X-Git-Url: https://deadsoftware.ru/gitweb?a=blobdiff_plain;ds=sidebyside;f=src%2Fshared%2Futils.pas;h=9e136495bc8792dc661c3e4fa254ee58d24974a6;hb=223356cbae3197afc861efa6241c4ae91bd92885;hp=36d59ff8bd1d5c22606d8b62e4b0727c0e389cba;hpb=7f60063e21b5764c01cf86883d2e1808af8b04ef;p=d2df-sdl.git

diff --git a/src/shared/utils.pas b/src/shared/utils.pas
index 36d59ff..9e13649 100644
--- a/src/shared/utils.pas
+++ b/src/shared/utils.pas
@@ -22,6 +22,36 @@ uses
   SysUtils, Classes;
 
 
+// ////////////////////////////////////////////////////////////////////////// //
+type
+  TUtf8DecoderFast = packed record
+  public
+    const Replacement = $FFFD; // replacement char for invalid unicode
+    const Accept = 0;
+    const Reject = 12;
+
+  private
+    state: LongWord;
+
+  public
+    codepoint: LongWord; // decoded codepoint (valid only when decoder is in "complete" state)
+
+  public
+    constructor Create (v: Boolean{fuck you, fpc});
+
+    procedure reset (); inline;
+
+    function complete (): Boolean; inline; // is current character complete? take `codepoint` then
+    function invalid (): Boolean; inline;
+    function completeOrInvalid (): Boolean; inline;
+
+    // process one byte, return `true` if codepoint is ready
+    function decode (b: Byte): Boolean; inline; overload;
+    function decode (c: AnsiChar): Boolean; inline; overload;
+  end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
 // does filename have one of ".wad", ".pk3", ".zip" extensions?
 function hasWadExtension (fn: AnsiString): Boolean;
 
@@ -98,10 +128,423 @@ type
 // returns formatted string if `writerCB` is `nil`, empty string otherwise
 function formatstrf (const fmt: AnsiString; args: array of const; writerCB: TFormatStrFCallback=nil): AnsiString;
 
+function wchar2win (wc: WideChar): AnsiChar; inline;
+function utf2win (const s: AnsiString): AnsiString;
+function win2utf (const s: AnsiString): AnsiString;
+function digitInBase (ch: AnsiChar; base: Integer): Integer;
+
+// returns string in single or double quotes
+// single quotes supports only pascal-style '' for single quote char
+// double quotes supports c-style escapes
+// function will select quote mode automatically
+function quoteStr (const s: AnsiString): AnsiString;
+
+
+type
+  generic TSimpleList<ItemT> = class
+  private
+    type PItemT = ^ItemT;
+
+  public
+    type
+      TEnumerator = record
+      private
+        mItems: PItemT;
+        mCount: Integer;
+        mCurrent: Integer;
+      public
+        constructor Create (aitems: PItemT; acount: Integer);
+        function MoveNext: Boolean;
+        function getCurrent (): ItemT;
+        property Current: ItemT read getCurrent;
+      end;
+
+  private
+    mItems: array of ItemT;
+    mCount: Integer; // can be less than `mItems` size
+
+  private
+    function getAt (idx: Integer): ItemT; inline;
+    procedure setAt (idx: Integer; const it: ItemT); inline;
+
+    function getCapacity (): Integer; inline;
+    procedure setCapacity (v: Integer); inline;
+
+  public
+    constructor Create (acapacity: Integer=-1);
+    destructor Destroy (); override;
+
+    //WARNING! don't change list contents in `for ... in`!
