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 // returns name (the same if no file found)
  53 function findFileCIStr (pathname: AnsiString; lastIsDir: Boolean=false): AnsiString;
  54
  55 // they throws
  56 function openDiskFileRO (pathname: AnsiString): TStream;
  57 function createDiskFile (pathname: AnsiString): TStream;
  58
  59 // little endian
  60 procedure writeInt (st: TStream; v: Byte); overload;
  61 procedure writeInt (st: TStream; v: ShortInt); overload;
  62 procedure writeInt (st: TStream; v: Word); overload;
  63 procedure writeInt (st: TStream; v: SmallInt); overload;
  64 procedure writeInt (st: TStream; v: LongWord); overload;
  65 procedure writeInt (st: TStream; v: LongInt); overload;
  66 procedure writeInt (st: TStream; v: Int64); overload;
  67 procedure writeInt (st: TStream; v: UInt64); overload;
  68
  69 function readByte (st: TStream): Byte;
  70 function readShortInt (st: TStream): ShortInt;
  71 function readWord (st: TStream): Word;
  72 function readSmallInt (st: TStream): SmallInt;
  73 function readLongWord (st: TStream): LongWord;
  74 function readLongInt (st: TStream): LongInt;
  75 function readInt64 (st: TStream): Int64;
  76 function readUInt64 (st: TStream): UInt64;
  77
  78 // big endian
  79 procedure writeIntBE (st: TStream; v: Byte); overload;
  80 procedure writeIntBE (st: TStream; v: ShortInt); overload;
  81 procedure writeIntBE (st: TStream; v: Word); overload;
  82 procedure writeIntBE (st: TStream; v: SmallInt); overload;
  83 procedure writeIntBE (st: TStream; v: LongWord); overload;
  84 procedure writeIntBE (st: TStream; v: LongInt); overload;
  85 procedure writeIntBE (st: TStream; v: Int64); overload;
  86 procedure writeIntBE (st: TStream; v: UInt64); overload;
  87
  88 function readByteBE (st: TStream): Byte;
  89 function readShortIntBE (st: TStream): ShortInt;
  90 function readWordBE (st: TStream): Word;
  91 function readSmallIntBE (st: TStream): SmallInt;
  92 function readLongWordBE (st: TStream): LongWord;
  93 function readLongIntBE (st: TStream): LongInt;
  94 function readInt64BE (st: TStream): Int64;
  95 function readUInt64BE (st: TStream): UInt64;
  96
  97
  98 implementation
  99
 100
 101 function hasWadExtension (fn: AnsiString): Boolean;
 102 begin
 103   fn := ExtractFileExt(fn);
 104   result := StrEquCI1251(fn, '.wad') or StrEquCI1251(fn, '.pk3') or StrEquCI1251(fn, '.zip');
 105 end;
 106
 107
 108 function addWadExtension (fn: AnsiString): AnsiString;
 109 begin
 110   result := fn;
 111   if not hasWadExtension(result) then result := result+'.wad';
 112 end;
 113
 114
 115 function isWadPath (fn: AnsiString): Boolean;
 116 var
 117   p: Integer;
 118   s: AnsiString;
 119 begin
 120   result := false;
 121   while true do
 122   begin
 123     p := Pos(':', fn);
 124     if (p = 0) or (length(fn)-p < 1) then break;
 125     if (p-4 > 1) and (fn[p-4] = '.') and ((fn[p+1] = '\') or (fn[p+1] = '/')) then
 126     begin
 127       s := Copy(fn, p-4, 4);
 128       if StrEquCI1251(s, '.wad') or StrEquCI1251(s, '.pk3') or StrEquCI1251(s, '.zip') then
 129       begin
 130         result := true;
 131         exit;
 132       end;
 133     end;
 134     Delete(fn, 1, p);
 135   end;
 136 end;
 137
 138
 139 function Int64ToStrComma (i: Int64): AnsiString;
 140 var
 141   f: Integer;
 142 begin
 143   Str(i, result);
 144   f := Length(result)+1;
 145   while f > 4 do
 146   begin
 147     Dec(f, 3); Insert(',', result, f);
 148   end;
 149 end;
 150
 151
 152 function UpCase1251 (ch: Char): Char;
 153 begin
 154   if ch < #128 then
