANMMIOMAOUMSMSAH.B.MSOIONARZMIZ"
=~/../g;1while s![]\w|,?]!$m{$&}!;print
Explanation:
Extract the morse dictionary. Each symbol is defined in terms of two chars, which can be either literal dots or dashes, or a reference to the value of another defined char. E and T contain dummy chars to avoid desyncing the decoder; we'll remove them later.
Read and format the input. "Hello world" becomes "H E L L O / W O R L D"
The next step depends on the input and output dictionaries being distinct, so turn dots in the input to an unused char (vertical bar, |)
Replace any char in the input that occurs in the morse dictionary with its value in the dictionary, until no replacements occur.
Remove the dummy char mentioned in step 1.
Print the output.
In the final version, the dictionary is optimized for runtime efficiency:
All one-symbol characters (E and T) and two-symbol characters (A, I, M, and N) are defined directly and decode in one pass.
All three-symbol characters are defined in terms of a two-symbol character and a literal symbol, decoding in two passes.
All four-symbol characters are defined in terms of two two-symbol characters, decoding in two passes with three replacements.
The five- and six-symbol characters (numbers and punctuation) decode in three passes, with four or five replacements respectively.
Since the golfed code only replaces one character per loop (to save one character of code!) the number of loops is limited to five times the length of the input (three times the length of the input if only alphabetics are used). But by adding a g to the s/// operation, the number of loops is limited to three (two if only alphabetics are used).
Example transformation:
Hello 123
H E L L O / 1 2 3
II .] AI AI M- / AO UM SM
.... . .-.. .-.. --- / .-M- .A-- I.--
.... . .-.. .-.. --- / .---- ..--- ...--
Python list comprehension, 159-character one-liner
for c in raw_input().upper():print c<","and"/"or bin(ord("•ƒwTaQIECBRZ^`šŒ#S#n|':<.$402&9/6)(18?,*%+3-;=>"[ord(c)-44])-34)[3:].translate(" "*47+"/.-"+" "*206),
Uses the similar data packing to P Daddy's C implementation, but does not store the bits in reverse order and uses bin() to extract the data rather than arithmetic. Note also that spaces are detected using inequality; it considers every character "less than comma" to be a space.
Python for loop, 205 chars including newlines
for a in raw_input().upper():
q='_ETIANMSURWDKGOHVF_L_PJBXCYZQ__54_3___2__+____16=/_____7___8_90'.find(a);s=''
while q>0:s='-.'[q%2]+s;q=~-q/2
print['/','--..--','..--..','.-.-.-',''][' ,?.'.find(a)]+s,
I was dorking around with a compact coding for the symbols, but I don't see if getting any better than the implicit trees already in use, so I present the coding here in case some one else can use it.
Consider the string:
--..--..-.-.-..--...----.....-----.--/
which contains all the needed sequences as substrings. We could code the symbols by offset and length like this:
ET RRRIIGGGJJJJ
--..--..-.-.-..--...----.....-----.--/
CCCC DD WWW 00000
,,,,,, AALLLL BBBB 11111
--..--..-.-.-..--...----.....-----.--/
?????? KKK MMSSS 22222
FFFF PPPP 33333
--..--..-.-.-..--...----.....-----.--/
UUU XXXX 44444
NN PPPP OOO 55555
--..--..-.-.-..--...----.....-----.--/
ZZZZ 66666
77777 YYYY
--..--..-.-.-..--...----.....-----.--/
...... 88888 HHHH
99999 VVVV QQQQ
--..--..-.-.-..--...----.....-----.--/
with the space (i.e. word boundary) starting and ending on the final character (the '/'). Feel free to use it, if you see a good way.
Most of the shorter symbols have several possible codings, of course.
P Daddy found a shorter version of this trick (and I can now see at least some of the redundancy here) and did a nice c implementation. Alec did a python implementation with the first (buggy and incomplete) version. Hobbs did a pretty compact perl version that I don't understand at all.
J, 124 130 134 characters
'.- /'{~;2,~&.>(]`(<&3:)#.(a:=])"0)}.&,&#:&.></.40-~a.i.')}ggWOKIHX`dfggggggg-#B4*:68,?5</.7>E20+193ACD'{~0>.45-~a.i.toupper
J beats C! Awesome!
Usage:
'.- /'{~;2,~&.>(]`(<&3:)#.(a:=])"0)}.&,&#:&.></.40-~a.i.')}ggWOKIHX`dfggggggg-#B4*:68,?5</.7>E20+193ACD'{~0>.45-~a.i.toupper 'Hello World'
.... . .-.. .-.. --- / .-- --- .-. .-.. -..
