[3.11] gh-101006: Improve error handling when read marshal data (GH-101007) (GH-106227)

* EOFError no longer overrides other errors such as MemoryError or OSError at
  the start of the object.
* Raise more relevant error when the NULL object occurs as a code object
  component.
* Minimize an overhead of calling PyErr_Occurred().
(cherry picked from commit 8bf6904b229583033035d91a3800da5604dcaad4)

Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>

1 files changed, 71 insertions, 61 deletions

diff --git a/Python/marshal.c b/Python/marshal.c
index bf607cb..29f3bab 100644
--- a/Python/marshal.c
+++ b/Python/marshal.c
@@ -750,23 +750,28 @@ r_string(Py_ssize_t n, RFILE *p)
 static int
 r_byte(RFILE *p)
 {
-    int c = EOF;
-
     if (p->ptr != NULL) {
-        if (p->ptr < p->end)
-            c = (unsigned char) *p->ptr++;
-        return c;
+        if (p->ptr < p->end) {
+            return (unsigned char) *p->ptr++;
+        }
     }
-    if (!p->readable) {
+    else if (!p->readable) {
         assert(p->fp);
-        c = getc(p->fp);
+        int c = getc(p->fp);
+        if (c != EOF) {
+            return c;
+        }
     }
     else {
         const char *ptr = r_string(1, p);
-        if (ptr != NULL)
-            c = *(const unsigned char *) ptr;
+        if (ptr != NULL) {
+            return *(const unsigned char *) ptr;
+        }
+        return EOF;
     }
-    return c;
+    PyErr_SetString(PyExc_EOFError,
+                    "EOF read where not expected");
+    return EOF;
 }
 
 static int
@@ -827,10 +832,11 @@ r_PyLong(RFILE *p)
     digit d;
 
     n = r_long(p);
-    if (PyErr_Occurred())
-        return NULL;
     if (n == 0)
         return (PyObject *)_PyLong_New(0);
+    if (n == -1 && PyErr_Occurred()) {
+        return NULL;
+    }
     if (n < -SIZE32_MAX || n > SIZE32_MAX) {
         PyErr_SetString(PyExc_ValueError,
                        "bad marshal data (long size out of range)");
@@ -849,10 +855,6 @@ r_PyLong(RFILE *p)
         d = 0;
         for (j=0; j < PyLong_MARSHAL_RATIO; j++) {
             md = r_short(p);
-            if (PyErr_Occurred()) {
-                Py_DECREF(ob);
-                return NULL;
-            }
             if (md < 0 || md > PyLong_MARSHAL_BASE)
                 goto bad_digit;
             d += (digit)md << j*PyLong_MARSHAL_SHIFT;
@@ -863,10 +865,6 @@ r_PyLong(RFILE *p)
     d = 0;
     for (j=0; j < shorts_in_top_digit; j++) {
         md = r_short(p);
-        if (PyErr_Occurred()) {
-            Py_DECREF(ob);
-            return NULL;
-        }
         if (md < 0 || md > PyLong_MARSHAL_BASE)
             goto bad_digit;
         /* topmost marshal digit should be nonzero */
@@ -878,18 +876,17 @@ r_PyLong(RFILE *p)
         }
         d += (digit)md << j*PyLong_MARSHAL_SHIFT;
     }
-    if (PyErr_Occurred()) {
-        Py_DECREF(ob);
-        return NULL;
-    }
+    assert(!PyErr_Occurred());
     /* top digit should be nonzero, else the resulting PyLong won't be
        normalized */
     ob->ob_digit[size-1] = d;
     return (PyObject *)ob;
   bad_digit:
     Py_DECREF(ob);
-    PyErr_SetString(PyExc_ValueError,
-                    "bad marshal data (digit out of range in long)");
+    if (!PyErr_Occurred()) {
+        PyErr_SetString(PyExc_ValueError,
+                        "bad marshal data (digit out of range in long)");
+    }
     return NULL;
 }
 
@@ -912,8 +909,6 @@ r_float_str(RFILE *p)
     const char *ptr;
     n = r_byte(p);
     if (n == EOF) {
-        PyErr_SetString(PyExc_EOFError,
-            "EOF read where object expected");
         return -1;
     }
     ptr = r_string(n, p);
@@ -992,8 +987,10 @@ r_object(RFILE *p)
     PyObject *retval = NULL;
 
     if (code == EOF) {
-        PyErr_SetString(PyExc_EOFError,
-                        "EOF read where object expected");
+        if (PyErr_ExceptionMatches(PyExc_EOFError)) {
+            PyErr_SetString(PyExc_EOFError,
+                            "EOF read where object expected");
+        }
         return NULL;
     }
 
@@ -1045,7 +1042,10 @@ r_object(RFILE *p)
 
     case TYPE_INT:
         n = r_long(p);
-        retval = PyErr_Occurred() ? NULL : PyLong_FromLong(n);
+        if (n == -1 && PyErr_Occurred()) {
+            break;
+        }
+        retval = PyLong_FromLong(n);
         R_REF(retval);
         break;
 