 155   begin
 156     if (ch >= 'a') and (ch <= 'z') then Dec(ch, 32);
 157   end
 158   else
 159   begin
 160     if (ch >= #224) and (ch <= #255) then
 161     begin
 162       Dec(ch, 32);
 163     end
 164     else
 165     begin
 166       case ch of
 167         #184, #186, #191: Dec(ch, 16);
 168         #162, #179: Dec(ch);
 169       end;
 170     end;
 171   end;
 172   result := ch;
 173 end;
 174
 175
 176 function LoCase1251 (ch: Char): Char;
 177 begin
 178   if ch < #128 then
 179   begin
 180     if (ch >= 'A') and (ch <= 'Z') then Inc(ch, 32);
 181   end
 182   else
 183   begin
 184     if (ch >= #192) and (ch <= #223) then
 185     begin
 186       Inc(ch, 32);
 187     end
 188     else
 189     begin
 190       case ch of
 191         #168, #170, #175: Inc(ch, 16);
 192         #161, #178: Inc(ch);
 193       end;
 194     end;
 195   end;
 196   result := ch;
 197 end;
 198
 199
 200 function StrEquCI1251 (const s0, s1: AnsiString): Boolean;
 201 var
 202   i: Integer;
 203 begin
 204   result := false;
 205   if length(s0) <> length(s1) then exit;
 206   for i := 1 to length(s0) do if UpCase1251(s0[i]) <> UpCase1251(s1[i]) then exit;
 207   result := true;
 208 end;
 209
 210
 211 // ////////////////////////////////////////////////////////////////////////// //
 212 // utils
 213 // `ch`: utf8 start
 214 // -1: invalid utf8
 215 function utf8CodeLen (ch: Word): Integer;
 216 begin
 217   if ch < $80 then begin result := 1; exit; end;
 218   if (ch and $FE) = $FC then begin result := 6; exit; end;
 219   if (ch and $FC) = $F8 then begin result := 5; exit; end;
 220   if (ch and $F8) = $F0 then begin result := 4; exit; end;
 221   if (ch and $F0) = $E0 then begin result := 3; exit; end;
 222   if (ch and $E0) = $C0 then begin result := 2; exit; end;
 223   result := -1; // invalid
 224 end;
 225
 226
 227 function utf8Valid (const s: AnsiString): Boolean;
 228 var
 229   pos, len: Integer;
 230 begin
 231   result := false;
 232   pos := 1;
 233   while pos <= length(s) do
 234   begin
 235     len := utf8CodeLen(Byte(s[pos]));
 236     if len < 1 then exit; // invalid sequence start
 237     if pos+len-1 > length(s) then exit; // out of chars in string
 238     Dec(len);
 239     Inc(pos);
 240     // check other sequence bytes
 241     while len > 0 do
 242     begin
 243       if (Byte(s[pos]) and $C0) <> $80 then exit;
 244       Dec(len);
 245       Inc(pos);
 246     end;
 247   end;
 248   result := true;
 249 end;
 250
 251
 252 // ////////////////////////////////////////////////////////////////////////// //
 253 const
 254   uni2wint: array [128..255] of Word = (
 255     $0402,$0403,$201A,$0453,$201E,$2026,$2020,$2021,$20AC,$2030,$0409,$2039,$040A,$040C,$040B,$040F,
 256     $0452,$2018,$2019,$201C,$201D,$2022,$2013,$2014,$003F,$2122,$0459,$203A,$045A,$045C,$045B,$045F,
 257     $00A0,$040E,$045E,$0408,$00A4,$0490,$00A6,$00A7,$0401,$00A9,$0404,$00AB,$00AC,$00AD,$00AE,$0407,
 258     $00B0,$00B1,$0406,$0456,$0491,$00B5,$00B6,$00B7,$0451,$2116,$0454,$00BB,$0458,$0405,$0455,$0457,
 259     $0410,$0411,$0412,$0413,$0414,$0415,$0416,$0417,$0418,$0419,$041A,$041B,$041C,$041D,$041E,$041F,
 260     $0420,$0421,$0422,$0423,$0424,$0425,$0426,$0427,$0428,$0429,$042A,$042B,$042C,$042D,$042E,$042F,
 261     $0430,$0431,$0432,$0433,$0434,$0435,$0436,$0437,$0438,$0439,$043A,$043B,$043C,$043D,$043E,$043F,
 262     $0440,$0441,$0442,$0443,$0444,$0445,$0446,$0447,$0448,$0449,$044A,$044B,$044C,$044D,$044E,$044F
 263   );
 264
 265
 266 function decodeUtf8Char (s: AnsiString; var pos: Integer): char;
 267 var
