mirror of
https://gitlab.com/freepascal.org/lazarus/lazarus.git
synced 2025-08-10 19:56:21 +02:00
Converter: change the character encoding of source files to UTF-8
git-svn-id: trunk@31023 -
This commit is contained in:
juha
parent
dfe60001a1
commit
ba43bbf08f
@ -452,9 +452,6 @@ begin
|
||||
end;
|
||||
|
||||
function TConvertDelphiUnit.CopyAndLoadFile: TModalResult;
|
||||
var
|
||||
// CurEncoding: String;
|
||||
Changed: Boolean;
|
||||
begin
|
||||
IDEMessagesWindow.AddMsg(Format(lisConvDelphiConvertingFile,
|
||||
[fOrigUnitFilename]), '', -1);
|
||||
@ -475,12 +472,11 @@ begin
|
||||
[lbfCheckIfText,lbfUpdateFromDisk],true);
|
||||
if Result<>mrOk then exit;
|
||||
// Change encoding to UTF-8
|
||||
{ CurEncoding:=GuessEncoding(fPascalBuffer.Source); //fPascalBuffer.DiskEncoding;
|
||||
if CurEncoding<>EncodingUTF8 then begin
|
||||
fPascalBuffer.Source:=ConvertEncoding(fPascalBuffer.Source, CurEncoding, EncodingUTF8);
|
||||
fPascalBuffer.DiskEncoding:=EncodingUTF8;
|
||||
fPascalBuffer.MemEncoding:=EncodingUTF8;
|
||||
end; }
|
||||
if fPascalBuffer.DiskEncoding<>EncodingUTF8 then begin
|
||||
IDEMessagesWindow.AddMsg(Format(lisConvDelphiChangedEncodingToUTF8,
|
||||
[fPascalBuffer.DiskEncoding]), '', -1);
|
||||
fPascalBuffer.DiskEncoding:=EncodingUTF8; // Takes effect when buffer is saved.
|
||||
end;
|
||||
// Create a shared link for codetools.
|
||||
Assert(fCTLink=Nil, 'fCTLink should be Nil in CopyAndLoadFile');
|
||||
fCTLink:=TCodeToolLink.Create(fPascalBuffer);
|
||||
@ -499,8 +495,6 @@ var
|
||||
LfmFilename: string; // Lazarus .LFM file name.
|
||||
DFMConverter: TDFMConverter;
|
||||
TempLFMBuffer: TCodeBuffer;
|
||||
// CurEncoding: String;
|
||||
Changed: Boolean;
|
||||
begin
|
||||
Result:=mrOK;
|
||||
fLFMBuffer:=nil;
|
||||
@ -538,23 +532,13 @@ begin
|
||||
DFMConverter.Free;
|
||||
end;
|
||||
// Change encoding to UTF-8
|
||||
if fSettings.FixEncoding then begin
|
||||
Result:=LoadCodeBuffer(TempLFMBuffer,LfmFilename,
|
||||
[lbfCheckIfText,lbfUpdateFromDisk],true);
|
||||
// Note: EnUnicode is meant to be a temporary solution.
|
||||
// LCL has other functions for char encoding.
