New submission from Will Seaver <will.sea...@gmail.com>:
In general, this is where my big problems began. My understanding of the
traversal algorithm was completely stymied by one thing: a lack of knowledge
about the __getitem__ method. If I understood this one piece of the puzzle
better, the whole thing would fall into place for me. I could glean some
information without understanding __getitem__, but without that knowledge, I
could not completely understand. With that out of the way, on to the line by
The analogy given on p. 62 was very helpful toward what understanding I do have
of traversal now. Thank you very much for that; it helped me to understand the
brute force nature of the protocol much better.
In the example url given on p. 62 (http://example.com:8080/a/b/c?foo=1, 'a' is
the root object, 'b' is the view name and 'c' is the context, correct?
p. 69: "Let's say that view lookup finds no matching view type. In this
circumstance, the repoze.bfg router returns the result of the not found view and
the request ends." I did not understand this at the time, but later reading
revealed the existence of the not_found view, and now I've got it.
object graph (p. 61: "It [traversal] is the act of finding a context and a view
name by walking over an object graph...") this has something to do with the
container node and leaf node (p. 65: "The object graph consists of container
nodes and leaf nodes.") Container nodes have children and leaf nodes do not,
title: chapter 7
topic: bfg book
Repoze Bugs <b...@bugs.repoze.org>
Repoze-dev mailing list