 268   b, c: Integer;
 269 begin
 270   (* The following encodings are valid, except for the 5 and 6 byte
 271    * combinations:
 272    *  0xxxxxxx
 273    *  110xxxxx 10xxxxxx
 274    *  1110xxxx 10xxxxxx 10xxxxxx
 275    *  11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
 276    *  111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
 277    *  1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
 278    *)
 279   result := '?';
 280   if pos > length(s) then exit;
 281
 282   b := Byte(s[pos]);
 283   Inc(pos);
 284   if b < $80 then begin result := char(b); exit; end;
 285
 286   // mask out unused bits
 287        if (b and $FE) = $FC then b := b and $01
 288   else if (b and $FC) = $F8 then b := b and $03
 289   else if (b and $F8) = $F0 then b := b and $07
 290   else if (b and $F0) = $E0 then b := b and $0F
 291   else if (b and $E0) = $C0 then b := b and $1F
 292   else exit; // invalid utf8
 293
 294   // now continue
 295   while pos <= length(s) do
 296   begin
 297     c := Byte(s[pos]);
 298     if (c and $C0) <> $80 then break; // no more
 299     b := b shl 6;
 300     b := b or (c and $3F);
 301     Inc(pos);
 302   end;
 303
 304   // done, try 1251
 305   for c := 128 to 255 do if uni2wint[c] = b then begin result := char(c and $FF); exit; end;
 306   // alas
 307 end;
 308
 309
 310 function utf8to1251 (s: AnsiString): AnsiString;
 311 var
 312   pos: Integer;
 313 begin
 314   if not utf8Valid(s) then begin result := s; exit; end;
 315   pos := 1;
 316   while pos <= length(s) do
 317   begin
 318     if Byte(s[pos]) >= $80 then break;
 319     Inc(pos);
 320   end;
 321   if pos > length(s) then begin result := s; exit; end; // nothing to do here
 322   result := '';
 323   pos := 1;
 324   while pos <= length(s) do result := result+decodeUtf8Char(s, pos);
 325 end;
 326
 327
 328 // ////////////////////////////////////////////////////////////////////////// //
 329 // `pathname` will be modified if path is valid
 330 // `lastIsDir` should be `true` if we are searching for directory
 331 // nobody cares about shitdoze, so i'll use the same code path for it
 332 function findFileCI (var pathname: AnsiString; lastIsDir: Boolean=false): Boolean;
 333 var
 334   sr: TSearchRec;
 335   npt: AnsiString;
 336   newname: AnsiString = '';
 337   curname: AnsiString;
 338   wantdir: Boolean;
 339   attr: LongInt;
 340   foundher: Boolean;
 341 begin
 342   npt := pathname;
 343   result := (length(npt) > 0);
 344   if (length(npt) > 0) and ((npt[1] = '/') or (npt[1] = '\')) then newname := '/';
 345   while length(npt) > 0 do
 346   begin
 347     // remove trailing slashes
 348     while (length(npt) > 0) and ((npt[1] = '/') or (npt[1] = '\')) do Delete(npt, 1, 1);
 349     if length(npt) = 0 then break;
 350     // extract name
 351     curname := '';
 352     while (length(npt) > 0) and (npt[1] <> '/') and (npt[1] <> '\') do
 353     begin
 354       curname := curname+npt[1];
 355       Delete(npt, 1, 1);
 356     end;
 357     // remove trailing slashes again
 358     while (length(npt) > 0) and ((npt[1] = '/') or (npt[1] = '\')) do Delete(npt, 1, 1);
 359     wantdir := lastIsDir or (length(npt) > 0); // do we want directory here?
 360     writeln(Format('0: npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d', [npt, newname, curname, Integer(wantdir)]));
 361     // try the easiest case first
 362     attr := FileGetAttr(newname+curname);
 363     if attr <> -1 then
 364     begin
 365       if wantdir = ((attr and faDirectory) <> 0) then
 366       begin
 367         // i found her!
