src/shared/utils.pas

   1 (* Copyright (C)  DooM 2D:Forever Developers
   2  *
   3  * This program is free software: you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation, either version 3 of the License, or
   6  * (at your option) any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License
  14  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  15  *)
  16 {$MODE DELPHI}
  17 unit utils;
  18
  19 interface
  20
  21 uses
  22   SysUtils, Classes;
  23
  24
  25 // does filename have one of ".wad", ".pk3", ".zip" extensions?
  26 function hasWadExtension (fn: AnsiString): Boolean;
  27
  28 // does filepath have ".XXX:\" in it?
  29 function isWadPath (fn: AnsiString): Boolean;
  30
  31 // adds ".wad" extension if filename doesn't have one of ".wad", ".pk3", ".zip"
  32 function addWadExtension (fn: AnsiString): AnsiString;
  33
  34 // convert number to strig with nice commas
  35 function Int64ToStrComma (i: Int64): AnsiString;
  36
  37 function UpCase1251 (ch: Char): Char;
  38 function LoCase1251 (ch: Char): Char;
  39
  40 // `true` if strings are equal; ignoring case for cp1251
  41 function StrEquCI1251 (const s0, s1: AnsiString): Boolean;
  42
  43 function utf8Valid (const s: AnsiString): Boolean;
  44
  45 function utf8to1251 (s: AnsiString): AnsiString;
  46
  47 // `pathname` will be modified if path is valid
  48 // `lastIsDir` should be `true` if we are searching for directory
  49 // nobody cares about shitdoze, so i'll use the same code path for it
  50 function findFileCI (var pathname: AnsiString; lastIsDir: Boolean=false): Boolean;
  51
  52 // they throws
  53 function openDiskFileRO (pathname: AnsiString): TStream;
  54 function createDiskFile (pathname: AnsiString): TStream;
  55
  56 // little endian
  57 procedure writeInt (st: TStream; v: Byte); overload;
  58 procedure writeInt (st: TStream; v: ShortInt); overload;
  59 procedure writeInt (st: TStream; v: Word); overload;
  60 procedure writeInt (st: TStream; v: SmallInt); overload;
  61 procedure writeInt (st: TStream; v: LongWord); overload;
  62 procedure writeInt (st: TStream; v: LongInt); overload;
  63 procedure writeInt (st: TStream; v: Int64); overload;
  64 procedure writeInt (st: TStream; v: UInt64); overload;
  65
  66 function readByte (st: TStream): Byte;
  67 function readShortInt (st: TStream): ShortInt;
  68 function readWord (st: TStream): Word;
  69 function readSmallInt (st: TStream): SmallInt;
  70 function readLongWord (st: TStream): LongWord;
  71 function readLongInt (st: TStream): LongInt;
  72 function readInt64 (st: TStream): Int64;
  73 function readUInt64 (st: TStream): UInt64;
  74
  75 // big endian
  76 procedure writeIntBE (st: TStream; v: Byte); overload;
  77 procedure writeIntBE (st: TStream; v: ShortInt); overload;
  78 procedure writeIntBE (st: TStream; v: Word); overload;
  79 procedure writeIntBE (st: TStream; v: SmallInt); overload;
  80 procedure writeIntBE (st: TStream; v: LongWord); overload;
  81 procedure writeIntBE (st: TStream; v: LongInt); overload;
  82 procedure writeIntBE (st: TStream; v: Int64); overload;
  83 procedure writeIntBE (st: TStream; v: UInt64); overload;
  84
  85 function readByteBE (st: TStream): Byte;
  86 function readShortIntBE (st: TStream): ShortInt;
  87 function readWordBE (st: TStream): Word;
  88 function readSmallIntBE (st: TStream): SmallInt;
  89 function readLongWordBE (st: TStream): LongWord;
  90 function readLongIntBE (st: TStream): LongInt;
  91 function readInt64BE (st: TStream): Int64;
  92 function readUInt64BE (st: TStream): UInt64;
  93
  94
  95 implementation
  96
  97
  98 function hasWadExtension (fn: AnsiString): Boolean;
  99 begin
 100   fn := ExtractFileExt(fn);
 101   result := StrEquCI1251(fn, '.wad') or StrEquCI1251(fn, '.pk3') or StrEquCI1251(fn, '.zip');
 102 end;
 103
 104
 105 function addWadExtension (fn: AnsiString): AnsiString;
 106 begin
 107   result := fn;