'.- /'{~;2,~&.>(]`(<&3:)#.(a:=])"0)}.&,&#:&.></.40-~a.i.')}ggWOKIHX`dfggggggg-#B4*:68,?5</.7>E20+193ACD'{~0>.45-~a.i.toupper 'Hello, Stackoverflow.'
.... . .-.. .-.. --- .-.-.- / ... - .- -.-. -.- --- ...- . .-. ..-. .-.. --- .-- --..--
Python 3 One Liner: 172 characters
print(' '.join('/'if c==' 'else''.join('.'if x=='0'else'-'for x in bin(ord("ijÁĕÁÿïçãáàðøüþÁÁÁÁÁČÁÅ×ÚÌÂÒÎÐÄ×ÍÔÇÆÏÖÝÊÈÃÉÑËÙÛÜ"[ord(c)-44])-192)[3:])for c in input().upper()))
(Encoding the tranlation table into unicode code points. Works fine, and they display here fine in my test on my Windows Vista machine.)
Edited to pare down to 184 characters by removing some unnecessary spaces and brackets (making list comps gen exps).
Edit again: More spaces removed that I didn't even know was possible before seeing other answers here - so down to 176.
Edit again down to 172 (woo woo!) by using ' '.join instead of ''.join and doing the spaces separately. (duh!)
C# 266 chars
The 131 char C solution translated to C# yields 266 characters:
foreach(var i in Encoding.ASCII.GetBytes(args[0].ToUpper())){var c=(int)i;for(c=(c-32!=0)?Encoding.ASCII.GetBytes("•ƒŒKa`^ZRBCEIQiw#S#nx(37+$6-2&#/4)'18=,*%.:0;?5")[c-44]-34:-3;c!=0;c/=2)Console.Write(Encoding.ASCII.GetChars(new byte[]{(byte)((c/2!=0)?46-c%2:0)}));}
which is more readable as:
foreach (var i in Encoding.ASCII.GetBytes(args[0].ToUpper()))
{
var c = (int)i;
for (c = ((c - 32) != 0) ? Encoding.ASCII.GetBytes("•ƒŒKa`^ZRBCEIQiw#S#nx(37+$6-2&#/4)'18=,*%.:0;?5")[c - 44] - 34 : -3
; c != 0
; c /= 2)
Console.Write(Encoding.ASCII.GetChars(new byte[] { (byte)((c / 2 != 0) ? 46 - c % 2 : 0) }));
}
Golfscript - 106 chars - NO FUNNY CHARS :)
newline at the end of the input is not supported, so use something like this
echo -n Hello, Stackoverflow| ../golfscript.rb morse.gs
' '/{{.32|"!etianmsurwdkgohvf!l!pjbxcyzq"?)"UsL?/'#! 08<>"#".,?0123456789"?=or
2base(;>{'.-'\=}%' '}%}%'/'*
Letters are a special case and converted to lowercase and ordered in their binary positions.
Everything else is done by a translation table
Python
Incomplete solution, but maybe somebody can make a full solution out of it. Doesn't handle digits or punctuation, but weighs in at only 154 chars.
def e(l):
i='_etianmsurwdkgohvf_l_pjbxcyzq'.find(l.lower());v=''
while i>0:v='-.'[i%2]+v;i=(i-1)/2;return v or '/'
def enc(s):return ' '.join(map(e,s))
C (248 characters)
Another tree-based solution.
#define O putchar
char z[99],*t=
" ETINAMSDRGUKWOHBL~FCPJVX~YZQ~~54~3~~~2~~+~~~~16=/~~.~~7,~~8~90";c,p,i=0;
main(){gets(z);while(c=z[i++]){c-46?c-44?c:O(45):O(c);c=c>96?c-32:c;p=-1;
while(t[++p]!=c);for(;p;p/=2){O(45+p--%2);}c-32?O(32):(O(47),O(c));}}
Could be errors in source tree because wikipedia seems to have it wrong or maybe I misunderstood something.
