This is an automated email from the ASF dual-hosted git repository.
airborne12 pushed a commit to branch clucene
in repository https://gitbox.apache.org/repos/asf/doris-thirdparty.git
The following commit(s) were added to refs/heads/clucene by this push:
new 08d45e541fe [fix](inverted index) reject null bytes in term middle to
avoid stack overflow (#368)
08d45e541fe is described below
commit 08d45e541fef1e48dc4ab00d2d5fde9569b49ee3
Author: zzzxl <[email protected]>
AuthorDate: Wed Aug 26 21:25:45 2026 +0800
[fix](inverted index) reject null bytes in term middle to avoid stack
overflow (#368)
---
src/core/CLucene/index/SDocumentWriter.cpp | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)
diff --git a/src/core/CLucene/index/SDocumentWriter.cpp
b/src/core/CLucene/index/SDocumentWriter.cpp
index 39ea1720d05..b4ded79b029 100644
--- a/src/core/CLucene/index/SDocumentWriter.cpp
+++ b/src/core/CLucene/index/SDocumentWriter.cpp
@@ -504,6 +504,26 @@ void
SDocumentsWriter<T>::ThreadState::FieldData::addPosition(Token *token) {
code = (code * 31) + tokenText[upto++];
uint32_t hashPos = code & postingsHashMask;
+ if (upto < tokenTextLen) {
+ for (int32_t i = upto; i < tokenTextLen; i++) {
+ if (tokenText[i] != CLUCENE_END_OF_WORD) {
+ char error_msg[512];
+ if constexpr (std::is_same_v<T, char>) {
+ snprintf(error_msg, sizeof(error_msg),
+ "Inverted index does not support terms with null
byte (\\0) in the "
+ "middle. "
+ "Found null at position %d, but non-null
character at position %d. "
+ "Term prefix: '%.*s'.",
+ upto, i, std::min(upto, 50), tokenText);
+ } else {
+ snprintf(error_msg, sizeof(error_msg),
+ "Term contains null character in the middle at
position %d.", upto);
+ }
+ _CLTHROWA(CL_ERR_IllegalArgument, error_msg);
+ }
+ }
+ }
+
assert(!postingsCompacted);
// Locate Posting in hash
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