 108   if not hasWadExtension(result) then result := result+'.wad';
 109 end;
 110
 111
 112 function isWadPath (fn: AnsiString): Boolean;
 113 var
 114   p: Integer;
 115   s: AnsiString;
 116 begin
 117   result := false;
 118   while true do
 119   begin
 120     p := Pos(':', fn);
 121     if (p = 0) or (length(fn)-p < 1) then break;
 122     if (p-4 > 1) and (fn[p-4] = '.') and ((fn[p+1] = '\') or (fn[p+1] = '/')) then
 123     begin
 124       s := Copy(fn, p-4, 4);
 125       if StrEquCI1251(s, '.wad') or StrEquCI1251(s, '.pk3') or StrEquCI1251(s, '.zip') then
 126       begin
 127         result := true;
 128         exit;
 129       end;
 130     end;
 131     Delete(fn, 1, p);
 132   end;
 133 end;
 134
 135
 136 function Int64ToStrComma (i: Int64): AnsiString;
 137 var
 138   f: Integer;
 139 begin
 140   Str(i, result);
 141   f := Length(result)+1;
 142   while f > 4 do
 143   begin
 144     Dec(f, 3); Insert(',', result, f);
 145   end;
 146 end;
 147
 148
 149 function UpCase1251 (ch: Char): Char;
 150 begin
 151   if ch < #128 then
 152   begin
 153     if (ch >= 'a') and (ch <= 'z') then Dec(ch, 32);
 154   end
 155   else
 156   begin
 157     if (ch >= #224) and (ch <= #255) then
 158     begin
 159       Dec(ch, 32);
 160     end
 161     else
 162     begin
 163       case ch of
 164         #184, #186, #191: Dec(ch, 16);
 165         #162, #179: Dec(ch);
 166       end;
 167     end;
 168   end;
 169   result := ch;
 170 end;
 171
 172
 173 function LoCase1251 (ch: Char): Char;
 174 begin
 175   if ch < #128 then
 176   begin
 177     if (ch >= 'A') and (ch <= 'Z') then Inc(ch, 32);
 178   end
 179   else
 180   begin
 181     if (ch >= #192) and (ch <= #223) then
 182     begin
 183       Inc(ch, 32);
 184     end
 185     else
 186     begin
 187       case ch of
 188         #168, #170, #175: Inc(ch, 16);
 189         #161, #178: Inc(ch);
 190       end;
 191     end;
 192   end;
 193   result := ch;
 194 end;
 195
 196
 197 function StrEquCI1251 (const s0, s1: AnsiString): Boolean;
 198 var
 199   i: Integer;
 200 begin
 201   result := false;
 202   if length(s0) <> length(s1) then exit;
 203   for i := 1 to length(s0) do if UpCase1251(s0[i]) <> UpCase1251(s1[i]) then exit;
 204   result := true;
 205 end;
 206
 207
 208 // ////////////////////////////////////////////////////////////////////////// //
 209 // utils
 210 // `ch`: utf8 start
 211 // -1: invalid utf8
 212 function utf8CodeLen (ch: Word): Integer;
 213 begin
 214   if ch < $80 then begin result := 1; exit; end;
 215   if (ch and $FE) = $FC then begin result := 6; exit; end;
 216   if (ch and $FC) = $F8 then begin result := 5; exit; end;
 217   if (ch and $F8) = $F0 then begin result := 4; exit; end;
 218   if (ch and $F0) = $E0 then begin result := 3; exit; end;
 219   if (ch and $E0) = $C0 then begin result := 2; exit; end;
 220   result := -1; // invalid
 221 end;
 222
 223
 224 function utf8Valid (const s: AnsiString): Boolean;
 225 var
 226   pos, len: Integer;
 227 begin
 228   result := false;
 229   pos := 1;
 230   while pos <= length(s) do
 231   begin
 232     len := utf8CodeLen(Byte(s[pos]));
 233     if len < 1 then exit; // invalid sequence start
 234     if pos+len-1 > length(s) then exit; // out of chars in string
 235     Dec(len);
 236     Inc(pos);
 237     // check other sequence bytes
 238     while len > 0 do
 239     begin
 240       if (Byte(s[pos]) and $C0) <> $80 then exit;
 241       Dec(len);
 242       Inc(pos);
 243     end;
 244   end;
 245   result := true;
 246 end;
 247
 248
 249 // ////////////////////////////////////////////////////////////////////////// //
 250 const
 251   uni2wint: array [128..255] of Word = (
 252     $0402,$0403,$201A,$0453,$201E,$2026,$2020,$2021,$20AC,$2030,$0409,$2039,$040A,$040C,$040B,$040F,
 253     $0452,$2018,$2019,$201C,$201D,$2022,$2013,$2014,$003F,$2122,$0459,$203A,$045A,$045C,$045B,$045F,
