Skip to content

LLM inference in C/C++

C++ ggml

common/peg : suppress incomplete escape sequences (#26780)

Aldehir Rojas48d22e295e2b
diff --git a/common/peg-parser.cpp b/common/peg-parser.cpp
index ef290ed7c0..4a4be7cf78 100644
--- a/common/peg-parser.cpp
+++ b/common/peg-parser.cpp
@@ -570,23 +570,34 @@ struct parser_executor {
}
static common_peg_parse_result handle_escape_sequence(common_peg_parse_context & ctx, size_t start, size_t & pos, const char delimiter) {
+ auto save = pos;
+
++pos; // consume '\'
if (pos >= ctx.input.size()) {
if (!ctx.is_lenient()) {
return common_peg_parse_result(COMMON_PEG_PARSE_RESULT_FAIL, start);
}
+ pos = save; // suppress unmatched '\'
return common_peg_parse_result(COMMON_PEG_PARSE_RESULT_NEED_MORE_INPUT, start, pos);
}
char c = ctx.input[pos];
+
if (c == delimiter || c == '\\' || c == '/' || c == 'b' || c == 'f' || c == 'n' || c == 'r' || c == 't') {
++pos;
return common_peg_parse_result(COMMON_PEG_PARSE_RESULT_SUCCESS, start, pos);
- } else if (c == 'u') {
- return handle_unicode_escape(ctx, start, pos);
- } else {
- return common_peg_parse_result(COMMON_PEG_PARSE_RESULT_FAIL, start);
}
+
+ if (c == 'u') {
+ auto result = handle_unicode_escape(ctx, start, pos);
+ if (result.need_more_input()) {
+ pos = save; // suppress incomplete sequence
+ return common_peg_parse_result(COMMON_PEG_PARSE_RESULT_NEED_MORE_INPUT, start, pos);
+ }
+ return result;
+ }
+
+ return common_peg_parse_result(COMMON_PEG_PARSE_RESULT_FAIL, start);
}
static common_peg_parse_result handle_unicode_escape(common_peg_parse_context & ctx, size_t start, size_t & pos) {
diff --git a/tests/peg-parser/test-json-parser.cpp b/tests/peg-parser/test-json-parser.cpp
index 5dd00115ce..ec7c2e668f 100644
--- a/tests/peg-parser/test-json-parser.cpp
+++ b/tests/peg-parser/test-json-parser.cpp
@@ -77,6 +77,30 @@ void test_json_parser(testing &t) {
t.assert_equal("result_is_need_more_input", true, result.need_more_input());
});
+ // Test need_more_input() parsing - incomplete escape sequence in a string value
+ t.test("need_more_input() parsing - incomplete escape sequence", [](testing &t) {
+ auto json = build_peg_parser([](common_peg_parser_builder & p) { return p.json(); });
+
+ std::vector<std::string> inputs {
+ R"({"text": "hello\)", // dangling backslash
+ R"({"text": "hello\u)", // incomplete unicode escape sequence
+ R"({"text": "hello\u00)",
+ };
+
+ for (const auto & input : inputs) {
+ t.test(input, [&](testing &t) {
+ common_peg_parse_context ctx(input, COMMON_PEG_PARSE_FLAG_LENIENT);
+
+ auto result = json.parse(ctx);
+
+ t.assert_equal("result_is_need_more_input", true, result.need_more_input());
+
+ // the incomplete escape sequence is not part of the partial value
+ t.assert_equal("result_end", input.find('\\'), result.end);
+ });
+ }
+ });
+
t.test("object member", [](testing &t) {
auto parser = build_peg_parser([](common_peg_parser_builder & p) {
return p.json_member("name", "\"" + p.chars("[a-z]") + "\"");