On Fri, Aug 1, 2008 at 11:10 PM, David Lutterkort <[EMAIL PROTECTED]>wrote:
> On Fri, 2008-08-01 at 21:56 +0200, Raphaël Pinson wrote: > > > > Or to have both last_modified_node and last_created_node. > > last_modified_node would would refer to the input PATH in the cases of > > aug_set, aug_mv, aug_rm... last_created_node would refer to the exact > > PATH of the last created node. Is it possible for a command to create > > more than one node in a go (I don't see how it would happen)? > > > > > > So there would be something like > > > > > > Previous call m (the modfied path) ; c (the created path) > > aug_init - NULL; NULL > > aug_get - no influence on value > > of /augeas/tree/last_modified; no influence on value > > of /augeas/tree/last_created > > aug_set - PATH passed to aug_set, fully qualified ; > > resulting PATH, only if the path was created > > aug_insert - fully qualified path to newly created node (or > > empty - nothing was _modified_ so to say?) ; fully qualified path to > > newly created node > > aug_rm - PATH passed to aug_rm ; NULL > > aug_mv - PATH passed to aug_mv as DST argument ; PATH, > > only if the path was created > > aug_match - no influence on value > > of /augeas/tree/last_modified ; no influence on value > > of /augeas/tree/last_created > > aug_save - either leave /augeas/tree/last_modified > > unchanged or set to NULL ; either leave /augeas/tree/last_created > > unchanged or set to NULL > > aug_print - no influence on value > > of /augeas/tree/last_modified ; no influence on value > > of /augeas/tree/last_created > > > > > > Perhaps that makes the whole thing a bit too complicated though... > > I think so ... I'd much rather stick with a last_created node; > last_modified doesn't add anything the caller doesn't know already. > > The calls that might set something in last_created (aug_set, aug_mv, > aug_insert) should probably also set last_created to NULL if they didn't > create anything. Seems like the cleanest solution to me. > OK, so I guess what there is to code now is a set_last_created call, and the functions that will guess the path to send it to the function :) I was telling you earlier that I thought in tree_insert I could probably use seg_parent to get the parent node of the PATH given to tree_insert, and then go through all the children of this parent node until I find the newly inserted tree. Do you confirm this is how to do it? How do I get the path after I've done this? Is format_path(found_seg) correct? Raphael
_______________________________________________ augeas-devel mailing list [email protected] https://www.redhat.com/mailman/listinfo/augeas-devel
