On Thu, Mar 21, 2013 at 9:14 AM, <[email protected]> wrote: > On 2013/03/21 08:04:42, janek wrote: > >> So, you think that we shouldn't report any programming error in these >> alignment funcitons since an empty extent doesn't prevent us from >> doing our job i.e. aligning)? > > Well, with an empty extent it would appear our job of aligning > _self_ is already done before we even start.
yes, in a way. >> (i'm just asking to make sure that i understood you correctly) >> If so, i'll just delete all these error reports - fair enough for me. > > For me the problem only starts once one tries aligning a non-empty > construct to an empty one. For _that_ we need a coping strategy. Not quite. I mean, grob's parent must exist (even if its extent is empty), and if the extent is empty, we can just return 0 offset. > That strategy might in some cases warrant flagging an error, in other > cases it might warrant escalating to aligning on the empty construct's > parent instead. Hmm. Maybe, but no example comes to my mind. > I don't know the logic of the code and its uses, so this is basically my > unqualified gut feeling and nothing else. Flagging and reporting a > "problem" without any user-visible consequences does not appear helpful > to me. What is the user or programmer supposed to do? > > Whether my characterization is accurate, I don't know either. If it > isn't, function name and warning message could possibly be made more > informative. It is telling that (presumably after looking at the code at least once) you don't know what the logic of the code is. I find the naming quite confusing indeed. Please take a look at this comment with which i hope to explain the idea of the code and its uses (you can see an older and newer version here): https://codereview.appspot.com/7768043/diff2/19001:25001/lily/self-alignment-interface.cc maybe it'll help you. I'll also write some more complete explanation and send it to the list. best, janek _______________________________________________ lilypond-devel mailing list [email protected] https://lists.gnu.org/mailman/listinfo/lilypond-devel
