| 1 | n/a | """ |
|---|
| 2 | n/a | Broken bytecode objects can easily crash the interpreter. |
|---|
| 3 | n/a | |
|---|
| 4 | n/a | This is not going to be fixed. It is generally agreed that there is no |
|---|
| 5 | n/a | point in writing a bytecode verifier and putting it in CPython just for |
|---|
| 6 | n/a | this. Moreover, a verifier is bound to accept only a subset of all safe |
|---|
| 7 | n/a | bytecodes, so it could lead to unnecessary breakage. |
|---|
| 8 | n/a | |
|---|
| 9 | n/a | For security purposes, "restricted" interpreters are not going to let |
|---|
| 10 | n/a | the user build or load random bytecodes anyway. Otherwise, this is a |
|---|
| 11 | n/a | "won't fix" case. |
|---|
| 12 | n/a | |
|---|
| 13 | n/a | """ |
|---|
| 14 | n/a | |
|---|
| 15 | n/a | import types |
|---|
| 16 | n/a | |
|---|
| 17 | n/a | co = types.CodeType(0, 0, 0, 0, 0, b'\x04\x71\x00\x00', |
|---|
| 18 | n/a | (), (), (), '', '', 1, b'') |
|---|
| 19 | n/a | exec(co) |
|---|
