summaryrefslogtreecommitdiff
path: root/src/cnode/cnode.cpp
diff options
context:
space:
mode:
Diffstat (limited to 'src/cnode/cnode.cpp')
-rw-r--r--src/cnode/cnode.cpp15
1 files changed, 13 insertions, 2 deletions
diff --git a/src/cnode/cnode.cpp b/src/cnode/cnode.cpp
index 279ee9a..70e14b8 100644
--- a/src/cnode/cnode.cpp
+++ b/src/cnode/cnode.cpp
@@ -62,8 +62,19 @@ CfgNode::CfgNode(Cpath& path_comps, char *name, char *val, char *comment,
_is_value = (getTmpl()->isValue() && !_is_leaf);
_is_multi = getTmpl()->isMulti();
- /* XXX given the current definition of "default", the concept of
- * "default" doesn't really apply to config files.
+ /* XXX given the current definition of "default" (i.e., the
+ * "post-bug 1219" definition), the concept of "default" doesn't
+ * really apply to config files. however, if in the future we
+ * do go back to the original, simpler definition of "default"
+ * (which IMO is the right thing to do), the "default handling"
+ * here and elsewhere in the backend library will need to be
+ * revamped.
+ *
+ * in fact, in that case pretty much the only place that need to
+ * worry about "default" is in the "output" (i.e., "show")
+ * processing, and even there the only thing that needs to be
+ * done is to compare the current value with the "default value"
+ * in the template.
*/
_is_default = false;
_is_deactivated = deact;