commit 03d6ee382340dc1db57d7bbc269de0e102ed1a7a
author Serhiy Storchaka <storchaka@gmail.com> Mon Jul 06 13:58:33 2015 +0300
committer Serhiy Storchaka <storchaka@gmail.com> Mon Jul 06 13:58:33 2015 +0300
tree ea512d959a8e61b90e1a1a13ed49f37928bd703f
parent 0357268d96d4ff3546cfd89f594a5630a3adf747

Issue #18684: Fixed reading out of the buffer in the re module.

Modules/sre_lib.h

diff --git a/Modules/sre_lib.h b/Modules/sre_lib.h
index 5c6c5a5..1049de4 100644
--- a/Modules/sre_lib.h
+++ b/Modules/sre_lib.h
@@ -30,7 +30,7 @@
                 SRE_IS_LINEBREAK((int) ptr[-1]));
 
     case SRE_AT_END:
-        return (((void*) (ptr+1) == state->end &&
+        return (((SRE_CHAR *)state->end - ptr == 1 &&
                  SRE_IS_LINEBREAK((int) ptr[0])) ||
                 ((void*) ptr == state->end));
 
@@ -1093,9 +1093,9 @@
             /* <ASSERT> <skip> <back> <pattern> */
             TRACE(("|%p|%p|ASSERT %d\n", ctx->pattern,
                    ctx->ptr, ctx->pattern[1]));
-            state->ptr = ctx->ptr - ctx->pattern[1];
-            if (state->ptr < state->beginning)
+            if (ctx->ptr - (SRE_CHAR *)state->beginning < (Py_ssize_t)ctx->pattern[1])
                 RETURN_FAILURE;
+            state->ptr = ctx->ptr - ctx->pattern[1];
             DO_JUMP0(JUMP_ASSERT, jump_assert, ctx->pattern+2);
             RETURN_ON_FAILURE(ret);
             ctx->pattern += ctx->pattern[0];
@@ -1106,8 +1106,8 @@
             /* <ASSERT_NOT> <skip> <back> <pattern> */
             TRACE(("|%p|%p|ASSERT_NOT %d\n", ctx->pattern,
                    ctx->ptr, ctx->pattern[1]));
-            state->ptr = ctx->ptr - ctx->pattern[1];
-            if (state->ptr >= state->beginning) {
+            if (ctx->ptr - (SRE_CHAR *)state->beginning >= (Py_ssize_t)ctx->pattern[1]) {
+                state->ptr = ctx->ptr - ctx->pattern[1];
                 DO_JUMP0(JUMP_ASSERT_NOT, jump_assert_not, ctx->pattern+2);
                 if (ret) {
                     RETURN_ON_ERROR(ret);
@@ -1199,12 +1199,20 @@
     SRE_CODE* overlap = NULL;
     int flags = 0;
 
+    if (ptr > end)
+        return 0;
+
     if (pattern[0] == SRE_OP_INFO) {
         /* optimization info block */
         /* <INFO> <1=skip> <2=flags> <3=min> <4=max> <5=prefix info>  */
 
         flags = pattern[2];
 
+        if (pattern[3] && end - ptr < (Py_ssize_t)pattern[3]) {
+            TRACE(("reject (got %u chars, need %u)\n",
+                   (unsigned int)(end - ptr), pattern[3]));
+            return 0;
+        }
         if (pattern[3] > 1) {
             /* adjust end point (but make sure we leave at least one
                character in there, so literal search will work) */
@@ -1322,15 +1330,18 @@
                 break;
             ptr++;
         }
-    } else
+    } else {
         /* general case */
-        while (ptr <= end) {
+        assert(ptr <= end);
+        while (1) {
             TRACE(("|%p|%p|SEARCH\n", pattern, ptr));
-            state->start = state->ptr = ptr++;
+            state->start = state->ptr = ptr;
             status = SRE(match)(state, pattern, 0);
-            if (status != 0)
+            if (status != 0 || ptr >= end)
                 break;
+            ptr++;
         }
+    }
 
     return status;
 }