Rev 5886: Comment why we can't trivially optimize the same-path case. in http://bazaar.launchpad.net/~jameinel/bzr/2.4-set-parent-trees-delta-282941

[John Arbash Meinel]

At http://bazaar.launchpad.net/~jameinel/bzr/2.4-set-parent-trees-delta-282941

------------------------------------------------------------
revno: 5886
revision-id: john at arbash-meinel.com-20110519151519-0p4f1yrn1co5ol1g
parent: john at arbash-meinel.com-20110519150825-sp71o82yxs356qoi
committer: John Arbash Meinel <john at arbash-meinel.com>
branch nick: 2.4-set-parent-trees-delta-282941
timestamp: Thu 2011-05-19 17:15:19 +0200
message:
  Comment why we can't trivially optimize the same-path case.
-------------- next part --------------
=== modified file 'bzrlib/dirstate.py'
--- a/bzrlib/dirstate.py	2011-05-19 15:08:25 +0000
+++ b/bzrlib/dirstate.py	2011-05-19 15:15:19 +0000
@@ -1543,6 +1543,13 @@
                 deletes.append((old_path_utf8, None, file_id, None, True))
             elif (old_path, new_path) == root_only:
                 # change things in-place
+                # Note: the case of a parent directory changing its file_id
+                #       tends to break optimizations here, because officially
+                #       the file has actually been moved, it just happens to
+                #       end up at the same path. If we can figure out how to
+                #       handle that case, we can avoid a lot of add+delete
+                #       pairs for objects that stay put.
+                # elif old_path == new_path:
                 changes.append((old_path_utf8, new_path_utf8, file_id,
                                 inv_to_entry(inv_entry)))
             else: