Here's the second outstanding issue that Jason brought up for discussion. I've also attached the latest iMesh.h and iBase.h to this email from the repository - I believe those are up to date with all the other changes we've agreed to.. Lori -------- Original Message -------- Subject: iMesh_getAdjEntities (yet another proposed iMesh change) Date: Wed, 04 Feb 2009 11:33:26 -0600 From: Jason Kraftcheck <[email protected]> To: TSTT Interface <[email protected]> I think that we should either drop the iMesh_getAdjEntities function entirely, or if not, then remove the 'offset' argument. This function returns entities of the "requested" type, but not the entities of the "requestor" type/topology. For the "offset" information to be useful, one must also have the entities of the "requestor" type/topology in the appropriate order. Presumably to get that list the application must call iMesh_getEntities with the implicit assumption that the order will be correct. But if the application has the ordered "requestor" list, then it could just as easily call iMesh_getEntAdjArr to get the adjacent entities without needing to make any assumptions about functions returning entities in a consistent order. The only case where that might be inefficient is if one wanted something like the list of vertices adjacent to all of the entities in a set but didn't care about the entities. But for this use case, presumably the application would rather have a single list of unique vertex handles and would have no use for the offset list. - jason