F#, 256 chars
let rec D i=if i=16 then" "else
let x=int"U*:+F8c]uWjGbJ0-0Dnmd0BiC5?\4o`h7f>9[1E=pr_".[i]-32
if x>43 then"-"+D(x-43)else"."+D x
let M(s:string)=s.ToUpper()|>Seq.fold(fun s c->s+match c with
|' '->"/ "|','->"--..-- "|'.'->".-.-.- "|_->D(int c-48))""
For example
M("Hello, Stack.") |> printfn "%s"
yields
.... . .-.. .-.. --- --..-- / ... - .- -.-. -.- .-.-.-
I think my technique may be unique so far. The idea is:
there is an ascii range of chars that covers most of what we want (0..Z)
there are only 43 chars in this range
thus we can encode one bit (dash or dot) plus a 'next character' in a range of 86 chars
the range ascii(32-117) is all 'printable' and can serve as this 86-char range
so the string literal encodes a table along those lines
There's a little more to it, but that's the gist. Comma, period, and space are not in the range 0..Z so they're handled specially by the 'match'. Some 'unused' characters in the range 0..Z (like ';') are used in the table as suffixes of other morse translations that aren't themselves morse 'letters'.
Here's my contribution as a console application in VB.Net
Module MorseCodeConverter
Dim M() As String = {".-", "-...", "-.-.", "-..", ".", "..-.", "--.", "....", "..", ".---", "-.-", ".-..", "--", "-.", "---", ".--.", "--.-", ".-.", "...", "-", "..-", "...-", ".--", "-..-", "-.--", "--..", "-----", ".----", "..---", "...--", "....-", ".....", "-....", "--...", "---..", "----."}
Sub Main()
Dim I, O
Dim a, b
While True
I = Console.ReadLine()
O = ""
For Each a In I
b = AscW(UCase(a))
If b > 64 And b < 91 Then
O &= M(b - 65) & " "
ElseIf b > 47 And b < 58 Then
O &= M(b - 22) & " "
ElseIf b = 46 Then
O &= ".-.-.- "
ElseIf b = 44 Then
O &= "--..-- "
ElseIf b = 63 Then
O &= "..--.. "
Else
O &= "/"
End If
Next
Console.WriteLine(O)
End While
End Sub
End Module
I left he white space in to make it readable. Totals 1100 characters. It will read the input from the command line, one line at a time, and send the corresponding output back to the output stream. The compressed version is below, with only 632 characters.
Module Q
Dim M() As String={".-","-...","-.-.","-..",".","..-.","--.","....","..",".---","-.-",".-..","--","-.","---",".--.","--.-",".-.","...","-","..-","...-",".--","-..-","-.--","--..","-----",".----","..---","...--","....-",".....","-....","--...","---..","----."}
Sub Main()
Dim I,O,a,b:While 1:I=Console.ReadLine():O="":For Each a In I:b=AscW(UCase(a)):If b>64 And b<91 Then:O &=M(b-65)&" ":ElseIf b>47 And b<58 Then:O &=M(b-22)&" ":ElseIf b=46 Then:O &=".-.-.- ":ElseIf b=44 Then:O &="--..-- ":ElseIf b=63 Then:O &= "..--.. ":Else:O &="/":End IF:Next:Console.WriteLine(O):End While
End Sub
End Module
C (233 characters)
W(n,p){while(n--)putch(".-.-.--.--..--..-.....-----..../"[p++]);}main(){
char*p,c,s[99];gets(s);for(p=s;*p;){c=*p++;c=toupper(c);c=c>90?35:c-32?
"È#À#¶µ´³²±°¹¸·#####Ê##i Že‘J•aEAv„…`q!j“d‰ƒˆ"[c-44]:63;c-35?
W(c>>5,c&31):0;putch(0);}}
This takes input from stdin. Taking input from the command line adds 2 characters. Instead of:
...main(){char*p,c,s[99];gets(s);for(p=s;...
you get:
...main(int i,char**s){char*p,c;for(p=s[1];...
I'm using Windows-1252 code page for characters above 127, and I'm not sure how they'll turn up in other people's browsers. I notice that, in my browser at least (Google Chrome), two of the characters (between "#" and "i") aren't showing up. If you copy out of the browser and paste into a text editor, though, they do show up, albeit as little boxes.
It can be converted to ASCII-only, but this adds 24 characters, increasing the character count to 257. To do this, I first offset each character in the string by -64, minimizing the number of characters that are greater than 127. Then I substitute \xXX character escapes where necessary. It changes this:
...c>90?35:c-32?"È#À#¶µ´³²±°¹¸·#####Ê##i Že‘J•aEAv„…`q!j“d‰ƒˆ"[c-44]:63;
c-35?W(...
to this:
...c>90?99:c-32?"\x88#\x80#vutsrqpyxw#####\x8A#\0PA)\xE0N%Q\nU!O\5\1\66DE 1
\xE1*S$ICH"[c-44]+64:63;c-99?W(...