|
||||
TempLFMBuffer.Source:=EnUnicode(TempLFMBuffer.Source, Changed);
|
||||
if Changed then
|
||||
IDEMessagesWindow.AddMsg(lisConvDelphiChangedEncodingToUTF8, '', -1);
|
||||
// TempLFMBuffer.SaveToFile(ChangeFileExt(TempLFMBuffer.Filename, '_utf8.lfm'));
|
||||
Result:=LoadCodeBuffer(TempLFMBuffer,LfmFilename,
|
||||
[lbfCheckIfText,lbfUpdateFromDisk],true);
|
||||
if TempLFMBuffer.DiskEncoding<>EncodingUTF8 then begin
|
||||
IDEMessagesWindow.AddMsg(Format(lisConvDelphiChangedEncodingToUTF8,
|
||||
[TempLFMBuffer.DiskEncoding]), '', -1);
|
||||
TempLFMBuffer.DiskEncoding:=EncodingUTF8;
|
||||
TempLFMBuffer.Save;
|
||||
{ CurEncoding:=GuessEncoding(TempLFMBuffer.Source);
|
||||
if CurEncoding<>EncodingUTF8 then begin
|
||||
ShowMessage('Encoding = ' + CurEncoding);
|
||||
TempLFMBuffer.Source:=ConvertEncoding(TempLFMBuffer.Source, CurEncoding, EncodingUTF8);
|
||||
TempLFMBuffer.DiskEncoding:=EncodingUTF8;
|
||||
TempLFMBuffer.MemEncoding:=EncodingUTF8;
|
||||
end; }
|
||||
end;
|
||||
// Read form file code in.
|
||||
if not fSettings.SameDfmFile then begin
|
||||
|
||||
@ -129,98 +129,11 @@ type
|
||||
destructor Destroy; override;
|
||||
end;
|
||||
|
||||
function EnUnicode(const TS: UTF8String; var Changed: Boolean): UTF8String;
|
||||
|
||||
implementation
|
||||
|
||||
{$R *.lfm}
|
||||
|
||||
{*******************************************************************************
|
||||
Function UTFEnc(S:ansiString):WideChar;
|
||||
S - string like '#1234' or like '1234'.
|
||||
It process only first 4-5 symbols. (Some kind of protection)
|
||||
Result - One Unicode symbol.
|
||||
If S isn't an unicode function will return symbol #0000.
|
||||
*******************************************************************************}
|
||||
function UTFEnc(S: ansistring): WideChar;
|
||||
var X: word; //word - to be sure that it will return Unicode symbol. Not ASCII.
|
||||
begin
|
||||
if (S[1]='#')and(Length(S)=5)then
|
||||
X:=StrToIntDef(Copy(S,2,4),0)
|
||||
else
|
||||
X:=StrToIntDef(Copy(S,1,4),0);
|
||||
Result:=WideChar(X);
|
||||
end;
|
||||
|
||||
{*******************************************************************************
|
||||
function EnUnicode(TS:UTF8String):UTF8String;
|
||||
TS:UTF8String - Processing string like <<Caption = #1234#1235':'#1258>>
|
||||
Function converts it to string like <<Caption = 'АБ:В'>>
|
||||
It have some troubles with strings that contains several pairs of apostrophes.
|
||||
(<<Form1.Caption := 'String1 '+'String2'>> will converts into
|
||||
<<Form1.Caption := 'String1 +String2'>>
|
||||
*******************************************************************************}
|
||||
function EnUnicode(const TS: UTF8String; var Changed: Boolean): UTF8String;
|
||||
var
|
||||
i,
|
||||
LPA, //LPA = Left Position of Apostroph. First pos of ' in TS
|
||||
LPS, //LPS = Left Position of Sharp. First pos of # in TS
|
||||
RPA, //RPA = Right Position of Apostroph. Last pos of ' in TS
|
||||
RPS: integer; //RPS = Right Position of Sharp. Last pos of # in TS
|
||||
insideAp: Boolean;//inside of two Apostrophes.
|
||||
S,WS: ansistring; //S copying "as is"(for better speed). WS - converts symbol by symbol
|
||||
begin
|
||||
Changed:=False;
|
||||
S:='';
|
||||
insideAp:=false;
|
||||
RPS:=0; RPA:=0;
|
||||
for i:=1 to Length(TS) do begin //find Last positions of ' and #
|
||||
if TS[i]='#' then
|
||||
RPS:=i; //May be there is a spec function to do it
|
||||
if TS[i]='''' then
|
||||
RPA:=i; //but I didn't find it.
