Re: [libvirt] [PATCH 04/10] parallels: Need to handle virJSONValueObjectGetString error
On 01/17/13 20:17, John Ferlan wrote:
---
src/parallels/parallels_driver.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/parallels/parallels_driver.c b/src/parallels/parallels_driver.c
index ea193af..1cf66ea 100644
--- a/src/parallels/parallels_driver.c
+++ b/src/parallels/parallels_driver.c
@@ -514,7 +514,10 @@ parallelsGetNetInfo(virDomainNetDefPtr net,
net->linkstate = VIR_DOMAIN_NET_INTERFACE_LINK_STATE_UP;
if (virJSONValueObjectHasKey(value, "state")) {
- tmp = virJSONValueObjectGetString(value, "state");
Hm calling this without checking should be safe as the
virJSONValueObjectHasKey basically guarantees successful return of a
value in virJSONValueObjectGetString.
+ if (!(tmp = virJSONValueObjectGetString(value,
"state"))) {
+ parallelsParseError();
+ goto error;
+ }
if STREQ(tmp, "disconnected")
net->linkstate = VIR_DOMAIN_NET_INTERFACE_LINK_STATE_DOWN;
}
I don't mind doing the change, but I think this is a false positive.
I give ACK to this patch, but it shouldn't be a problem to go without it.
Peter