Here's a more nicely formatted and commented version of the code:
/* writes `n` characters from internal string to stdout, starting with
* index `p` */
W(n,p){
while(n--)
/* warning for using putch without declaring it */
putch(".-.-.--.--..--..-.....-----..../"[p++]);
/* dmckee noticed (http://tinyurl.com/n4eart) the overlap of the
* various morse codes and created a 37-character-length string that
* contained the morse code for every required character (except for
* space). You just have to know the start index and length of each
* one. With the same idea, I came up with this 32-character-length
* string. This not only saves 5 characters here, but means that I
* can encode the start indexes with only 5 bits below.
*
* The start and length of each character are as follows:
*
* A: 0,2 K: 1,3 U: 10,3 4: 18,5
* B: 16,4 L: 15,4 V: 19,4 5: 17,5
* C: 1,4 M: 5,2 W: 4,3 6: 16,5
* D: 9,3 N: 1,2 X: 9,4 7: 25,5
* E: 0,1 O: 22,3 Y: 3,4 8: 24,5
* F: 14,4 P: 4,4 Z: 8,4 9: 23,5
* G: 5,3 Q: 5,4 0: 22,5 .: 0,6
* H: 17,4 R: 0,3 1: 21,5 ,: 8,6
* I: 20,2 S: 17,3 2: 20,5 ?: 10,6
* J: 21,4 T: 1,1 3: 19,5
*/
}
main(){ /* yuck, but it compiles and runs */
char *p, c, s[99];
/* p is a pointer within the input string */
/* c saves from having to do `*p` all the time */
/* s is the buffer for the input string */
gets(s); /* warning for use without declaring */
for(p=s; *p;){ /* begin with start of input, go till null character */
c = *p++; /* grab *p into c, increment p.
* incrementing p here instead of in the for loop saves
* one character */
c=toupper(c); /* warning for use without declaring */
c = c > 90 ? 35 : c - 32 ?
"È#À#¶µ´³²±°¹¸·#####Ê##i Že‘J•aEAv„…`q!j“d‰ƒˆ"[c - 44] : 63;
/**** OR, for the ASCII version ****/
c = c > 90 ? 99 : c - 32 ?
"\x88#\x80#vutsrqpyxw#####\x8A#\0PA)\xE0N%Q\nU!O\5\1\66DE 1\xE1"
"*S$ICH"[c - 44] + 64 : 63;
/* Here's where it gets hairy.
*
* What I've done is encode the (start,length) values listed in the
* comment in the W function into one byte per character. The start
* index is encoded in the low 5 bits, and the length is encoded in
* the high 3 bits, so encoded_char = (char)(length << 5 | position).
* For the longer, ASCII-only version, 64 is subtracted from the
* encoded byte to reduce the necessity of costly \xXX representations.
*
* The character array includes encoded bytes covering the entire range
* of characters covered by the challenge, except for the space
* character, which is checked for separately. The covered range
* starts with comma, and ends with capital Z (the call to `toupper`
* above handles lowercase letters). Any characters not supported are
* represented by the "#" character, which is otherwise unused and is
* explicitly checked for later. Additionally, an explicit check is
* done here for any character above 'Z', which is changed to the
* equivalent of a "#" character.
*
* The encoded byte is retrieved from this array using the value of
* the current character minus 44 (since the first supported character
* is ASCII 44 and index 0 in the array). Finally, for the ASCII-only
* version, the offset of 64 is added back in.
*/
c - 35 ? W(c >> 5, c & 31) : 0;
/**** OR, for the ASCII version ****/
c - 99 ? W(c >> 5, c & 31) : 0;
/* Here's that explicit check for the "#" character, which, as
* mentioned above, is for characters which will be ignored, because
* they aren't supported. If c is 35 (or 99 for the ASCII version),
* then the expression before the ? evaluates to 0, or false, so the
* expression after the : is evaluated. Otherwise, the expression
* before the ? is non-zero, thus true, so the expression before
* the : is evaluated.
*
* This is equivalent to:
*
* if(c != 35) // or 99, for the ASCII version
* W(c >> 5, c & 31);
*
* but is shorter by 2 characters.
*/
putch(0);
/* This will output to the screen a blank space. Technically, it's not
* the same as a space character, but it looks like one, so I think I
* can get away with it. If a real space character is desired, this
* must be changed to `putch(32);`, which adds one character to the
* overall length.