+    function GetEnumerator (): TEnumerator;
+
+    procedure reset (); inline; // won't resize `mItems`
+    procedure clear (); inline;
+
+    procedure append (constref it: ItemT); inline;
+
+  public
+    property count: Integer read mCount;
+    property capacity: Integer read getCapacity write setCapacity;
+    property at[idx: Integer]: ItemT read getAt write setAt; default;
+  end;
+
 
 implementation
 
 
+// ////////////////////////////////////////////////////////////////////////// //
+constructor TSimpleList.TEnumerator.Create (aitems: PItemT; acount: Integer);
+begin
+  mItems := aitems;
+  mCount := acount;
+  mCurrent := -1;
+end;
+
+function TSimpleList.TEnumerator.MoveNext: Boolean;
+begin
+  Inc(mCurrent);
+  result := (mCurrent < mCount);
+end;
+
+function TSimpleList.TEnumerator.getCurrent (): ItemT;
+begin
+  result := mItems[mCurrent];
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+constructor TSimpleList.Create (acapacity: Integer=-1);
+begin
+  mItems := nil;
+  if (acapacity > 0) then SetLength(mItems, acapacity);
+  mCount := 0;
+end;
+
+
+destructor TSimpleList.Destroy ();
+begin
+  mItems := nil;
+  inherited;
+end;
+
+
+function TSimpleList.getCapacity (): Integer; inline;
+begin
+  result := Length(mItems);
+end;
+
+
+procedure TSimpleList.setCapacity (v: Integer); inline;
+begin
+  if (v < mCount) then v := mCount;
+  if (v <> Length(mItems)) then SetLength(mItems, v);
+end;
+
+
+function TSimpleList.GetEnumerator (): TEnumerator;
+begin
+  if (Length(mItems) > 0) then result := TEnumerator.Create(@mItems[0], mCount)
+  else result := TEnumerator.Create(nil, -1);
+end;
+
+
+procedure TSimpleList.reset (); inline;
+begin
+  mCount := 0;
+end;
+
+
+procedure TSimpleList.clear (); inline;
+begin
+  mItems := nil;
+  mCount := 0;
+end;
+
+
+function TSimpleList.getAt (idx: Integer): ItemT; inline;
+begin
+  if (idx >= 0) and (idx < mCount) then result := mItems[idx] else result := Default(ItemT);
+end;
+
+
+procedure TSimpleList.setAt (idx: Integer; const it: ItemT); inline;
+begin
+  if (idx >= 0) and (idx < mCount) then mItems[idx] := it;
+end;
+
+
+procedure TSimpleList.append (constref it: ItemT); inline;
+begin
+  if (mCount = Length(mItems)) then
+  begin
+    if (mCount = 0) then SetLength(mItems, 128) else SetLength(mItems, mCount*2);
+  end;
+  mItems[mCount] := it;
+  Inc(mCount);
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+var
+  wc2shitmap: array[0..65535] of AnsiChar;
+  wc2shitmapInited: Boolean = false;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+const
+  cp1251: array[0..127] of Word = (
+    $0402,$0403,$201A,$0453,$201E,$2026,$2020,$2021,$20AC,$2030,$0409,$2039,$040A,$040C,$040B,$040F,
+    $0452,$2018,$2019,$201C,$201D,$2022,$2013,$2014,$003F,$2122,$0459,$203A,$045A,$045C,$045B,$045F,
+    $00A0,$040E,$045E,$0408,$00A4,$0490,$00A6,$00A7,$0401,$00A9,$0404,$00AB,$00AC,$00AD,$00AE,$0407,
+    $00B0,$00B1,$0406,$0456,$0491,$00B5,$00B6,$00B7,$0451,$2116,$0454,$00BB,$0458,$0405,$0455,$0457,
+    $0410,$0411,$0412,$0413,$0414,$0415,$0416,$0417,$0418,$0419,$041A,$041B,$041C,$041D,$041E,$041F,
+    $0420,$0421,$0422,$0423,$0424,$0425,$0426,$0427,$0428,$0429,$042A,$042B,$042C,$042D,$042E,$042F,
+    $0430,$0431,$0432,$0433,$0434,$0435,$0436,$0437,$0438,$0439,$043A,$043B,$043C,$043D,$043E,$043F,
+    $0440,$0441,$0442,$0443,$0444,$0445,$0446,$0447,$0448,$0449,$044A,$044B,$044C,$044D,$044E,$044F
+  );
+
+
+procedure initShitMap ();
+var
+  f: Integer;
+begin
+  for f := 0 to High(wc2shitmap) do wc2shitmap[f] := '?';
+  for f := 0 to 127 do wc2shitmap[f] := AnsiChar(f);
+  for f := 0 to 127 do wc2shitmap[cp1251[f]] := AnsiChar(f+128);
+  wc2shitmapInited := true;
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+// fast state-machine based UTF-8 decoder; using 8 bytes of memory
+// code points from invalid range will never be valid, this is the property of the state machine
