On 04/05/2010 12:01 PM, Eric Blake wrote:
And I just got called away; rest of the review later after I eat.
Resuming my review...
@@ -203,9 +204,8 @@ static int interfaceListInterfaces(virConnectPtr conn, char **const names, int n const char *errmsg, *details; int errcode = ncf_error(driver->netcf, &errmsg, &details); interfaceReportError(netcf_to_vir_err(errcode), - "%s (netcf: %s - %s)", - _("failed to list host interfaces"), - errmsg, details ? details : ""); + _("failed to list host interfaces (netcf: %s - %s)"), + errmsg, details ? details : "");
If details is empty, this results in "msg (netcf: msg - )" which looks a bit awkward. Perhaps this construct should be written as: "%s (netcf: %s%s%s)", errmsg, details ? " - " : "", details ? details : "" But that is a separate issue from your patch.
if (nparams != 1) { - testError(VIR_ERR_INVALID_ARG, "nparams"); + testError(VIR_ERR_INVALID_ARG, "%s", _("nparams")); goto cleanup; } if (STRNEQ(params[0].field, "weight")) { - testError(VIR_ERR_INVALID_ARG, "field"); + testError(VIR_ERR_INVALID_ARG, "%s", _("field")); goto cleanup; } if (params[0].type != VIR_DOMAIN_SCHED_FIELD_UINT) { - testError(VIR_ERR_INVALID_ARG, "type"); + testError(VIR_ERR_INVALID_ARG, "%s", _("type")); goto cleanup;
Translators tend to hate strings like this. Single words are hard to translate without any context. I'm not sure how to improve it, though. So, I guess that means ACK to the content of this patch, after addressing my nit in part 1 of the review about splitting the cfg.mk changes into two commits. -- Eric Blake eblake@redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org