1 What to expect after 0.99.5
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 * Link howto docs from the git.txt page. [DONE]
19 * Update the SubmittingPatches document to add MUA specific
20 hints on how to disable unwanted MIME and flowed-text by
21 collecting past list postings [DONE].
23 * Accept patches from people who were hit by shiny blue bat to
24 update the SubmittingPatches [ONGOING].
26 * Talk about using rsync just once at the beginning when
27 initializing a remote repository so that local packs do not
28 need to be expanded. I personally do not think we need tool
29 support for this (but see below about optimized cloning).
31 * Review the existing docs and see if the repository
32 organization needs to be clarified further [DONE].
34 * Update tutorial to cover shared repository style a bit more
37 * Maybe update tutorial with a toy project that involves two or
40 * Document the hooks [DONE].
42 * Update tutorial to cover setting up repository hooks to do
45 * Get help to properly use asciidoc in tutorial [Heck, ended up
46 learning it myself, DONE].
48 * Maybe justify and demonstrate an Octopus in the tutorial
51 * Add Octopus to the glossary [Thanks Johannes, DONE].
57 * Tony Luck reported an unfortunate glitch in the 3-way merge.
58 Encourage discussions to come up with a not-so-expensive way
59 to catch the kind of ambiguities that led to his misery.
60 [Daniel's patch looks quite promising.]
62 * HPA has two projects, klibc and klibc-kbuild, that have large
63 set of overlapping files in different paths (i.e. one has many
64 renames from the other). There currently is no way for git to
65 help keep these two trees in sync, merging criss-cross between
66 them. The merge logic should be able to take advantage of
67 rename/copy detection smarts git-diff-* family has [Discussion
70 * We might want to optimize cloning with GIT native transport
71 not to explode the pack, and store it in objects/pack instead.
72 We would need a tool to generate an idx file out of a pack
73 file for this. Also this itself may turn out to be a bad
74 idea, making the set of packs in repositories everybody has
75 different from each other.
77 * Maybe a pack optimizer. I am not convinced that packing all
78 objects into a single pack and removing all the existing panck
79 is the right way to go, since that would work against people
80 who already have those packs.
82 * Maybe an Emacs VC backend.
88 * When the branch head pointed by $GIT_DIR/HEAD changes while
89 the index file and working tree are looking the other way
90 (e.g. somebody pushed into your repository, or you ran "git
91 fetch" to update the ref your working tree is on), "git
92 checkout" without -f gets confused. Figure out a good way to
95 * We still have the same issue with "git fetch". Fetching into
96 the branch one is on _may_ need to do the same thing as
97 fetching into anonymous head and then do the resolve. At
98 least it needs a warning. [DONE]
100 * "git commit -m" should work for initial commits and perhaps
101 merge commits as well. Warning about merge is still a good
102 thing to do, while -m is useful in scripted non-interactive
103 use, so we need to be careful. [DONE]
105 * Encourage concrete proposals to commit log message templates
106 we discussed some time ago.
108 * Test apply.c patches from Robert Fitzsmons and have them
109 graduate to "master" branch [IN PROGRESS].
111 * Test read-tree reorganization patches from Daniel and have
112 them graduate to "master" branch [TESTED].
114 * Bug Ryan and work with him to update send-email easier to use.
115 [Ryan's patch pushed into "master". Will see how well it is
118 * Look at portability fixes from Jason Riedy
120 http://www.cs.berkeley.edu/~ejr/gits/git.git#portable
122 [Applied. Burned by one of them and backed out, but otherwise
125 * Accept patches to cause "read-tree -u" delete a directory when
128 * Perhaps accept patches to introduce the concept of "patch flow
129 expressed as ref mappings" Josef has been advocating about.
131 * Perhaps accept patches to do undo/redo.
133 * MIMEified applymbox to grok B and Q encodings in headers and
134 turn them into UTF-8; unwrap QP; explode multipart. [DONE]
136 * Maybe grok PGP signed text/plain in applymbox as well.
138 * Perhaps a tool to revert a single file to pre-modification
139 state? git-cat-file blob `git-ls-files | grep foo` >foo or
140 git-cat-file blob `git-ls-tree HEAD foo` >foo? What should
141 the command be called? git-revert is taken so is
144 * "git cherry-pick" that applies the patch an existing commit
145 introduces in its ancestry chain, possibly using the 3-way
146 merge machinery. [DONE]
148 * Update rebase using the cherry-pick command. [DONE]
150 * Redo "git revert" using 3-way merge machinery. [DONE]
152 * A tool to detect, show and prune already merged topic
155 * Set up an automated documentation rebuilding procedure at
156 kernel.org, using update hook mechanism. [DONE]
158 * Describe automated documentation rebuilding in a howto form
161 * Enhance "git repack" to not always use --all; this would be
162 handy if the repository contains wagging heads like "pu" in
165 * Accept and apply "git repack --all" patch, except the part
166 that removes the existing packs. [DONE]
168 * Internally split the project into non-doc and doc parts; add
169 an extra root for the doc part and merge from it; move the
170 internal doc source to a separate repository, like the +Meta
171 repository; experiment if this results in a reasonable
172 workflow, and document it in howto form if it does.
174 * Add names to all nodes in show-branch [DONE].
176 * Option to limit rename detection for more than N paths.
178 * Tool renames. [Discussion almost concluded, scheduled to
179 happen 0.99.7 with backward compatibility links, removing them
186 * Look at and merge Debian fixes from Tommi [Done].
188 * Perhaps "git branch -d" to delete a branch.
190 * Remove "git clone-dumb-http" [DONE].
192 * We would want test scripts for the relative directory path
193 stuff Linus has been working on. So far, the following
194 commands should be usable with relative directory paths:
205 Initial repository Gotchas
206 --------------------------
208 * `git fetch foo:bar` works OK, but `git checkout bar`
209 afterwards does not (missing `.git/HEAD`).
211 * `git commit -s` did not add signoff for initial commit [DONE].
213 * `git status` does not work well when missing `.git/HEAD` [DONE].
215 * `git log` either; git-log-script uses die which does not exist