4 The latest copy of this document is found at
6 http://kernel.org/git/?p=git/git.git;a=blob;hb=todo;f=TODO
12 - Immedately after 0.99.9, the backward compatible symbolic
13 links in $(bindir) will be removed.
15 git-ssh-push and git-ssh-pull pair is not going away if ever.
16 Each of these old-name commands continues to invoke its
17 old-name counterpart on the other end.
20 What to expect after 0.99.9
21 ===========================
23 This is written in a form of to-do list for me, so if I say
24 "accept patch", it means I do not currently plan to do that
25 myself. People interested in seeing it materialize please take
26 a hint. Also whatever I marked "Perhaps" do not have to happen
27 if ever -- only if somebody cares enough and submits a clean
34 * Help Jon Loeliger to find place in the documentation to place
37 * Accept patches from people who actually have done CVS
38 migration and update the cvs-migration documentation.
39 Link the documentation from the main git.txt page.
41 * Talk about using rsync just once at the beginning when
42 initializing a remote repository so that local packs do not
43 need to be expanded. I personally do not think we need tool
44 support for this (but see below about optimized cloning).
46 * Maybe update tutorial with a toy project that involves two or
49 * Update tutorial to cover setting up repository hooks to do
52 * Do we still have missing docs? If so accept patches to finish
55 * Accept patches to talk about "Whoops, it broke. What's
58 * Accept patches to make formatted tables in asciidoc to work
59 well in both html and man pages (see git-diff(1)).
65 * Git daemon, when deployed at kernel.org, might turn out to be
66 quite a burden, since it needs to generate customized packs
67 every time a new request comes in. It may be worthwhile to
68 precompute some packs for popular sets of heads downloaders
69 have and serve that, even if that could give more than the
70 client asks for in some cases. We will know about this soon
71 enough [DONE part to allow using precomputed pack, but not
72 dynamic caching part].
74 * Libification. There are many places "run once" mentality is
75 ingrained in the management of basic data structures, which
76 need to be fixed. [Matthias Urlichs is already working on
77 this: <pan.2005.10.03.20.48.52.132570@smurf.noris.de>; Post
80 * Maybe a pack optimizer.
82 Given a set of objects and a set of refs (probably a handful
83 branch heads and point release tags), find a set of packs to
84 allow reasonably minimum download for all of these classes of
85 people: (1) somebody cloning the repository from scratch, (2)
86 somebody who tends to follow the master branch head reasonably
87 closely, (3) somebody who tends to follow only the point
90 * Maybe an Emacs VC backend.
92 * 'git split-projects'? This requires updated 'git-rev-list' to
93 skip irrelevant commits.
94 Message-ID: <Pine.LNX.4.63.0509221617300.23242@iabervon.org>
96 * Look at libified GNU diff CVS seems to use, or libxdiff.
97 [Daniel has his own diff tool almost ready to start
98 integrating and testing; Post 1.0]
100 * Plug-in file-level merges [Post 1.0].
106 * Maybe look at Cogito and see if I can help Pasky to adjust to
107 the later core features? Zack Brown's "cg-seek leaving empty
108 directories" problem is a good example of this.
110 * Perhaps detect cloning request in upload-pack and cache the
111 result for next cloning request until any of our refs change.
113 * Encourage concrete proposals to commit log message templates
114 we discussed some time ago.
116 * Perhaps accept patches to do undo/redo.
118 * Perhaps accept patch to optionally allow '--fuzz' in
121 * Allow 'git apply' to accept GNU diff 2.7 output that forgets
122 to say '\No newline' if both input ends with incomplete
125 * Maybe grok PGP signed text/plain in applymbox as well.
127 * Enhance "git repack" to not always use --all; this would be
128 handy if the repository contains wagging heads like "pu" in
131 * Make rebase restartable; instead of skipping what cannot be
132 automatically forward ported, leave the conflicts in the work
133 tree, have the user resolve it, and then restart from where it
134 left off [mechanism mostly done].
136 * Output full path in the "git-rev-list --objects" output, not
137 just the basename, and see the improved clustering results in
138 better packing [Tried, but did not work out well].
140 * Updated git-changes-script Jeff Garzik needs [Inquiry for
141 external spec sent out with a quick hack. Will know if that
142 is what he needs hopefully soon].
148 * Stop installing the old-name symlinks [IN 0.99.9].
150 * 'git merge-projects'?
152 Subject: Re: Merges without bases
153 References: <1125004228.4110.20.camel@localhost.localdomain>
154 Date: Thu, 25 Aug 2005 15:26:36 -0700
155 Message-ID: <7vvf1tps9v.fsf@assigned-by-dhcp.cox.net>
157 * 'git lost-and-found'? Link dangling commits found by
158 fsck-objects under $GIT_DIR/refs/lost-found/. Then
159 show-branch or gitk can be used to find any lost commit. [A
160 feeler patch sent out. Very underwhelming response X-<.]
162 Do not name it /lost+found/; that would probably confuse
163 things that mistake it a mount point (not our code but
166 * Add simple globbing rules to git-show-branch so that I can
167 say 'git show-branch --heads "ko-*"' (ko-master, ko-pu, and
168 ko-rc are in refs/tags/).
170 * We would want test scripts for the relative directory path
171 stuff Linus has been working on. So far, the following
172 commands should be usable with relative directory paths:
182 * In a freashly created empty repository, `git fetch foo:bar`
183 works OK, but `git checkout bar` afterwards does not (missing