diff options
| author | Mike Kaganski <mike.kaganski@collabora.com> | 2021-12-11 12:53:26 +0300 |
|---|---|---|
| committer | Mike Kaganski <mike.kaganski@collabora.com> | 2021-12-11 12:21:57 +0100 |
| commit | 21154ea8c450f9f5568b32123d34a20e498a9290 (patch) | |
| tree | 2c4650bd7303153e32105524264666512657dfc3 /svtools | |
| parent | 692bc46b25db61176b4ced7b7beffeca7d55068e (diff) | |
tdf#146173: combine non-BMP characters' surrogates correctly
Change-Id: Ib3af1f9e461f133d2f5b09b9db4fb87c1ede0b9f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126658
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Diffstat (limited to 'svtools')
| -rw-r--r-- | svtools/source/svrtf/svparser.cxx | 20 |
1 files changed, 20 insertions, 0 deletions
diff --git a/svtools/source/svrtf/svparser.cxx b/svtools/source/svrtf/svparser.cxx index 0fec7a97097e..dd5068976ff7 100644 --- a/svtools/source/svrtf/svparser.cxx +++ b/svtools/source/svrtf/svparser.cxx @@ -247,7 +247,27 @@ sal_uInt32 SvParser<T>::GetNextChar() rInput.ReadUtf16(cUC); bErr = !rInput.good(); if( !bErr ) + { c = cUC; + if (rtl::isHighSurrogate(cUC)) + { + const sal_uInt64 nPos = rInput.Tell(); + rInput.ReadUtf16(cUC); + bErr = !rInput.good(); + if (!bErr) + { + if (rtl::isLowSurrogate(cUC)) + c = rtl::combineSurrogates(c, cUC); + else + rInput.Seek(nPos); // process lone high surrogate + } + else + { + bErr = false; // process lone high surrogate + rInput.Seek(nPos); // maybe step 1 byte back + } + } + } } else { |