diff --git a/src/libnetdata/line_splitter/fuzzer.sh b/src/libnetdata/line_splitter/fuzzer.sh
new file mode 100755
index 0000000000..1fad49ba70
--- /dev/null
+++ b/src/libnetdata/line_splitter/fuzzer.sh
@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+
+set -exu -o pipefail
+
+CFLAGS="-Wall -Wextra -O2 -g -fsanitize=fuzzer,undefined,address"
+
+re2c -b re2c_line_splitter.re.c -o re2c_line_splitter.c
+clang $CFLAGS -c re2c_line_splitter.c -o re2c_line_splitter.o
+clang++ $CFLAGS -c lines_splitter_pluginsd_fuzzer.cc -o lines_splitter_pluginsd_fuzzer.o
+clang++ $CFLAGS re2c_line_splitter.o lines_splitter_pluginsd_fuzzer.o -o pluginsd_line_splitter_fuzzer
+
+mkdir -p /tmp/corpus
+echo 'simple word test' > /tmp/corpus/simple.txt
+echo '"quoted string" test' > /tmp/corpus/quoted.txt
+echo "'single quoted' test" > /tmp/corpus/single_quoted.txt
+echo 'word1=word2 word3="quoted value"' > /tmp/corpus/mixed.txt
+./pluginsd_line_splitter_fuzzer -workers=12 -jobs=16 /tmp/corpus
diff --git a/src/libnetdata/line_splitter/lines_splitter_pluginsd_fuzzer.cc b/src/libnetdata/line_splitter/lines_splitter_pluginsd_fuzzer.cc
new file mode 100644
index 0000000000..229b28a9b9
--- /dev/null
+++ b/src/libnetdata/line_splitter/lines_splitter_pluginsd_fuzzer.cc
@@ -0,0 +1,49 @@
+#include <cstddef>
+#include <cstdint>
+#include <cstring>
+#include <vector>
+#include <cassert>
+
+// Function declaration
+extern "C" size_t quoted_strings_splitter_pluginsd_re2c(char *start, char **words, size_t max_words);
+
+
+extern "C" int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size)
+{
+ if (Size == 0)
+ return 0;
+
+ const size_t MAX_WORDS = 5;
+
+ std::vector<char> buffer(Size + 1);
+ std::memcpy(buffer.data(), Data, Size);
+ buffer[Size] = '\0'; // Ensure null-termination
+
+ // Create array to store word pointers
+ std::vector<char *> words(MAX_WORDS, nullptr);
+
+ // Run the splitter
+ size_t word_count = quoted_strings_splitter_pluginsd_re2c(buffer.data(), words.data(), MAX_WORDS);
+
+ // Basic invariant checks
+ assert(word_count <= MAX_WORDS && "Returned more words than max_words");
+
+ // Verify all returned word pointers are within our buffer
+ for (size_t i = 0; i < word_count; i++) {
+ if (words[i]) {
+ assert(
+ words[i] >= buffer.data() && words[i] < (buffer.data() + buffer.size()) &&
+ "Word pointer outside buffer bounds");
+
+ // Verify null-termination of each word
+ assert(strlen(words[i]) < buffer.size() && "Word not properly null-terminated");
+ }
+ }
+
+ // Verify remaining array elements are null
+ for (size_t i = word_count; i < MAX_WORDS; i++) {
+ assert(words[i] == nullptr && "Non-null pointer beyond word_count");
+ }
+
+ return 0;
+}
diff --git a/src/libnetdata/line_splitter/re2c_line_splitter.c b/src/libnetdata/line_splitter/re2c_line_splitter.c
index ae6f66624c..2452550ffa 100644
--- a/src/libnetdata/line_splitter/re2c_line_splitter.c
+++ b/src/libnetdata/line_splitter/re2c_line_splitter.c
@@ -1,8 +1,8 @@
-/* Generated by re2c 3.0 on Tue Nov 26 12:06:58 2024 */
-#line 1 "src/libnetdata/line_splitter/re2c_line_splitter.re.c"
-#line 4 "src/libnetdata/line_splitter/re2c_line_splitter.c"
-#define YYMAXFILL 1
-#line 1 "src/libnetdata/line_splitter/re2c_line_splitter.re.c"
+/* Generated by re2c 3.0 on Tue Dec 3 14:14:31 2024 */
+#line 1 "re2c_line_splitter.re.c"
+#line 4 "re2c_line_splitter.c"
+#define YYMAXFILL 2
+#line 1 "re2c_line_splitter.re.c"
#include <stddef.h>
@@ -10,15 +10,16 @@
size_t quoted_strings_splitter_pluginsd_re2c(char *start, char **words, size_t max_words)
{
size_t count = 0;
+ char *YYMARKER = NULL;
- const char *YYCURSOR = start;
+ char *YYCURSOR = start;
for (;;) {
-#line 18 "src/libnetdata/line_splitter/re2c_line_splitter.c"
+#line 19 "re2c_line_splitter.c"
{
char yych;
