Extensive unit tests for DiffNodes, Diffelements, and DiffLeafs data structure. The lack of unity in these three data structures is a bit gross, to be honest, but it might may not be a significant factor when I reach implementing the generic diff functions. The problem is that ideally these would look like the scheme structures:

Browse Source

(A B (C D E))

which is a nested list containing A and B items and a sublist of C D E.  Here there are only two classes: lists and everything else.  Right now we have three.  DiffNodes, which contain both atomic fields (A B) as well as the subnodes ((C D E)) here.  These a specific class for DiffLeaf, which is really just a pair mapping name=value.  And DiffElement contains a named item, since all objected in the hierarchy have a name.  It's just doesn't feel right to me right now.  Ultimately the problem is that you want the objects to be self-describing, so the DiffElement and DiffLeaf are a clean factoring the need for names in both the values and the nodes.

...

This commit is contained in:

Mark A. DePristo 2011-07-04 19:34:15 -04:00

parent 38740b0ff5

commit 527fbeaf3c

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Show Stats Download Patch File Download Diff File Expand all files Collapse all files

Diff Content Not Available