 254     $00A0,$040E,$045E,$0408,$00A4,$0490,$00A6,$00A7,$0401,$00A9,$0404,$00AB,$00AC,$00AD,$00AE,$0407,
 255     $00B0,$00B1,$0406,$0456,$0491,$00B5,$00B6,$00B7,$0451,$2116,$0454,$00BB,$0458,$0405,$0455,$0457,
 256     $0410,$0411,$0412,$0413,$0414,$0415,$0416,$0417,$0418,$0419,$041A,$041B,$041C,$041D,$041E,$041F,
 257     $0420,$0421,$0422,$0423,$0424,$0425,$0426,$0427,$0428,$0429,$042A,$042B,$042C,$042D,$042E,$042F,
 258     $0430,$0431,$0432,$0433,$0434,$0435,$0436,$0437,$0438,$0439,$043A,$043B,$043C,$043D,$043E,$043F,
 259     $0440,$0441,$0442,$0443,$0444,$0445,$0446,$0447,$0448,$0449,$044A,$044B,$044C,$044D,$044E,$044F
 260   );
 261
 262
 263 function decodeUtf8Char (s: AnsiString; var pos: Integer): char;
 264 var
 265   b, c: Integer;
 266 begin
 267   (* The following encodings are valid, except for the 5 and 6 byte
 268    * combinations:
 269    *  0xxxxxxx
 270    *  110xxxxx 10xxxxxx
 271    *  1110xxxx 10xxxxxx 10xxxxxx
 272    *  11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
 273    *  111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
 274    *  1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
 275    *)
 276   result := '?';
 277   if pos > length(s) then exit;
 278
 279   b := Byte(s[pos]);
 280   Inc(pos);
 281   if b < $80 then begin result := char(b); exit; end;
 282
 283   // mask out unused bits
 284        if (b and $FE) = $FC then b := b and $01
 285   else if (b and $FC) = $F8 then b := b and $03
 286   else if (b and $F8) = $F0 then b := b and $07
 287   else if (b and $F0) = $E0 then b := b and $0F
 288   else if (b and $E0) = $C0 then b := b and $1F
 289   else exit; // invalid utf8
 290
 291   // now continue
 292   while pos <= length(s) do
 293   begin
 294     c := Byte(s[pos]);
 295     if (c and $C0) <> $80 then break; // no more
 296     b := b shl 6;
 297     b := b or (c and $3F);
 298     Inc(pos);
 299   end;
 300
 301   // done, try 1251
 302   for c := 128 to 255 do if uni2wint[c] = b then begin result := char(c and $FF); exit; end;
 303   // alas
 304 end;
 305
 306
 307 function utf8to1251 (s: AnsiString): AnsiString;
 308 var
 309   pos: Integer;
 310 begin
 311   if not utf8Valid(s) then begin result := s; exit; end;
 312   pos := 1;
 313   while pos <= length(s) do
 314   begin
 315     if Byte(s[pos]) >= $80 then break;
 316     Inc(pos);
 317   end;
 318   if pos > length(s) then begin result := s; exit; end; // nothing to do here
 319   result := '';
 320   pos := 1;
 321   while pos <= length(s) do result := result+decodeUtf8Char(s, pos);
 322 end;
 323
 324
 325 // ////////////////////////////////////////////////////////////////////////// //
 326 // `pathname` will be modified if path is valid
 327 // `lastIsDir` should be `true` if we are searching for directory
 328 // nobody cares about shitdoze, so i'll use the same code path for it
 329 function findFileCI (var pathname: AnsiString; lastIsDir: Boolean=false): Boolean;
 330 var
 331   sr: TSearchRec;
 332   npt: AnsiString;
 333   newname: AnsiString = '';
 334   curname: AnsiString;
 335   wantdir: Boolean;
 336   attr: LongInt;
 337   foundher: Boolean;
 338 begin
 339   npt := pathname;
 340   result := (length(npt) > 0);
 341   if (length(npt) > 0) and ((npt[1] = '/') or (npt[1] = '\')) then newname := '/';
 342   while length(npt) > 0 do
 343   begin
 344     // remove trailing slashes
 345     while (length(npt) > 0) and ((npt[1] = '/') or (npt[1] = '\')) do Delete(npt, 1, 1);
 346     if length(npt) = 0 then break;
 347     // extract name
 348     curname := '';
 349     while (length(npt) > 0) and (npt[1] <> '/') and (npt[1] <> '\') do
 350     begin
 351       curname := curname+npt[1];
 352       Delete(npt, 1, 1);
 353     end;
 354     // remove trailing slashes again
 355     while (length(npt) > 0) and ((npt[1] = '/') or (npt[1] = '\')) do Delete(npt, 1, 1);
 356     wantdir := lastIsDir or (length(npt) > 0); // do we want directory here?
