That's all I need to hear. I'll see what kind of scaffolding I can come up with. I think that's a better way to go, my interests lie outside what is Leo core's responsibility, and I'd probably end up complaining about it anyway.
(-; I just didn't want to do work which would become obsolete when Leo replicated it in core. Thanks, Kent On Thu, Jul 10, 2014 at 9:41 AM, Edward K. Ream <[email protected]> wrote: > The recent discussions of SQLAlchemy, http://www.sqlalchemy.org/ and > camlistore, https://camlistore.org/ have been fascinating. Both look like > excellent projects that may have important benefits to Leo. > > Here, however, I'll explain why it looks to me like no db of any kind can > possibly solve the fundamental problem of reliably associating gnx's (or > equivalently uA's) with nodes in **foreign files**, that is, nodes created > by @auto, @org-mode or @vim-outline trees. > > The essence of the problem is easily stated. We assume (as part of the > meaning of @auto, @org-mode or @vim-outline trees) that people who do > **not** use Leo will modify such foreign files. If this were not so, we > could just use @file and have no problems at all recreating gnx's, clone > links and uA's. There seems to be **no way** to update any db (of whatever > kind, local or global, and regardless of the keys used) when non-Leonine > users modify the file. > > In particular, Fidel mentioned some scheme of "tracking" changes to nodes. > That will work when we Leo users modify the file, but how could such scheme > possibly track changes when non-Leonine users modify the file? > > Instead, it looks like we must accept the possibility that changes to a > foreign file will modify a node in ways that will break *any* possible > bookmark (or key) to the node. That can't be helped. The recent posts > about bookmark clones suggest workarounds, namely that bookmark node will > remain uncloned. Not exactly the end of the world. > > In short, I see no way to track changes to nodes made by people who do not > use Leo. Given that fundamental fact, it seems that bookmarks are the way > to go. We'll want to create flexible ways of connecting bookmarks to nodes, > but we must accept the fact that bookmark nodes might remain unconnected > (uncloned). > > Your comments, please. > > Edward > > -- > You received this message because you are subscribed to the Google Groups > "leo-editor" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/leo-editor. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
