[svk-commit] r2851 - branches/path-resolver
nobody at bestpractical.com
nobody at bestpractical.com
Fri May 9 19:57:21 EDT 2008
Author: ruz
Date: Fri May 9 19:57:21 2008
New Revision: 2851
Modified:
branches/path-resolver/README.path-resolver
Log:
* update README notes
Modified: branches/path-resolver/README.path-resolver
==============================================================================
--- branches/path-resolver/README.path-resolver (original)
+++ branches/path-resolver/README.path-resolver Fri May 9 19:57:21 2008
@@ -16,22 +16,31 @@
add it with all its children without questions
2) For changes it's possible to use the resolver we already have.
+* implemented, but without (old, base, new) resolver, just add or skip
=head2 M1 TODO
+=head3 Bugs
+
+* t/22status-conflict.t fails and there is no way to fix it without
+ adjusting the way we distinguish tree conflicts vs. skips
+* fix t/smerge/path-resolver/change_in_deleted_file_co.t, it
+ has some issues I don't know how treat them
+* eol style is not handled when we return back a file
+
+=head3 Features
+
* resolver help
-* when you're deciding if you want to add dir w/o children then
- you usually want to list all new nodes in the dir, something like
- > a)add all, o)nly this, ..., l)ist [a]: l
- X/file
- X/dir/file
- ...
+* l)ist action for add_directory
+* dt - diff thiers action for a file change
+* c)at action for add_file
=head1 SIDENOTES
* if I resolve a conflict then I do want svk to commit after all even
if overall merge is empty. I don't want to resolve things again and
this is not only about path resolving.
+** done
=head1 CASES
More information about the svk-commit
mailing list