 357     //writeln(Format('npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d', [npt, newname, curname, Integer(wantdir)]));
 358     // try the easiest case first
 359     attr := FileGetAttr(newname+curname);
 360     if attr <> -1 then
 361     begin
 362       if wantdir = ((attr and faDirectory) <> 0) then
 363       begin
 364         // i found her!
 365         newname := newname+curname;
 366         if wantdir then newname := newname+'/';
 367         continue;
 368       end;
 369     end;
 370     //writeln(Format('npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d', [npt, newname, curname, Integer(wantdir)]));
 371     // alas, either not found, or invalid attributes
 372     foundher := false;
 373     try
 374       if FindFirst(newname+'*', faAnyFile, sr) = 0 then
 375       repeat
 376         if (wantdir = ((sr.attr and faDirectory) <> 0)) and StrEquCI1251(sr.name, curname) then
 377         begin
 378           // i found her!
 379           newname := newname+sr.name;
 380           if wantdir then newname := newname+'/';
 381           foundher := true;
 382           break;
 383         end;
 384       until FindNext(sr) <> 0;
 385     finally
 386       FindClose(sr);
 387     end;
 388     if not foundher then begin newname := ''; result := false; break; end;
 389   end;
 390   if result then pathname := newname;
 391 end;
 392
 393
 394 function openDiskFileRO (pathname: AnsiString): TStream;
 395 begin
 396   if not findFileCI(pathname) then raise Exception.Create('can''t open file "'+pathname+'"');
 397   result := TFileStream.Create(pathname, fmOpenRead or {fmShareDenyWrite}fmShareDenyNone);
 398 end;
 399
 400 function createDiskFile (pathname: AnsiString): TStream;
 401 var
 402   path: AnsiString;
 403 begin
 404   path := ExtractFilePath(pathname);
 405   if length(path) > 0 then
 406   begin
 407     if not findFileCI(path, true) then raise Exception.Create('can''t create file "'+pathname+'"');
 408   end;
 409   result := TFileStream.Create(path+ExtractFileName(pathname), fmCreate);
 410 end;
 411
 412
 413 procedure writeIntegerLE (st: TStream; vp: Pointer; size: Integer);
 414 {$IFDEF ENDIAN_LITTLE}
 415 begin
 416   st.writeBuffer(vp^, size);
 417 end;
 418 {$ELSE}
 419 var
 420   p: PByte;
 421 begin
 422   p := PByte(vp)+size-1;
 423   while size > 0 do
 424   begin
 425     st.writeBuffer(p^, 1);
 426     Dec(size);
 427     Dec(p);
 428   end;
 429 end;
 430 {$ENDIF}
 431
 432 procedure writeIntegerBE (st: TStream; vp: Pointer; size: Integer);
 433 {$IFDEF ENDIAN_LITTLE}
 434 var
 435   p: PByte;
 436 begin
 437   p := PByte(vp)+size-1;
 438   while size > 0 do
 439   begin
 440     st.writeBuffer(p^, 1);
 441     Dec(size);
 442     Dec(p);
 443   end;
 444 end;
 445 {$ELSE}
 446 begin
 447   st.writeBuffer(vp^, size);
 448 end;
 449 {$ENDIF}
 450
 451 procedure writeInt (st: TStream; v: Byte); overload; begin writeIntegerLE(st, @v, 1); end;
 452 procedure writeInt (st: TStream; v: ShortInt); overload; begin writeIntegerLE(st, @v, 1); end;
 453 procedure writeInt (st: TStream; v: Word); overload; begin writeIntegerLE(st, @v, 2); end;
 454 procedure writeInt (st: TStream; v: SmallInt); overload; begin writeIntegerLE(st, @v, 2); end;
 455 procedure writeInt (st: TStream; v: LongWord); overload; begin writeIntegerLE(st, @v, 4); end;