 368         writeln(Format('3: npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d; found=tan', [npt, newname, curname, Integer(wantdir)]));
 369         newname := newname+curname;
 370         if wantdir then newname := newname+'/';
 371         continue;
 372       end;
 373     end;
 374     writeln(Format('1: npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d', [npt, newname, curname, Integer(wantdir)]));
 375     // alas, either not found, or invalid attributes
 376     foundher := false;
 377     try
 378       if FindFirst(newname+'*', faAnyFile, sr) = 0 then
 379       repeat
 380         if (wantdir = ((sr.attr and faDirectory) <> 0)) and StrEquCI1251(sr.name, curname) then
 381         begin
 382           // i found her!
 383           newname := newname+sr.name;
 384           if wantdir then newname := newname+'/';
 385           foundher := true;
 386           break;
 387         end;
 388       until FindNext(sr) <> 0;
 389     finally
 390       FindClose(sr);
 391     end;
 392     if (foundher) then
 393     begin
 394       writeln(Format('2: npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d; found=tan', [npt, newname, curname, Integer(wantdir)]));
 395     end
 396     else
 397     begin
 398       writeln(Format('2: npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d; found=ona', [npt, newname, curname, Integer(wantdir)]));
 399     end;
 400     if not foundher then begin newname := ''; result := false; break; end;
 401   end;
 402   writeln(Format('4: npt=[%s]; newname=[%s]; curname=[%s]; wantdir=%d; found=ona', [npt, newname, curname, Integer(wantdir)]));
 403   if result then pathname := newname;
 404 end;
 405
 406 function findFileCIStr (pathname: AnsiString; lastIsDir: Boolean=false): AnsiString;
 407 var
 408   s: AnsiString;
 409 begin
 410   s := pathname;
 411   if not findFileCI(s, lastIsDir) then s := pathname;
 412   writeln(Format('***: pathname=[%s]; s=[%s]', [pathname, s]));
 413   result := s;
 414 end;
 415
 416 function openDiskFileRO (pathname: AnsiString): TStream;
 417 begin
 418   if not findFileCI(pathname) then raise Exception.Create('can''t open file "'+pathname+'"');
 419   result := TFileStream.Create(pathname, fmOpenRead or {fmShareDenyWrite}fmShareDenyNone);
 420 end;
 421
 422 function createDiskFile (pathname: AnsiString): TStream;
 423 var
 424   path: AnsiString;
 425 begin
 426   path := ExtractFilePath(pathname);
 427   if length(path) > 0 then
 428   begin
 429     if not findFileCI(path, true) then raise Exception.Create('can''t create file "'+pathname+'"');
 430   end;
 431   result := TFileStream.Create(path+ExtractFileName(pathname), fmCreate);
 432 end;
 433
 434
 435 procedure writeIntegerLE (st: TStream; vp: Pointer; size: Integer);
 436 {$IFDEF ENDIAN_LITTLE}
 437 begin
 438   st.writeBuffer(vp^, size);
 439 end;
 440 {$ELSE}
 441 var
 442   p: PByte;
 443 begin
 444   p := PByte(vp)+size-1;
 445   while size > 0 do
 446   begin
 447     st.writeBuffer(p^, 1);
 448     Dec(size);
 449     Dec(p);
 450   end;
 451 end;
 452 {$ENDIF}
 453
 454 procedure writeIntegerBE (st: TStream; vp: Pointer; size: Integer);
 455 {$IFDEF ENDIAN_LITTLE}
 456 var
 457   p: PByte;
 458 begin
 459   p := PByte(vp)+size-1;
 460   while size > 0 do
 461   begin
 462     st.writeBuffer(p^, 1);
 463     Dec(size);
 464     Dec(p);
 465   end;
 466 end;
 467 {$ELSE}
 468 begin
 469   st.writeBuffer(vp^, size);
 470 end;
 471 {$ENDIF}
 472
 473 procedure writeInt (st: TStream; v: Byte); overload; begin writeIntegerLE(st, @v, 1); end;
 474 procedure writeInt (st: TStream; v: ShortInt); overload; begin writeIntegerLE(st, @v, 1); end;
