mir/libreoffice
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Make BOM detection slightly more straightforward

Browse Source 
Without taking system endianness and current stream endianness
into account - just read and check single bytes.

Change-Id: I9273d8f403caad7adb5e11cecc04e326919dad1f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126595
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
This commit is contained in:
Mike Kaganski
2021-12-10 12:01:19 +02:00
parent 16376cae68
commit 1f1ce06a18
1 changed files with 36 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

70
tools/source/stream/stream.cxx
View File

@@ -718,52 +718,54 @@ void SvStream::StartReadingUnicodeText( rtl_TextEncoding eReadBomCharSet )
eReadBomCharSet == RTL_TEXTENCODING_UTF8)) eReadBomCharSet == RTL_TEXTENCODING_UTF8))
return; // nothing to read return; // nothing to read
bool bTryUtf8 = false; const sal_uInt64 nOldPos = Tell();
sal_uInt16 nFlag(0); bool bGetBack = true;
sal_sSize nBack = sizeof(nFlag); unsigned char nFlag(0);
ReadUInt16( nFlag ); ReadUChar( nFlag );
switch ( nFlag ) switch ( nFlag )
{ {
case 0xfeff : case 0xfe: // UTF-16BE?
// native UTF-16
if ( eReadBomCharSet == RTL_TEXTENCODING_DONTKNOW ||
eReadBomCharSet == RTL_TEXTENCODING_UNICODE)
nBack = 0;
break;
case 0xfffe :
// swapped UTF-16
if ( eReadBomCharSet == RTL_TEXTENCODING_DONTKNOW || if ( eReadBomCharSet == RTL_TEXTENCODING_DONTKNOW ||
eReadBomCharSet == RTL_TEXTENCODING_UNICODE) eReadBomCharSet == RTL_TEXTENCODING_UNICODE)
{ {
SetEndian( m_nEndian == SvStreamEndian::BIG ? SvStreamEndian::LITTLE : SvStreamEndian::BIG ); ReadUChar(nFlag);
nBack = 0; if (nFlag == 0xff)
{
SetEndian(SvStreamEndian::BIG);
bGetBack = false;
}
} }
break; break;
case 0xefbb : case 0xff: // UTF-16LE?
if (m_nEndian == SvStreamEndian::BIG && if ( eReadBomCharSet == RTL_TEXTENCODING_DONTKNOW ||
(eReadBomCharSet == RTL_TEXTENCODING_DONTKNOW || eReadBomCharSet == RTL_TEXTENCODING_UNICODE)
eReadBomCharSet == RTL_TEXTENCODING_UTF8)) {
bTryUtf8 = true; ReadUChar(nFlag);
if (nFlag == 0xfe)
{
SetEndian(SvStreamEndian::LITTLE);
bGetBack = false;
}
}
break; break;
case 0xbbef : case 0xef: // UTF-8?
if (m_nEndian == SvStreamEndian::LITTLE && if ( eReadBomCharSet == RTL_TEXTENCODING_DONTKNOW ||
(eReadBomCharSet == RTL_TEXTENCODING_DONTKNOW || eReadBomCharSet == RTL_TEXTENCODING_UTF8)
eReadBomCharSet == RTL_TEXTENCODING_UTF8)) {
bTryUtf8 = true; ReadUChar(nFlag);
if (nFlag == 0xbb)
{
ReadUChar(nFlag);
if (nFlag == 0xbf)
bGetBack = false; // it is UTF-8
}
}
break; break;
default: default:
; // nothing ; // nothing
} }
if (bTryUtf8) if (bGetBack)
{ Seek(nOldPos); // no BOM, pure data
unsigned char nChar(0);
nBack += sizeof(nChar);
ReadUChar( nChar );
if (nChar == 0xbf)
nBack = 0; // it is UTF-8
}
if (nBack)
SeekRel( -nBack ); // no BOM, pure data
} }
sal_uInt64 SvStream::SeekRel(sal_Int64 const nPos) sal_uInt64 SvStream::SeekRel(sal_Int64 const nPos)