diff options
| author | Michael Stahl <michael.stahl@allotropia.de> | 2021-11-16 14:28:15 +0100 |
|---|---|---|
| committer | Michael Stahl <michael.stahl@allotropia.de> | 2021-11-16 17:49:12 +0100 |
| commit | a7116b890ccd6dd1721413b4de6591a8057668ef (patch) | |
| tree | b2e98702f0f8772866638edab709f195b7ab50ba /svtools | |
| parent | 75a14a1e962793de1483ceea1bc36cc78144e386 (diff) | |
ofz#40766 svtools, sw: HTMLParser: really stop inserting control chars
35d248cab1f0d4800f72abb5cb6afb56f40d9083 forgot to fix one place where
control characters were in a presumed XML declaration.
Another place looks missing where comments are handled, but it's not
clear if these can be passed on to Writer.
Revert the previous fix from commit
b3325ef8cdfc2c82eec34e747106f75a9fccb7e4.
Change-Id: I11ad13de9122533626e512ce0384051e3e5bd97f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125306
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Diffstat (limited to 'svtools')
| -rw-r--r-- | svtools/source/svhtml/parhtml.cxx | 6 |
1 files changed, 5 insertions, 1 deletions
diff --git a/svtools/source/svhtml/parhtml.cxx b/svtools/source/svhtml/parhtml.cxx index d9f64c4541d9..e38afaa8ec2b 100644 --- a/svtools/source/svhtml/parhtml.cxx +++ b/svtools/source/svhtml/parhtml.cxx @@ -1058,6 +1058,7 @@ HtmlTokenId HTMLParser::GetNextToken_() sTmpBuffer.appendUtf32( nNextCh ); nNextCh = GetNextChar(); } while( '>' != nNextCh && '/' != nNextCh && !rtl::isAsciiWhiteSpace( nNextCh ) && + !linguistic::IsControlChar(nNextCh) && IsParserWorking() && !rInput.eof() ); if( !sTmpBuffer.isEmpty() ) @@ -1135,8 +1136,11 @@ HtmlTokenId HTMLParser::GetNextToken_() if( !bDone ) sTmpBuffer.appendUtf32(nNextCh); } - else + else if (!linguistic::IsControlChar(nNextCh) + || nNextCh == '\r' || nNextCh == '\n' || nNextCh == '\t') + { sTmpBuffer.appendUtf32(nNextCh); + } if( !bDone ) nNextCh = GetNextChar(); } |