 475 procedure writeInt (st: TStream; v: Word); overload; begin writeIntegerLE(st, @v, 2); end;
 476 procedure writeInt (st: TStream; v: SmallInt); overload; begin writeIntegerLE(st, @v, 2); end;
 477 procedure writeInt (st: TStream; v: LongWord); overload; begin writeIntegerLE(st, @v, 4); end;
 478 procedure writeInt (st: TStream; v: LongInt); overload; begin writeIntegerLE(st, @v, 4); end;
 479 procedure writeInt (st: TStream; v: Int64); overload; begin writeIntegerLE(st, @v, 8); end;
 480 procedure writeInt (st: TStream; v: UInt64); overload; begin writeIntegerLE(st, @v, 8); end;
 481
 482 procedure writeIntBE (st: TStream; v: Byte); overload; begin writeIntegerBE(st, @v, 1); end;
 483 procedure writeIntBE (st: TStream; v: ShortInt); overload; begin writeIntegerBE(st, @v, 1); end;
 484 procedure writeIntBE (st: TStream; v: Word); overload; begin writeIntegerBE(st, @v, 2); end;
 485 procedure writeIntBE (st: TStream; v: SmallInt); overload; begin writeIntegerBE(st, @v, 2); end;
 486 procedure writeIntBE (st: TStream; v: LongWord); overload; begin writeIntegerBE(st, @v, 4); end;
 487 procedure writeIntBE (st: TStream; v: LongInt); overload; begin writeIntegerBE(st, @v, 4); end;
 488 procedure writeIntBE (st: TStream; v: Int64); overload; begin writeIntegerBE(st, @v, 8); end;
 489 procedure writeIntBE (st: TStream; v: UInt64); overload; begin writeIntegerBE(st, @v, 8); end;
 490
 491
 492 procedure readIntegerLE (st: TStream; vp: Pointer; size: Integer);
 493 {$IFDEF ENDIAN_LITTLE}
 494 begin
 495   st.readBuffer(vp^, size);
 496 end;
 497 {$ELSE}
 498 var
 499   p: PByte;
 500 begin
 501   p := PByte(vp)+size-1;
 502   while size > 0 do
 503   begin
 504     st.readBuffer(p^, 1);
 505     Dec(size);
 506     Dec(p);
 507   end;
 508 end;
 509 {$ENDIF}
 510
 511 procedure readIntegerBE (st: TStream; vp: Pointer; size: Integer);
 512 {$IFDEF ENDIAN_LITTLE}
 513 var
 514   p: PByte;
 515 begin
 516   p := PByte(vp)+size-1;
 517   while size > 0 do
 518   begin
 519     st.readBuffer(p^, 1);
 520     Dec(size);
 521     Dec(p);
 522   end;
 523 end;
 524 {$ELSE}
 525 begin
 526   st.readBuffer(vp^, size);
 527 end;
 528 {$ENDIF}
 529
 530 function readByte (st: TStream): Byte; begin readIntegerLE(st, @result, 1); end;
 531 function readShortInt (st: TStream): ShortInt; begin readIntegerLE(st, @result, 1); end;
 532 function readWord (st: TStream): Word; begin readIntegerLE(st, @result, 2); end;
 533 function readSmallInt (st: TStream): SmallInt; begin readIntegerLE(st, @result, 2); end;
 534 function readLongWord (st: TStream): LongWord; begin readIntegerLE(st, @result, 4); end;
 535 function readLongInt (st: TStream): LongInt; begin readIntegerLE(st, @result, 4); end;
 536 function readInt64 (st: TStream): Int64; begin readIntegerLE(st, @result, 8); end;
 537 function readUInt64 (st: TStream): UInt64; begin readIntegerLE(st, @result, 8); end;
 538
 539 function readByteBE (st: TStream): Byte; begin readIntegerBE(st, @result, 1); end;
 540 function readShortIntBE (st: TStream): ShortInt; begin readIntegerBE(st, @result, 1); end;
 541 function readWordBE (st: TStream): Word; begin readIntegerBE(st, @result, 2); end;
 542 function readSmallIntBE (st: TStream): SmallInt; begin readIntegerBE(st, @result, 2); end;
 543 function readLongWordBE (st: TStream): LongWord; begin readIntegerBE(st, @result, 4); end;
 544 function readLongIntBE (st: TStream): LongInt; begin readIntegerBE(st, @result, 4); end;
 545 function readInt64BE (st: TStream): Int64; begin readIntegerBE(st, @result, 8); end;
 546 function readUInt64BE (st: TStream): UInt64; begin readIntegerBE(st, @result, 8); end;
 547
 548
 549 end.