@@ -1111,10 +1111,11 @@ r_object(RFILE *p)
         {
             const char *ptr;
             n = r_long(p);
-            if (PyErr_Occurred())
-                break;
             if (n < 0 || n > SIZE32_MAX) {
-                PyErr_SetString(PyExc_ValueError, "bad marshal data (bytes object size out of range)");
+                if (!PyErr_Occurred()) {
+                    PyErr_SetString(PyExc_ValueError,
+                        "bad marshal data (bytes object size out of range)");
+                }
                 break;
             }
             v = PyBytes_FromStringAndSize((char *)NULL, n);
@@ -1136,10 +1137,11 @@ r_object(RFILE *p)
         /* fall through */
     case TYPE_ASCII:
         n = r_long(p);
-        if (PyErr_Occurred())
-            break;
         if (n < 0 || n > SIZE32_MAX) {
-            PyErr_SetString(PyExc_ValueError, "bad marshal data (string size out of range)");
+            if (!PyErr_Occurred()) {
+                PyErr_SetString(PyExc_ValueError,
+                    "bad marshal data (string size out of range)");
+            }
             break;
         }
         goto _read_ascii;
@@ -1150,8 +1152,6 @@ r_object(RFILE *p)
     case TYPE_SHORT_ASCII:
         n = r_byte(p);
         if (n == EOF) {
-            PyErr_SetString(PyExc_EOFError,
-                "EOF read where object expected");
             break;
         }
     _read_ascii:
@@ -1178,10 +1178,11 @@ r_object(RFILE *p)
         const char *buffer;
 
         n = r_long(p);
-        if (PyErr_Occurred())
-            break;
         if (n < 0 || n > SIZE32_MAX) {
-            PyErr_SetString(PyExc_ValueError, "bad marshal data (string size out of range)");
+            if (!PyErr_Occurred()) {
+                PyErr_SetString(PyExc_ValueError,
+                    "bad marshal data (string size out of range)");
+            }
             break;
         }
         if (n != 0) {
@@ -1203,16 +1204,18 @@ r_object(RFILE *p)
         }
 
     case TYPE_SMALL_TUPLE:
-        n = (unsigned char) r_byte(p);
-        if (PyErr_Occurred())
+        n = r_byte(p);
+        if (n == EOF) {
             break;
+        }
         goto _read_tuple;
     case TYPE_TUPLE:
         n = r_long(p);
-        if (PyErr_Occurred())
-            break;
         if (n < 0 || n > SIZE32_MAX) {
-            PyErr_SetString(PyExc_ValueError, "bad marshal data (tuple size out of range)");
+            if (!PyErr_Occurred()) {
+                PyErr_SetString(PyExc_ValueError,
+                    "bad marshal data (tuple size out of range)");
+            }
             break;
         }
     _read_tuple:
@@ -1238,10 +1241,11 @@ r_object(RFILE *p)
 
     case TYPE_LIST:
         n = r_long(p);
-        if (PyErr_Occurred())
-            break;
         if (n < 0 || n > SIZE32_MAX) {
-            PyErr_SetString(PyExc_ValueError, "bad marshal data (list size out of range)");
+            if (!PyErr_Occurred()) {
+                PyErr_SetString(PyExc_ValueError,
+                    "bad marshal data (list size out of range)");
+            }
             break;
         }
         v = PyList_New(n);
@@ -1296,10 +1300,11 @@ r_object(RFILE *p)
     case TYPE_SET:
     case TYPE_FROZENSET:
         n = r_long(p);
-        if (PyErr_Occurred())
-            break;
         if (n < 0 || n > SIZE32_MAX) {
-            PyErr_SetString(PyExc_ValueError, "bad marshal data (set size out of range)");
+            if (!PyErr_Occurred()) {
+                PyErr_SetString(PyExc_ValueError,
+                    "bad marshal data (set size out of range)");
+            }
             break;
         }
 
@@ -1377,20 +1382,20 @@ r_object(RFILE *p)
 
             /* XXX ignore long->int overflows for now */
             argcount = (int)r_long(p);
-            if (PyErr_Occurred())
+            if (argcount == -1 && PyErr_Occurred())
                 goto code_error;
             posonlyargcount = (int)r_long(p);
-            if (PyErr_Occurred()) {
+            if (posonlyargcount == -1 && PyErr_Occurred()) {
                 goto code_error;
             }
             kwonlyargcount = (int)r_long(p);
-            if (PyErr_Occurred())
+            if (kwonlyargcount == -1 && PyErr_Occurred())
                 goto code_error;
             stacksize = (int)r_long(p);
-            if (PyErr_Occurred())
+            if (stacksize == -1 && PyErr_Occurred())
                 goto code_error;
             flags = (int)r_long(p);
-            if (PyErr_Occurred())
+            if (flags == -1 && PyErr_Occurred())
                 goto code_error;
             code = r_object(p);
             if (code == NULL)
@@ -1463,6 +1468,10 @@ r_object(RFILE *p)
             v = r_ref_insert(v, idx, flag, p);
 
           code_error:
+            if (v == NULL && !PyErr_Occurred()) {
+                PyErr_SetString(PyExc_TypeError,
+                    "NULL object in marshal data for code object");
+            }
             Py_XDECREF(code);
             Py_XDECREF(consts);
             Py_XDECREF(names);
@@ -1480,9 +1489,10 @@ r_object(RFILE *p)
     case TYPE_REF:
         n = r_long(p);
         if (n < 0 || n >= PyList_GET_SIZE(p->refs)) {
-            if (n == -1 && PyErr_Occurred())
-                break;
-            PyErr_SetString(PyExc_ValueError, "bad marshal data (invalid reference)");
+            if (!PyErr_Occurred()) {
+                PyErr_SetString(PyExc_ValueError,
+                    "bad marshal data (invalid reference)");
+            }
             break;
         }
         v = PyList_GET_ITEM(p->refs, n);