} /* end for loop, continue with the rest of the input string */
} /* end main */
This beats everything here except for a couple of the Python implementations. I keep thinking that it can't get any shorter, but then I find some way to shave off a few more characters. If anybody can find any more room for improvement, let me know.
EDIT:
I noticed that, although this routine rejects any invalid characters above ASCII 44 (outputting just a blank space for each one), it doesn't check for invalid characters below this value. To check for these adds 5 characters to the overall length, changing this:
...c>90?35:c-32?"...
to this:
...c-32?c>90|c<44?35:"...
REBOL (118 characters)
A roughly 10 year-old implementation
foreach c ask""[l: index? find" etinamsdrgukwohblzfcpövxäqüyj"c while[l >= 2][prin pick"-."odd? l l: l / 2]prin" "]
Quoted from: http://www.rebol.com/oneliners.html
(no digits though and words are just separated by double spaces :/ ...)
Python (210 characters)
This is a complete solution based on Alec's one
def e(l):
i=(' etianmsurwdkgohvf_l_pjbxcyzq__54_3___2%7s16%7s7___8_90%12s?%8s.%29s,'%tuple('_'*5)).find(l.lower());v=''
while i>0:v='-.'[i%2]+v;i=(i-1)/2
return v or '/'
def enc(s):return ' '.join(map(e,s))
C, 338 chars
338 with indentation and all removable linebreaks removed:
#define O putchar
#define W while
char*l="x#####ppmmmmm##FBdYcbcbSd[Kcd`\31(\b1g_<qCN:_'|\25D$W[QH0";
int c,b,o;
main(){
W(1){
W(c<32)
c=getchar()&127;
W(c>96)
c^=32;
c-=32;
o=l[c/2]-64;
b=203+(c&1?o>>3:0);
o=c&1?o&7:o>>3;
W(o>6)
O(47),o=0;
c/=2;
W(c--)
b+=(l[c]-64&7)+(l[c]-64>>3);
b=(((l[b/7]<<7)+l[b/7+1])<<(b%7))>>14-o;
W(o--)
O(b&(1<<o)?46:45);
O(32);
}
}
This isn't based on the tree approach other people have been taking. Instead, l first encodes the lengths of all bytes between 32 and 95 inclusive, two bytes to a character. As an example, D is -.. for a length of 3 and E is . for a length of 1. This is encoded as 011 and 001, giving 011001. To make more characters encodable and avoid escapes, 64 is then added to the total, giving 1011001 - 89, ASCII Y. Non-morse characters are assigned a length of 0. The second half of l (starting with \031) are the bits of the morse code itself, with a dot being 1 and a dash 0. To avoid going into high ASCII, this data is encoded 7 bits/byte.
The code first sanitises c, then works out the morse length of c (in o), then adds up the lengths of all the previous characters to produce b, the bit index into the data.
Finally, it loops through the bits, printing dots and dashes.
The length '7' is used as a special flag for printing a / when encountering a space.
There are probably some small gains to be had from removing brackets, but I'm way off from some of the better results and I'm hungry, so...
C# Using Linq (133 chars)
static void Main()
{
Console.WriteLine(String.Join(" ", (from c in Console.ReadLine().ToUpper().ToCharArray()
select m[c]).ToArray()));
}
OK, so I cheated. You also need to define a dictionary as follows (didn't bother counting the chars, since this blows me out of the game):
static Dictionary<char, string> m = new Dictionary<char, string>() {
{'A', ".-"},
{'B', "-.."},
{'C', "-.-."},
{'D', "-.."},
{'E', "."},
{'F', "..-."},
{'G', "--."},
{'H', "...."},
{'I', ".."},
{'J', ".---"},
{'K', "-.-"},
{'L', ".-.."},
{'M', "--"},
{'N', "-."},
{'O', "---"},
{'P', ".--."},
{'Q', "--.-"},
{'R', ".-."},
{'S', "..."},
{'T', "-"},
{'U', "..-"},
{'V', "...-"},
{'W', ".--"},
{'X', "-..-"},
{'Y', "-.--"},
{'Z', "--.."},
{'0', "-----"},
{'1', ".----"},
{'2', "..---"},
{'3', "...--"},
{'4', "....-"},
{'5', "....."},
{'6', "-...."},
{'7', "--..."},
{'8', "---.."},
{'9', "----."},
{' ', "/"},
{'.', ".-.-.-"},
{',', "--..--"},
{'?', "..--.."},
};
Still, can someone provide a more concise C# implementation which is also as easy to understand and maintain as this?
