/src/fuzz_ruby_parser.cpp
Line | Count | Source |
1 | | /* Copyright 2025 Google LLC |
2 | | Licensed under the Apache License, Version 2.0 (the "License"); |
3 | | you may not use this file except in compliance with the License. |
4 | | You may obtain a copy of the License at |
5 | | http://www.apache.org/licenses/LICENSE-2.0 |
6 | | Unless required by applicable law or agreed to in writing, software |
7 | | distributed under the License is distributed on an "AS IS" BASIS, |
8 | | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
9 | | See the License for the specific language governing permissions and |
10 | | limitations under the License. |
11 | | */ |
12 | | |
13 | | /* |
14 | | * Fuzzer for Ruby's parser (ruby_parser.c) |
15 | | * Tests parsing of Ruby source code with malformed/random input |
16 | | * to find bugs in the Ruby parser implementation. |
17 | | */ |
18 | | |
19 | | #include <stdint.h> |
20 | | #include <stdlib.h> |
21 | | #include <string> |
22 | | #include <fuzzer/FuzzedDataProvider.h> |
23 | | #include "ruby.h" |
24 | | #include "ruby/ruby.h" |
25 | | #include "ruby/encoding.h" |
26 | | |
27 | | /* Forward declarations for parser functions */ |
28 | | extern "C" { |
29 | | extern VALUE rb_parser_new(void); |
30 | | extern VALUE rb_parser_compile_string(VALUE vparser, const char *f, VALUE s, int line); |
31 | | extern VALUE ruby_verbose; |
32 | | } |
33 | | |
34 | | /* Silence stderr output during fuzzing */ |
35 | | static VALUE |
36 | | silenced_parse(VALUE parser_str_pair) |
37 | 16.8k | { |
38 | 16.8k | VALUE *args = (VALUE *)parser_str_pair; |
39 | 16.8k | VALUE parser = args[0]; |
40 | 16.8k | VALUE code_str = args[1]; |
41 | | |
42 | | /* Compile the string - this will parse it */ |
43 | 16.8k | return rb_parser_compile_string(parser, "(fuzz)", code_str, 1); |
44 | 16.8k | } |
45 | | |
46 | | extern "C" int |
47 | | LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) |
48 | 16.8k | { |
49 | 16.8k | static int initialized = 0; |
50 | 16.8k | int state; |
51 | | |
52 | 16.8k | if (!initialized) { |
53 | 1 | ruby_init(); |
54 | 1 | ruby_init_loadpath(); |
55 | 1 | initialized = 1; |
56 | | |
57 | | // Suppress Ruby warnings to avoid log noise |
58 | 1 | ruby_verbose = Qfalse; |
59 | 1 | } |
60 | | |
61 | 16.8k | if (size == 0) { |
62 | 0 | return 0; |
63 | 0 | } |
64 | | |
65 | | // Limit input size to avoid pathologically slow parsing |
66 | | // Ruby parser can be exponentially slow with deeply nested |
67 | | // structures resulting in timeouts. |
68 | 16.8k | if (size > 50000) { |
69 | 6 | return 0; |
70 | 6 | } |
71 | | // Use FuzzedDataProvider for structured data consumption |
72 | 16.8k | FuzzedDataProvider fdp(data, size); |
73 | | |
74 | | /* Create a Ruby string from the fuzz input */ |
75 | 16.8k | std::string code_data = fdp.ConsumeRemainingBytesAsString(); |
76 | 16.8k | VALUE code_str = rb_str_new(code_data.data(), code_data.size()); |
77 | | |
78 | | /* Create a new parser instance */ |
79 | 16.8k | VALUE parser = rb_parser_new(); |
80 | 16.8k | if (NIL_P(parser)) { |
81 | 0 | return 0; |
82 | 0 | } |
83 | | |
84 | | /* Prepare arguments for protected call */ |
85 | 16.8k | VALUE args[2]; |
86 | 16.8k | args[0] = parser; |
87 | 16.8k | args[1] = code_str; |
88 | | |
89 | | /* Parse the code with exception protection */ |
90 | 16.8k | rb_protect(silenced_parse, (VALUE)args, &state); |
91 | | |
92 | | /* If an exception occurred, clear it and continue */ |
93 | 16.8k | if (state) { |
94 | 398 | rb_set_errinfo(Qnil); |
95 | 398 | } |
96 | | |
97 | | /* Force GC to clean up */ |
98 | 16.8k | rb_gc_start(); |
99 | | |
100 | 16.8k | return 0; |
101 | 16.8k | } |