static const unsigned char yybm[] = {
- 192, 208, 208, 208, 208, 208, 208, 208,
+ 0, 208, 208, 208, 208, 208, 208, 208,
208, 224, 224, 224, 224, 224, 208, 208,
208, 208, 208, 208, 208, 208, 208, 208,
208, 208, 208, 208, 208, 208, 208, 208,
@@ -58,56 +59,75 @@ size_t quoted_strings_splitter_pluginsd_re2c(char *start, char **words, size_t m
if (yych <= 0x00) goto yy1;
if (yych <= ' ') goto yy3;
if (yych <= '"') goto yy4;
- if (yych <= '\'') goto yy5;
+ if (yych <= '\'') goto yy6;
goto yy3;
yy1:
++YYCURSOR;
-#line 23 "src/libnetdata/line_splitter/re2c_line_splitter.re.c"
+#line 26 "re2c_line_splitter.re.c"
{
+ if (count < max_words)
+ words[count] = NULL;
+
return count;
}
-#line 70 "src/libnetdata/line_splitter/re2c_line_splitter.c"
+#line 74 "re2c_line_splitter.c"
yy2:
yych = *++YYCURSOR;
if (yybm[0+yych] & 16) {
goto yy2;
}
-#line 35 "src/libnetdata/line_splitter/re2c_line_splitter.re.c"
+#line 41 "re2c_line_splitter.re.c"
{
if (count == max_words)
return count;
- start[YYCURSOR - start] = '\0';
+ if (*YYCURSOR != '\0')
+ *YYCURSOR++ = '\0';
+
words[count++] = start;
- start = (char *) ++YYCURSOR;
+ start = YYCURSOR;
continue;
}
-#line 86 "src/libnetdata/line_splitter/re2c_line_splitter.c"
+#line 92 "re2c_line_splitter.c"
yy3:
yych = *++YYCURSOR;
if (yybm[0+yych] & 32) {
goto yy3;
}
-#line 44 "src/libnetdata/line_splitter/re2c_line_splitter.re.c"
+#line 52 "re2c_line_splitter.re.c"
{
start = (char *) YYCURSOR;
continue;
}
-#line 97 "src/libnetdata/line_splitter/re2c_line_splitter.c"
+#line 103 "re2c_line_splitter.c"
yy4:
- yych = *++YYCURSOR;
- if (yybm[0+yych] & 64) {
- goto yy4;
- }
- goto yy6;
+ yych = *(YYMARKER = ++YYCURSOR);
+ if (yych >= 0x01) goto yy8;
yy5:
- yych = *++YYCURSOR;
- if (yybm[0+yych] & 128) {
- goto yy5;
- }
+#line 21 "re2c_line_splitter.re.c"
+ {
+ if (count < max_words)
+ words[count] = NULL;
+ return count;
+ }
+#line 114 "re2c_line_splitter.c"
yy6:
+ yych = *(YYMARKER = ++YYCURSOR);
+ if (yych <= 0x00) goto yy5;
+ goto yy12;
+yy7:
+ yych = *++YYCURSOR;
+yy8:
+ if (yybm[0+yych] & 64) {
+ goto yy7;
+ }
+ if (yych >= 0x01) goto yy10;
+yy9:
+ YYCURSOR = YYMARKER;
+ goto yy5;
+yy10:
++YYCURSOR;
-#line 26 "src/libnetdata/line_splitter/re2c_line_splitter.re.c"
+#line 32 "re2c_line_splitter.re.c"
{
if (count == max_words)
return count;
@@ -117,9 +137,17 @@ yy6:
words[count++] = start;
continue;
}
-#line 121 "src/libnetdata/line_splitter/re2c_line_splitter.c"
+#line 141 "re2c_line_splitter.c"
+yy11:
+ yych = *++YYCURSOR;
+yy12:
+ if (yybm[0+yych] & 128) {
+ goto yy11;
+ }
+ if (yych <= 0x00) goto yy9;
+ goto yy10;
}
-#line 48 "src/libnetdata/line_splitter/re2c_line_splitter.re.c"
+#line 56 "re2c_line_splitter.re.c"
}
}
diff --git a/src/libnetdata/line_splitter/re2c_line_splitter.re.c b/src/libnetdata/line_splitter/re2c_line_splitter.re.c
index d5159ea4a2..62571cb887 100644
--- a/src/libnetdata/line_splitter/re2c_line_splitter.re.c
+++ b/src/libnetdata/line_splitter/re2c_line_splitter.re.c
@@ -5,15 +5,16 @@
size_t quoted_strings_splitter_pluginsd_re2c(char *start, char **words, size_t max_words)
{
size_t count = 0;
+ char *YYMARKER = NULL;
- const char *YYCURSOR = start;
+ char *YYCURSOR = start;
for (;;) {
/*!re2c
re2c:define:YYCTYPE = char;
re2c:yyfill:enable = 0;
- single_quotes_word = ["] [^"]* ["];
- double_quotes_word = ['] [^']* ['];
+ single_quotes_word = ["] [^"\x00]* ["];
+ double_quotes_word = ['] [^'\x00]* ['];
unquoted_word = [^= "'\t\n\v\f\r\x00]+;
whitespace = [= \t\n\v\f\r]+;
@@ -41,9 +42,11 @@ size_t quoted_strings_splitter_pluginsd_re2c(char *start, char **words, size_t m
if (count == max_words)
return count;
- start[YYCURSOR - start] = '\0';
+ if (*YYCURSOR != '\0')
+ *YYCURSOR++ = '\0';
+
words[count++] = start;
- start = (char *) ++YYCURSOR;
+ start = YYCURSOR;
continue;
}
whitespace {