 456 procedure writeInt (st: TStream; v: LongInt); overload; begin writeIntegerLE(st, @v, 4); end;
 457 procedure writeInt (st: TStream; v: Int64); overload; begin writeIntegerLE(st, @v, 8); end;
 458 procedure writeInt (st: TStream; v: UInt64); overload; begin writeIntegerLE(st, @v, 8); end;
 459
 460 procedure writeIntBE (st: TStream; v: Byte); overload; begin writeIntegerBE(st, @v, 1); end;
 461 procedure writeIntBE (st: TStream; v: ShortInt); overload; begin writeIntegerBE(st, @v, 1); end;
 462 procedure writeIntBE (st: TStream; v: Word); overload; begin writeIntegerBE(st, @v, 2); end;
 463 procedure writeIntBE (st: TStream; v: SmallInt); overload; begin writeIntegerBE(st, @v, 2); end;
 464 procedure writeIntBE (st: TStream; v: LongWord); overload; begin writeIntegerBE(st, @v, 4); end;
 465 procedure writeIntBE (st: TStream; v: LongInt); overload; begin writeIntegerBE(st, @v, 4); end;
 466 procedure writeIntBE (st: TStream; v: Int64); overload; begin writeIntegerBE(st, @v, 8); end;
 467 procedure writeIntBE (st: TStream; v: UInt64); overload; begin writeIntegerBE(st, @v, 8); end;
 468
 469
 470 procedure readIntegerLE (st: TStream; vp: Pointer; size: Integer);
 471 {$IFDEF ENDIAN_LITTLE}
 472 begin
 473   st.readBuffer(vp^, size);
 474 end;
 475 {$ELSE}
 476 var
 477   p: PByte;
 478 begin
 479   p := PByte(vp)+size-1;
 480   while size > 0 do
 481   begin
 482     st.readBuffer(p^, 1);
 483     Dec(size);
 484     Dec(p);
 485   end;
 486 end;
 487 {$ENDIF}
 488
 489 procedure readIntegerBE (st: TStream; vp: Pointer; size: Integer);
 490 {$IFDEF ENDIAN_LITTLE}
 491 var
 492   p: PByte;
 493 begin
 494   p := PByte(vp)+size-1;
 495   while size > 0 do
 496   begin
 497     st.readBuffer(p^, 1);
 498     Dec(size);
 499     Dec(p);
 500   end;
 501 end;
 502 {$ELSE}
 503 begin
 504   st.readBuffer(vp^, size);
 505 end;
 506 {$ENDIF}
 507
 508 function readByte (st: TStream): Byte; begin readIntegerLE(st, @result, 1); end;
 509 function readShortInt (st: TStream): ShortInt; begin readIntegerLE(st, @result, 1); end;
 510 function readWord (st: TStream): Word; begin readIntegerLE(st, @result, 2); end;
 511 function readSmallInt (st: TStream): SmallInt; begin readIntegerLE(st, @result, 2); end;
 512 function readLongWord (st: TStream): LongWord; begin readIntegerLE(st, @result, 4); end;
 513 function readLongInt (st: TStream): LongInt; begin readIntegerLE(st, @result, 4); end;
 514 function readInt64 (st: TStream): Int64; begin readIntegerLE(st, @result, 8); end;
 515 function readUInt64 (st: TStream): UInt64; begin readIntegerLE(st, @result, 8); end;
 516
 517 function readByteBE (st: TStream): Byte; begin readIntegerBE(st, @result, 1); end;
 518 function readShortIntBE (st: TStream): ShortInt; begin readIntegerBE(st, @result, 1); end;
 519 function readWordBE (st: TStream): Word; begin readIntegerBE(st, @result, 2); end;
 520 function readSmallIntBE (st: TStream): SmallInt; begin readIntegerBE(st, @result, 2); end;
 521 function readLongWordBE (st: TStream): LongWord; begin readIntegerBE(st, @result, 4); end;
 522 function readLongIntBE (st: TStream): LongInt; begin readIntegerBE(st, @result, 4); end;
 523 function readInt64BE (st: TStream): Int64; begin readIntegerBE(st, @result, 8); end;
 524 function readUInt64BE (st: TStream): UInt64; begin readIntegerBE(st, @result, 8); end;
 525
 526
 527 end.