Perl, 206 characters, using dmckee's idea
This is longer than the first one I submitted, but I still think it's interesting. And/or awful. I'm not sure yet. This makes use of dmckee's coding idea, plus a couple other good ideas that I saw around. Initially I thought that the "length/offset in a fixed string" thing couldn't come out to less data than the scheme in my other solution, which uses a fixed two bytes per char (and all printable bytes, at that). I did in fact manage to get the data down to considerably less (one byte per char, plus four bytes to store the 26-bit pattern we're indexing into) but the code to get it out again is longer, despite my best efforts to golf it. (Less complex, IMO, but longer anyway).
Anyway, 206 characters; newlines are removable except the first.
#!perl -lp
($a,#b)=unpack"b32C*",
"\264\202\317\0\31SF1\2I.T\33N/G\27\308XE0=\x002V7HMRfermlkjihgx\207\205";
$a=~y/01/-./;#m{A..Z,0..9,qw(. , ?)}=map{substr$a,$_%23,1+$_/23}#b;
$_=join' ',map$m{uc$_}||"/",/./g
Explanation:
There are two parts to the data. The first four bytes ("\264\202\317\0") represent 32 bits of morse code ("--.-..-.-.-----.....--..--------") although only the first 26 bits are used. This is the "reference string".
The remainder of the data string stores the starting position and length of substrings of the reference string that represent each character -- one byte per character, in the order (A, B, ... Z, 0, 1, ... 9, ".", ",", "?"). The values are coded as 23 * (length - 1) + pos, and the decoder reverses that. The last starting pos is of course 22.
So the unpack does half the work of extracting the data and the third line (as viewed here) does the rest, now we have a hash with $m{'a'} = '.-' et cetera, so all there is left is to match characters of the input, look them up in the hash, and format the output, which the last line does... with some help from the shebang, which tells perl to remove the newline on input, put lines of input in $_, and when the code completes running, write $_ back to output with newlines added again.
Python 2; 171 characters
Basically the same as Andrea's solution, but as a complete program, and using stupid tricks to make it shorter.
for c in raw_input().lower():print"".join(".-"[int(d)]for d in bin(
(' etianmsurwdkgohvf_l_pjbxcyzq__54_3___2%7s16%7s7___8_90%12s?%8s.%29s,'
%(('',)*5)).find(c))[3:])or'/',
(the added newlines can all be removed)
Or, if you prefer not to use the bin() function in 2.6, we can get do it in 176:
for c in raw_input():C=lambda q:q>0and C(~-q/2)+'-.'[q%2]or'';print C(
(' etianmsurwdkgohvf_l_pjbxcyzq__54_3___2%7s16%7s7___8_90%12s?%8s.%29s,'%
(('',)*5)).find(c.lower()))or'/',
(again, the added newlines can all be removed)
C89 (293 characters)
Based off some of the other answers.
EDIT: Shrunk the tree (yay).
#define P putchar
char t['~']="~ETIANMSURWDKGOHVF~L~PJBXCYZQ~~54~3",o,q[9],Q=10;main(c){for(;Q;)t[
"&./7;=>KTr"[--Q]]="2167890?.,"[Q];while((c=getchar())>=0){c-=c<'{'&c>96?32:0;c-
10?c-32?0:P(47):P(10);for(o=1;o<'~';++o)if(t[o]==c){for(;o;o/=2)q[Q++]=45+(o--&1
);for(;Q;P(q[--Q]));break;}P(32);}}
Here's another approach, based on dmckee's work, demonstrating just how readable Python is:
Python
244 characters
def h(l):p=2*ord(l.upper())-88;a,n=map(ord,"AF__GF__]E\\E[EZEYEXEWEVEUETE__________CF__IBPDJDPBGAHDPC[DNBSDJCKDOBJBTCND`DKCQCHAHCZDSCLD??OD"[p:p+2]);return "--..--..-.-.-..--...----.....-----.-"[a-64:a+n-128]
def e(s):return ' '.join(map(h,s))
Limitations:
dmckee's string missed the 'Y' character, and I was too lazy to add it. I think you'd just have to change the "??" part, and add a "-" at the end of the second string literal
it doesn't put '/' between words; again, lazy
Since the rules called for fewest characters, not fewest bytes, you could make at least one of my lookup tables smaller (by half) if you were willing to go outside the printable ASCII characters.
