1 What to expect after 0.99.6
2 ===========================
4 This is written in a form of to-do list for me, so if I say
5 "accept patch", it means I do not currently plan to do that
6 myself. People interested in seeing it materialize please take
13 * Accept patches from people who actually have done CVS
14 migration and update the cvs-migration documentation.
15 Link the documentation from the main git.txt page.
17 * Accept patches from people who were hit by shiny blue bat to
18 update the SubmittingPatches [ONGOING].
20 * Talk about using rsync just once at the beginning when
21 initializing a remote repository so that local packs do not
22 need to be expanded. I personally do not think we need tool
23 support for this (but see below about optimized cloning).
25 * Maybe update tutorial with a toy project that involves two or
28 * Update tutorial to cover setting up repository hooks to do
31 * Accept patches to finish missing docs.
37 * Tony Luck reported an unfortunate glitch in the 3-way merge.
38 Encourage discussions to come up with a not-so-expensive way
39 to catch the kind of ambiguities that led to his misery.
40 [Daniel's patch looks quite promising, so is the one from
43 * HPA has two projects, klibc and klibc-kbuild, that have large
44 set of overlapping files in different paths (i.e. one has many
45 renames from the other). There currently is no way for git to
46 help keep these two trees in sync, merging criss-cross between
47 them. The merge logic should be able to take advantage of
48 rename/copy detection smarts git-diff-* family has. Linus,
49 me, and Daniel outlined a smarter merge strategy for this.
52 * To make it easier to experiment with different merge
53 strategies, make git-merge driver that will run merge backends
54 for the best merge [Outlined the idea; just do it].
56 * We might want to optimize cloning with GIT native transport
57 not to explode the pack, and store it in objects/pack instead.
58 We would need a tool to generate an idx file out of a pack
59 file for this. Also this itself may turn out to be a bad
60 idea, making the set of packs in repositories everybody has
61 different from each other.
63 * Maybe a pack optimizer. I am not convinced that packing all
64 objects into a single pack and removing all the existing panck
65 is the right way to go, since that would work against people
66 who already have those packs.
68 * Maybe an Emacs VC backend.
74 * Tool renames [STARTED].
76 * Have Daniel's read-tree graduate from "pu" after plugging leaks.
78 * Implement a merge backend using Daniel's read-tree.
80 * Accept Fredrik merge after renaming it (I want to name the
81 driver 'git merge'). Suggest where to place *.py stuff --
82 probably in $(share)/git-core/ and add Makefile entry for
85 * Encourage concrete proposals to commit log message templates
86 we discussed some time ago.
88 * Bug Martin for archimport script documentation.
90 * More portability. I dropped a SunOS patch on the floor by
93 * Accept patches to cause "read-tree -u" delete a directory when
96 * Perhaps accept patches to introduce the concept of "patch flow
97 expressed as ref mappings" Josef has been advocating about.
99 * Perhaps accept patches to do undo/redo.
101 * Maybe grok PGP signed text/plain in applymbox as well.
103 * Perhaps a tool to revert a single file to pre-modification
104 state? git-cat-file blob `git-ls-files | grep foo` >foo or
105 git-cat-file blob `git-ls-tree HEAD foo` >foo? What should
106 the command be called? git-revert is taken so is
109 * A tool to detect, show and prune already merged topic
112 * Enhance "git repack" to not always use --all; this would be
113 handy if the repository contains wagging heads like "pu" in
116 * Internally split the project into non-doc and doc parts; add
117 an extra root for the doc part and merge from it; move the
118 internal doc source to a separate repository, like the +Meta
119 repository; experiment if this results in a reasonable
120 workflow, and document it in howto form if it does.
122 * Option to limit rename detection for more than N paths.
128 * Perhaps "git branch -d" to delete a branch.
130 * We would want test scripts for the relative directory path
131 stuff Linus has been working on. So far, the following
132 commands should be usable with relative directory paths:
142 * In a freashly created empty repository, `git fetch foo:bar`
143 works OK, but `git checkout bar` afterwards does not (missing