|
||||
end;
|
||||
RPS:=RPS+4; //Actually no need for position of #, but pos of last symbol of sequence #1234
|
||||
i:=1; //Now let's find FIRST pos of ' and #
|
||||
LPA:=Pos('''',TS); //If ' not found I must throw out processing of
|
||||
if LPA=0 then
|
||||
LPA:=Length(TS)+1; //any apostrophes in the TS.
|
||||
LPS:=Pos('#',TS); //Also for #
|
||||
if LPS=0 then LPS:=Length(TS)+1;
|
||||
if (LPA<LPS) AND (LPA<=Length(TS)) then begin //Now I must define position of first
|
||||
i:=LPA; //symbol either ' or #.
|
||||
end else if (LPS<LPA) AND (LPS<=Length(TS)) then begin
|
||||
i:=LPS;
|
||||
end;
|
||||
if (RPS<=LPS) OR (RPS<=4) OR (RPS>Length(TS)+4) then
|
||||
RPS:=0; //More hiding processing if ' or #
|
||||
if (RPA<=LPA) OR (RPA=0) OR (RPA>Length(TS)) then
|
||||
RPA:=0; // not found
|
||||
if (LPA<=Length(TS)) OR (LPS<=Length(TS)) then begin //if we've found ' or # or both - start process
|
||||
S:=LeftStr(TS,i-1); //first part of TS (before ' or #) copy "as is"
|
||||
WS:='';
|
||||
while (i<=Length(TS)) AND ((i<=RPA) OR (i<=RPS)) do begin //process
|
||||
if TS[i]='''' then begin //current symbol = '
|
||||
insideAp:=not insideAp; //part inside of '' will be copy as is
|
||||
end; //even it contains #1234 sequenses
|
||||
if (not insideAp) AND (TS[i]='#') then begin //part outside of ''
|
||||
WS:=WS+UTFEnc(Copy(TS,i,5)); //send to UTFEnc
|
||||
i:=i+4; //skipping nummbers
|
||||
end else
|
||||
if TS[i]<>'''' then //skipping apostrophes themselves
|
||||
WS:=WS+TS[i];
|
||||
i:=i+1;
|
||||
end;
|
||||
//Form1.Memo1.Lines.Add(S+'|'+WS+'|'+Copy(TS,i,Length(TS)));//It was an debug output
|
||||
S:=S+''''+WS+''''; //adding apostrophes around processed part
|
||||
S:=S+Copy(TS,i,Length(TS)); //adding rest of string as is
|
||||
Changed:=True;
|
||||
end {if (LPA<=Length(TS))OR(LPS<=Length(TS))}
|
||||
else
|
||||
S:=TS; //TS doesn't contain neither ' nor #. Copy as is.
|
||||
Result:=AnsiToUtf8(S); //Result must be an UTF8-string
|
||||
end;
|
||||
|
||||
function IsMissingType(LFMError: TLFMError): boolean;
|
||||
begin
|
||||
with LFMError do
|
||||
|
||||
@ -5139,7 +5139,7 @@ resourcestring
|
||||
lisConvDelphiConvertingUnitFiles = '*** Converting unit files ... ***';
|
||||
lisConvDelphiConvertingFile = '* Converting file %s *';
|
||||
lisConvDelphiFixingUsedUnits = '* Fixing used units for file %s *';
|
||||
lisConvDelphiChangedEncodingToUTF8 = 'Changed encoding to UTF-8';
|
||||
lisConvDelphiChangedEncodingToUTF8 = 'Changed encoding from %s to UTF-8';
|
||||
lisConvDelphiErrorCanTFindUnit = '%s(%s,%s) Error: Can''t find unit %s';
|
||||
lisConvDelphiAllSubDirsScanned = 'All sub-directories will be scanned for unit files';
|
||||
lisConvDelphiMissingIncludeFile = '%s(%s,%s) missing include file';
|
||||
|
||||
Loading…
Reference in New Issue
Block a user