EDIT: If I use naïvely-chosen Unicode chars but just keep them in escaped ASCII in the source file, it still gets a tad shorter because the decoder is simpler:
Python
240 characters
def h(l):a,n=divmod(ord(u'\x06_7_\xd0\xc9\xc2\xbb\xb4\xad\xa6\x9f\x98\x91_____\x14_AtJr2<s\xc1d\x89IQdH\x8ff\xe4Pz9;\xba\x88X_f'[ord(l.upper())-44]),7);return "--..--..-.-.-..--...----.....-----.-"[a:a+n]
def e(s):return ' '.join(map(h,s))
I think it also makes the intent of the program much clearer.
If you saved this as UTF-8, I believe the program would be down to 185 characters, making it the shortest complete Python solution, and second only to Perl. :-)
Here's a third, completely different way of encoding morse code:
Python
232 characters
def d(c):
o='';b=ord("Y_j_?><80 !#'/_____f_\x06\x11\x15\x05\x02\x15\t\x1c\x06\x1e\r\x12\x07\x05\x0f\x16\x1b\n\x08\x03\r\x18\x0e\x19\x01\x13"[ord(c.upper())-44])
while b!=1:o+='.-'[b&1];b/=2
return o
e=lambda s:' '.join(map(d,s))
If you can figure out a way to map this onto some set of printable characters, you could save quite a few characters. This is probably my most direct solution, though I don't know if it's the most readable.
OK, now I've wasted way too much time on this.
Haskell
type MorseCode = String
program :: String
program = "__5__4H___3VS__F___2 UI__L__+_ R__P___1JWAE"
++ "__6__=B__/_XD__C__YKN__7_Z__QG__8_ __9__0 OMT "
decode :: MorseCode -> String
decode = interpret program
where
interpret = head . foldl exec []
exec xs '_' = undefined : xs
exec (x:y:xs) c = branch : xs
where
branch (' ':ds) = c : decode ds
branch ('-':ds) = x ds
branch ('.':ds) = y ds
branch [] = [c]
For example, decode "-- --- .-. ... . -.-. --- -.. ." returns "MORSE CODE".
This program is from taken from the excellent article Fun with Morse Code.
PHP
I modified the previous PHP entry to be slightly more efficient. :)
$a=array(32=>"/",44=>"--..--",1,".-.-.-",48=>"-----",".----","..---","...--","....-",".....","-....","--...","---..","----.",63=>"..--..",1,".-","-...","-.-.","-..",".","..-.","--.","....","..",".---","-.-",".-..","--","-.","---",".--.","--.-",".-.","...","-","..-","...-",".--","-..-","-.--","--..");
foreach(str_split(strtoupper("hello world?"))as$k=>$v){echo $a[ord($v)]." ";}
Komodo says 380 characters on 2 lines - the extra line is just for readability. ;D
The interspersed 1s in the array is just to save 2 bytes by filling that array position with data instead of manually jumping to the array position after that.
Consider the first vs. the second. The difference is clearly visible. :)
array(20=>"data",22=>"more data";
array(20=>"data",1,"more data";
The end result, however, is exactly as long as you use the array positions rather than loop through the contents, which we don't do on this golf course.
End result: 578 characters, down to 380 (198 characters, or ~34.26% savings).
Bash, a script I wrote a while ago (time-stamp says last year) weighing in at a hefty 1661 characters. Just for fun really :)
#!/bin/sh
txt=''
res=''
if [ "$1" == '' ]; then
read -se txt
else
txt="$1"
fi;
len=$(echo "$txt" | wc -c)
k=1
while [ "$k" -lt "$len" ]; do
case "$(expr substr "$txt" $k 1 | tr '[:upper:]' '[:lower:]')" in
'e') res="$res"'.' ;;
't') res="$res"'-' ;;
'i') res="$res"'..' ;;
'a') res="$res"'.-' ;;
'n') res="$res"'-.' ;;
'm') res="$res"'--' ;;
's') res="$res"'...' ;;
'u') res="$res"'..-' ;;
'r') res="$res"'.-.' ;;
'w') res="$res"'.--' ;;
'd') res="$res"'-..' ;;
'k') res="$res"'-.-' ;;
'g') res="$res"'--.' ;;
'o') res="$res"'---' ;;
'h') res="$res"'....' ;;
'v') res="$res"'...-' ;;
'f') res="$res"'..-.' ;;
'l') res="$res"'.-..' ;;
'p') res="$res"'.--.' ;;
'j') res="$res"'.---' ;;
'b') res="$res"'-...' ;;
'x') res="$res"'-..-' ;;
'c') res="$res"'-.-.' ;;
'y') res="$res"'-.--' ;;
'z') res="$res"'--..' ;;
'q') res="$res"'--.-' ;;
'5') res="$res"'.....' ;;
'4') res="$res"'....-' ;;
'3') res="$res"'...--' ;;
'2') res="$res"'..---' ;;
'1') res="$res"'.----' ;;
'6') res="$res"'-....' ;;
'7') res="$res"'--...' ;;
'8') res="$res"'---..' ;;
'9') res="$res"'----.' ;;
'0') res="$res"'-----' ;;
esac;
[ ! "$(expr substr "$txt" $k 1)" == " " ] && [ ! "$(expr substr "$txt" $(($k+1)) 1)" == ' ' ] && res="$res"' '
k=$(($k+1))
done;
echo "$res"
C89 (388 characters)
This is incomplete as it doesn't handle comma, fullstop, and query yet.
