On Wed, Apr 21, 2010 at 11:25:48AM -0600, Eric Blake wrote:
On 04/21/2010 10:01 AM, Chris Lalancette wrote:
+static void qemuDomainDefNamespaceFree(void *nsdata) +{ + qemuDomainCmdlineDefPtr cmd = nsdata; + int i;
s/int/unsigned int/
+static int qemuDomainDefNamespaceParse(xmlDocPtr xml, + xmlNodePtr root, + xmlXPathContextPtr ctxt, + void **data) +{ + qemuDomainCmdlineDefPtr cmd = NULL; + xmlNsPtr ns; + xmlNodePtr *nodes = NULL; + int n, i;
n must be signed, but can i be unsigned?
+ +static int qemuDomainDefNamespaceFormatXML(virBufferPtr buf, + void *nsdata) +{ + qemuDomainCmdlineDefPtr cmd = nsdata; + int i;
unsigned
+ + if (cmd->num_extra || cmd->num_env) + virBufferAddLit(buf, " <qemu:commandline>\n"); + for (i = 0; i < cmd->num_extra; i++) + virBufferVSprintf(buf, " <qemu:arg>%s</qemu:arg>\n", cmd->extra[i]);
Is there any chance that cmd->extra[i] might contain content that needs escaping before it is valid as XML?
Yes, this should use virBufferEscapeString() instead.
+ for (i = 0; i < cmd->num_env; i++) { + virBufferVSprintf(buf, " <qemu:env name='%s'", cmd->env_name[i]); + if (cmd->env_value[i]) + virBufferVSprintf(buf, " value='%s'", cmd->env_value[i]);
Likewise for env_value[i]? (I'm assuming that env_name[i] is immune, since it was parsed using virXPathString("string(./@name)",), which should have rejected strings not valid as environment variable names.)
Yep, likewise here Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://deltacloud.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|