+const
+  // see http://bjoern.hoehrmann.de/utf-8/decoder/dfa/
+  utf8dfa: array[0..$16c-1] of Byte = (
+    // maps bytes to character classes
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 00-0f
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 10-1f
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 20-2f
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 30-3f
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 40-4f
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 50-5f
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 60-6f
+    $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00, // 70-7f
+    $01,$01,$01,$01,$01,$01,$01,$01,$01,$01,$01,$01,$01,$01,$01,$01, // 80-8f
+    $09,$09,$09,$09,$09,$09,$09,$09,$09,$09,$09,$09,$09,$09,$09,$09, // 90-9f
+    $07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07, // a0-af
+    $07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07,$07, // b0-bf
+    $08,$08,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02, // c0-cf
+    $02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02,$02, // d0-df
+    $0a,$03,$03,$03,$03,$03,$03,$03,$03,$03,$03,$03,$03,$04,$03,$03, // e0-ef
+    $0b,$06,$06,$06,$05,$08,$08,$08,$08,$08,$08,$08,$08,$08,$08,$08, // f0-ff
+    // maps a combination of a state of the automaton and a character class to a state
+    $00,$0c,$18,$24,$3c,$60,$54,$0c,$0c,$0c,$30,$48,$0c,$0c,$0c,$0c, // 100-10f
+    $0c,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$00,$0c,$0c,$0c,$0c,$0c,$00, // 110-11f
+    $0c,$00,$0c,$0c,$0c,$18,$0c,$0c,$0c,$0c,$0c,$18,$0c,$18,$0c,$0c, // 120-12f
+    $0c,$0c,$0c,$0c,$0c,$0c,$0c,$18,$0c,$0c,$0c,$0c,$0c,$18,$0c,$0c, // 130-13f
+    $0c,$0c,$0c,$0c,$0c,$18,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$24, // 140-14f
+    $0c,$24,$0c,$0c,$0c,$24,$0c,$0c,$0c,$0c,$0c,$24,$0c,$24,$0c,$0c, // 150-15f
+    $0c,$24,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$0c,$0c);
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+constructor TUtf8DecoderFast.Create (v: Boolean{fuck you, fpc}); begin state := Accept; codepoint := 0; end;
+
+procedure TUtf8DecoderFast.reset (); inline; begin state := Accept; codepoint := 0; end;
+
+function TUtf8DecoderFast.complete (): Boolean; inline; begin result := (state = Accept); end;
+function TUtf8DecoderFast.invalid (): Boolean; inline; begin result := (state = Reject); end;
+function TUtf8DecoderFast.completeOrInvalid (): Boolean; inline; begin result := (state = Accept) or (state = Reject); end;
+
+function TUtf8DecoderFast.decode (c: AnsiChar): Boolean; inline; overload; begin result := decode(Byte(c)); end;
+
+function TUtf8DecoderFast.decode (b: Byte): Boolean; inline; overload;
+var
+  tp: LongWord;
+begin
+  if (state = Reject) then begin state := Accept; codepoint := 0; end;
+  tp := utf8dfa[b];
+  if (state <> Accept) then codepoint := (b and $3f) or (codepoint shl 6) else codepoint := ($ff shr tp) and b;
+  state := utf8dfa[256+state+tp];
+  if (state = Reject) then begin codepoint := Replacement; state := Accept; end;
+  result := (state = Accept);
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+function wchar2win (wc: WideChar): AnsiChar; inline;
+begin
+  if not wc2shitmapInited then initShitMap();
+  if (LongWord(wc) > 65535) then result := '?' else result := wc2shitmap[LongWord(wc)];
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+function utf2win (const s: AnsiString): AnsiString;
+var
+  f, c: Integer;
+  ud: TUtf8DecoderFast;
+begin
+  for f := 1 to Length(s) do
+  begin
+    if (Byte(s[f]) > 127) then
+    begin
+      ud := TUtf8DecoderFast.Create(true);
+      result := '';
+      for c := 1 to Length(s) do
+      begin
+        if ud.decode(s[c]) then result += wchar2win(WideChar(ud.codepoint));
+      end;
+      exit;
+    end;
+  end;
+  result := s;
+end;
+
+