#define P putchar
char q[10],Q,tree[]=
"EISH54V 3UF 2ARL + WP J 1TNDB6=X/ KC Y MGZ7 Q O 8 90";s2;e(x){q[Q++]
=x;}p(){for(;Q--;putchar(q[Q]));Q=0;}T(int x,char*t,int s){s2=s/2;return s?*t-x
?t[s2]-x?T(x,++t+s2,--s/2)?e(45):T(x,t,--s/2)?e(46):0:e(45):e(46):0;}main(c){
while((c=getchar())>=0){c-=c<123&&c>96?32:0;if(c==10)P(10);if(c==32)P(47);else
T(c,tree,sizeof(tree)),p();P(' ');}}
Wrapped for readability. Only two of the linebreaks are required (one for the #define, one after else, which could be a space). I've added a few non-standard characters but didn't add non-7-bit ones.
C, 533 characters
I took advice from some comments and switched to stdin. Killed another 70 characters roughly.
#include <stdio.h>
#include <ctype.h>
char *u[36] = {".-","-...","-.-.","-..",".","..-.","--.","....","..",".---","-.-",".-..","--","-.","---",".--.","--.-",".-.","...","-","..-","...-",".--","-..-","-.--","--..","-----",".----","..---","...--","....-",".....","-....","--...","---..","----."};
main(){
char*v;int x;char o;
do{
o = toupper(getc(stdin));v=0;if(o>=65&&o<=90)v=u[o-'A'];if(o>=48&&o<=57)v=u[o-'0'+26];if(o==46)v=".-.-.-";if(o==44)v="--..--";if(o==63)v="..--..";if(o==32)v="/";if(v)printf("%s ", v);} while (o != EOF);
}
C (381 characters)
char*p[36]={".-","-...","-.-.","-..",".","..-.","--.","....","..",".---","-.-",".-..","--","-.","---",".--.","--.-",".-.","...","-","..-","...-",".--","-..-","-.--","--..","-----",".----","..---","...--","....-",".....","-....","--...","---..","----."};
main(){int c;while((c=tolower(getchar()))!=10)printf("%s ",c==46?".-.-.-":c==44?"--..--":c==63?"..--..":c==32?"/":*(p+(c-97)));}
C, 448 bytes using cmdline arguments:
char*a[]={".-.-.-","--..--","..--..","/",".-","-...","-.-.","-..",".","..-.","--.","....","..",".---","-.-",".-..","--","-.","---",".--.","--.-",".-.","...","-","..-","...-",".--","-..-","-.--","--..","-----",".----","..---","...--","....-",".....","-....","--...","---..","----."},*k=".,? ",*s,*p,x;main(int _,char**v){for(;s=*++v;putchar(10))for(;x=*s++;){p=strchr(k,x);printf("%s ",p?a[p-k]:isdigit(x)?a[x-18]:isalpha(x=toupper(x))?a[x-61]:0);}}
C, 416 bytes using stdin:
char*a[]={".-.-.-","--..--","..--..","/",".-","-...","-.-.","-..",".","..-.","--.","....","..",".---","-.-",".-..","--","-.","---",".--.","--.-",".-.","...","-","..-","...-",".--","-..-","-.--","--..","-----",".----","..---","...--","....-",".....","-....","--...","---..","----."},*k=".,? ",*p,x;main(){while((x=toupper(getchar()))-10){p=strchr(k,x);printf("%s ",p?a[p-k]:isdigit(x)?a[x-18]:isalpha(x)?a[x-61]:0);}}