Add new parser error code, E_OVERFLOW. This error is returned when

the number of children of a node exceeds the max possible value for
the short that is used to count them.  The Python runtime converts
this parser error into the SyntaxError "expression too long."

5 files changed, 28 insertions, 18 deletions

diff --git a/Include/errcode.h b/Include/errcode.h
index 5655782..f94c11c 100644
--- a/Include/errcode.h
+++ b/Include/errcode.h
@@ -52,6 +52,7 @@ PERFORMANCE OF THIS SOFTWARE.
 #define E_DONE		16	/* Parsing complete */
 #define E_ERROR		17	/* Execution error */
 #define E_INDENT	18	/* Invalid indentation detected */
+#define E_OVERFLOW      19      /* Node had too many children */
 
 #ifdef __cplusplus
 }
diff --git a/Include/node.h b/Include/node.h
index 7f30923..3a564f7 100644
--- a/Include/node.h
+++ b/Include/node.h
@@ -46,7 +46,7 @@ typedef struct _node {
 } node;
 
 extern DL_IMPORT(node *) PyNode_New Py_PROTO((int type));
-extern DL_IMPORT(node *) PyNode_AddChild Py_PROTO((node *n, int type, char *str, int lineno));
+extern DL_IMPORT(int) PyNode_AddChild Py_PROTO((node *n, int type, char *str, int lineno));
 extern DL_IMPORT(void) PyNode_Free Py_PROTO((node *n));
 
 /* Node access functions */
diff --git a/Parser/node.c b/Parser/node.c
index c2308fd..3f593c8 100644
--- a/Parser/node.c
+++ b/Parser/node.c
@@ -29,10 +29,13 @@ PERFORMANCE OF THIS SOFTWARE.
 
 ******************************************************************/
 
+#include <limits.h>
+
 /* Parse tree node implementation */
 
 #include "pgenheaders.h"
 #include "node.h"
+#include "errcode.h"
 
 node *
 PyNode_New(type)
@@ -52,7 +55,7 @@ PyNode_New(type)
 #define XXX 3 /* Node alignment factor to speed up realloc */
 #define XXXROUNDUP(n) ((n) == 1 ? 1 : ((n) + XXX - 1) / XXX * XXX)
 
-node *
+int
 PyNode_AddChild(n1, type, str, lineno)
 	register node *n1;
 	int type;
@@ -62,12 +65,14 @@ PyNode_AddChild(n1, type, str, lineno)
 	register int nch = n1->n_nchildren;
 	register int nch1 = nch+1;
 	register node *n;
+	if (nch == SHRT_MAX || nch < 0)
+		return E_OVERFLOW;
 	if (XXXROUNDUP(nch) < nch1) {
 		n = n1->n_child;
 		nch1 = XXXROUNDUP(nch1);
 		PyMem_RESIZE(n, node, nch1);
 		if (n == NULL)
-			return NULL;
+			return E_NOMEM;
 		n1->n_child = n;
 	}
 	n = &n1->n_child[n1->n_nchildren++];
@@ -76,7 +81,7 @@ PyNode_AddChild(n1, type, str, lineno)
 	n->n_lineno = lineno;
 	n->n_nchildren = 0;
 	n->n_child = NULL;
-	return n;
+	return 0;
 }
 
 /* Forward */
diff --git a/Parser/parser.c b/Parser/parser.c
index 3b75dbc..00299af 100644
--- a/Parser/parser.c
+++ b/Parser/parser.c
@@ -153,11 +153,11 @@ shift(s, type, str, newstate, lineno)
 	int newstate;
 	int lineno;
 {
+	int err;
 	assert(!s_empty(s));
-	if (PyNode_AddChild(s->s_top->s_parent, type, str, lineno) == NULL) {
-		fprintf(stderr, "shift: no mem in addchild\n");
-		return -1;
-	}
+	err = PyNode_AddChild(s->s_top->s_parent, type, str, lineno);
+	if (err)
+		return err;
 	s->s_top->s_state = newstate;
 	return 0;
 }
@@ -172,13 +172,13 @@ push(s, type, d, newstate, lineno)
 	int newstate;
 	int lineno;
 {
+	int err;
 	register node *n;
 	n = s->s_top->s_parent;
 	assert(!s_empty(s));
-	if (PyNode_AddChild(n, type, (char *)NULL, lineno) == NULL) {
-		fprintf(stderr, "push: no mem in addchild\n");
-		return -1;
-	}
+	err = PyNode_AddChild(n, type, (char *)NULL, lineno);
+	if (err)
+		return err;
 	s->s_top->s_state = newstate;
 	return s_push(s, d, CHILD(n, NCH(n)-1));
 }
@@ -233,6 +233,7 @@ PyParser_AddToken(ps, type, str, lineno)
 	int lineno;
 {
 	register int ilabel;
+	int err;
 	
 	D(printf("Token %s/'%s' ... ", _PyParser_TokenNames[type], str));
 	
@@ -260,20 +261,20 @@ PyParser_AddToken(ps, type, str, lineno)
 					int arrow = x & ((1<<7)-1);
 					dfa *d1 = PyGrammar_FindDFA(
 						ps->p_grammar, nt);
-					if (push(&ps->p_stack, nt, d1,
-						arrow, lineno) < 0) {
+					if ((err = push(&ps->p_stack, nt, d1,
+						arrow, lineno)) > 0) {
 						D(printf(" MemError: push\n"));
-						return E_NOMEM;
+						return err;
 					}
 					D(printf(" Push ...\n"));
 					continue;
 				}
 				
 				/* Shift the token */
-				if (shift(&ps->p_stack, type, str,
-						x, lineno) < 0) {
+				if ((err = shift(&ps->p_stack, type, str,
+						x, lineno)) > 0) {
 					D(printf(" MemError: shift.\n"));
-					return E_NOMEM;
+					return err;
 				}
 				D(printf(" Shift.\n"));
 				/* Pop while we are in an accept-only state */
diff --git a/Python/pythonrun.c b/Python/pythonrun.c
index 7760714..3a60e47 100644
--- a/Python/pythonrun.c
+++ b/Python/pythonrun.c
@@ -1033,6 +1033,9 @@ err_input(err)
 	case E_INDENT:
 		msg = "inconsistent use of tabs and spaces in indentation";
 		break;
+	case E_OVERFLOW:
+		msg = "expression too long";
+		break;
 	default:
 		fprintf(stderr, "error=%d\n", err->error);
 		msg = "unknown parsing error";