paweld/lazarus
1
0
Fork 0
mirror of https://gitlab.com/freepascal.org/lazarus/lazarus.git synced 2025-08-08 08:16:18 +02:00
Code Issues Packages Projects Releases Wiki Activity

LazUtf8: faster implementation Utf8EscapeControlChars based upon idea by Alexey Torgashin. Issue #39573.

Browse Source
This commit is contained in:
Bart 2022-01-19 15:38:43 +01:00
parent a80c5252e0
commit 242f0ac056
1 changed files with 38 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
components/lazutils/lazutf8.pas
View File

@ -2937,31 +2937,62 @@ const
'[CAN]', '[EM]' , '[SUB]', '[ESC]', '[FS]' , '[GS]' , '[RS]' , '[US]'); '[CAN]', '[EM]' , '[SUB]', '[ESC]', '[FS]' , '[GS]' , '[RS]' , '[US]');
var var
Ch: Char; Ch: Char;
i: Integer; i,ResLen: Integer;
SubLen: SizeInt;
const
MaxGrowFactor: array[TEscapeMode] of integer = (3, 4, 5, 5, 5);
begin begin
if FindInvalidUTF8Codepoint(PChar(S), Length(S)) <> -1 then if FindInvalidUTF8Codepoint(PChar(S), Length(S)) <> -1 then
begin begin
UTF8FixBroken(S); UTF8FixBroken(S);
end; end;
Result := ''; Result := '';
SetLength(Result, Length(S)*MaxGrowFactor[EscapeMode]);
ResLen := 0;
//a byte < 127 cannot be part of a multi-byte codepoint, so this is safe //a byte < 127 cannot be part of a multi-byte codepoint, so this is safe
for i := 1 to Length(S) do for i := 1 to Length(S) do
begin begin
Inc(ResLen);
Ch := S[i]; Ch := S[i];
if (Ch < #32) then if (Ch < #32) then
begin begin
case EscapeMode of case EscapeMode of
emPascal: Result := Result + PascalEscapeStrings[Ch]; emPascal:
emHexPascal: Result := Result + HexEscapePascalStrings[Ch]; begin
emHexC: Result := Result + HexEscapeCStrings[Ch]; Move(PascalEscapeStrings[Ch][1], Result[ResLen], 3);
emC: Result := Result + CEscapeStrings[Ch]; Inc(ResLen, 3-1);
emAsciiControlNames: Result := Result + AsciiControlStrings[Ch]; end;
emHexPascal:
begin
Move(HexEscapePascalStrings[Ch][1], Result[ResLen], 4);
Inc(ResLen, 4-1);
end;
emHexC:
begin
Move(HexEscapeCStrings[Ch][1], Result[ResLen], 5);
Inc(ResLen, 5-1);
end;
emC:
begin
SubLen := Length(CEscapeStrings[Ch]);
Move(CEscapeStrings[Ch][1], Result[ResLen], SubLen);
Inc(ResLen, SubLen-1);
end;
emAsciiControlNames:
begin
SubLen := Length(AsciiControlStrings[Ch]);
Move(AsciiControlStrings[Ch][1], Result[ResLen], SubLen);
Inc(ResLen, SubLen-1);
end;
end;//case end;//case
end end
else else
Result := Result + Ch; begin
Result[ResLen] := Ch;
end; end;
end; end;
SetLength(Result, ResLen);
end;
function UTF8StringOfChar(AUtf8Char: String; N: Integer): String; function UTF8StringOfChar(AUtf8Char: String; N: Integer): String;
var var