+function win2utf (const s: AnsiString): AnsiString;
+var
+  f, c: Integer;
+
+  function utf8Encode (code: Integer): AnsiString;
+  begin
+    if (code < 0) or (code > $10FFFF) then begin result := '?'; exit; end;
+    if (code <= $7f) then
+    begin
+      result := Char(code and $ff);
+    end
+    else if (code <= $7FF) then
+    begin
+      result := Char($C0 or (code shr 6));
+      result += Char($80 or (code and $3F));
+    end
+    else if (code <= $FFFF) then
+    begin
+      result := Char($E0 or (code shr 12));
+      result += Char($80 or ((code shr 6) and $3F));
+      result += Char($80 or (code and $3F));
+    end
+    else if (code <= $10FFFF) then
+    begin
+      result := Char($F0 or (code shr 18));
+      result += Char($80 or ((code shr 12) and $3F));
+      result += Char($80 or ((code shr 6) and $3F));
+      result += Char($80 or (code and $3F));
+    end
+    else
+    begin
+      result := '?';
+    end;
+  end;
+
+begin
+  for f := 1 to Length(s) do
+  begin
+    if (Byte(s[f]) > 127) then
+    begin
+      result := '';
+      for c := 1 to Length(s) do
+      begin
+        if (Byte(s[c]) < 128) then
+        begin
+          result += s[c];
+        end
+        else
+        begin
+          result += utf8Encode(cp1251[Byte(s[c])-128])
+        end;
+      end;
+      exit;
+    end;
+  end;
+  result := s;
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+function digitInBase (ch: AnsiChar; base: Integer): Integer;
+begin
+  result := -1;
+  if (base < 1) or (base > 36) then exit;
+  if (ch < '0') then exit;
+  if (base <= 10) then
+  begin
+    if (Integer(ch) >= 48+base) then exit;
+    result := Integer(ch)-48;
+  end
+  else
+  begin
+    if (ch >= '0') and (ch <= '9') then begin result := Integer(ch)-48; exit; end;
+    if (ch >= 'a') and (ch <= 'z') then Dec(ch, 32); // poor man's tolower()
+    if (ch < 'A') or (Integer(ch) >= 65+(base-10)) then exit;
+    result := Integer(ch)-65+10;
+  end;
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
+function quoteStr (const s: AnsiString): AnsiString;
+
+  function squote (const s: AnsiString): AnsiString;
+  var
+    f: Integer;
+  begin
+    result := '''';
+    for f := 1 to Length(s) do
+    begin
+      if (s[f] = '''') then result += '''';
+      result += s[f];
+    end;
+    result += '''';
+  end;
+
+  function dquote (const s: AnsiString): AnsiString;
+  var
+    f: Integer;
+    ch: AnsiChar;
+  begin
+    result := '"';
+    for f := 1 to Length(s) do
+    begin
+      ch := s[f];
+           if (ch = #0) then result += '\z'
+      else if (ch = #9) then result += '\t'
+      else if (ch = #10) then result += '\n'
+      else if (ch = #13) then result += '\r'
+      else if (ch = #27) then result += '\e'
+      else if (ch < ' ') or (ch = #127) then
+      begin
+        result += '\x';
+        result += LowerCase(IntToHex(Integer(ch), 2));
+      end
+      else if (ch = '"') or (ch = '\') then
+      begin
+        result += '\';
+        result += ch;
+      end
+      else
+      begin
+        result += ch;
+      end;
+    end;
+    result += '"';
+  end;
+
+var
+  needSingle: Boolean = false;
+  f: Integer;
+begin
+  for f := 1 to Length(s) do
+  begin
+    if (s[f] = '''') then begin needSingle := true; continue; end;
+    if (s[f] < ' ') or (s[f] = #127) then begin result := dquote(s); exit; end;
+  end;
+  if needSingle then result := squote(s) else result := ''''+s+'''';
+end;
+
+
+// ////////////////////////////////////////////////////////////////////////// //
 function hasWadExtension (fn: AnsiString): Boolean;
 begin
   fn := ExtractFileExt(fn);
@@ -596,7 +1039,7 @@ var
       while (len > 0) do
       begin
         if (len > 255) then slen := 255 else slen := Integer(len);
-        Move(b^, ss[1], len);
+        Move(b^, ss[1], slen);
         ss[0] := AnsiChar(slen);
         result += ss;
         b += slen;