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
7 http://repo.or.cz/w/alt-git.git?a=blob;hb=todo;f=TODO
9 This is primarily meant for my personal reminder, but feel free
10 to pick an item from the list and work on it.
12 ----------------------------------------------------------------
17 * parse-remote.sh has POSIXLY incorrect shell construct.
19 Message-ID: <20070505080313.GA12170@gondor.apana.org.au>
21 * Use 'git diff' not 'git diff-tree' in merge and rebase
23 From: James Bowes <jbowes@dangerouslyinc.com>
24 Message-ID: <1178398134288-git-send-email-jbowes@dangerouslyinc.com>
28 From: Linus Torvalds <torvalds@linux-foundation.org>
29 Message-ID: <alpine.LFD.0.98.0705051524300.17381@woody.linux-foundation.org>
30 From: Junio C Hamano <junkio@cox.net>
31 Message-ID: <7vabwifl23.fsf@assigned-by-dhcp.cox.net>
33 * Pushing into a non-bare repository more gracefully.
35 When git-push is done to a non-bare repository and updates the
36 branch that is currently checked out, we currently do not do
39 From: Linus Torvalds <torvalds@linux-foundation.org>
40 Message-ID: <Pine.LNX.4.64.0704160931550.5473@woody.linux-foundation.org>
44 From: Franck Bui-Huu <vagabon.xyz@gmail.com>
45 Message-ID: <450EABD0.1040102@innova-card.com>
47 Repeated requests against git-daemon makes it stuck under --syslog
49 [jc: does not reproduce easily for me; has anybody seen it?]
52 * AsciiDoc 8 would break our documentation.
54 From: Stefan Richter <stefanr@s5r6.in-berlin.de>
55 Message-ID: <4523EC14.6070806@s5r6.in-berlin.de>
57 AsciiDoc 8 does not grok documents written for AsciiDoc 7 out of
62 * Delegate gitweb part to somebody else.
64 * Use gitattributes for more things.
66 - 'precious' files that are not tracked but not
67 build-products. Currently people seem to put them in
68 .gitignore, but that is not quite right, as .gitignore is
69 meant for ignoring things that can be lost (build products,
70 editor backup files). "git clean -x" and "git checkout" to
71 another branch that has a file where the current branch has a
72 directory could lose such 'precious' files.
74 - Customized "diff -p" markers per path (Johannes, on #git
77 I think it makes sense to give an extra parameter to xdiff
78 machinery to affect how "diff -p" markers are constructed (as
79 opposed to teach xdiff machinery to read gitattributes -- the
80 code does not have path information at that level). The
81 simplest interface would be to pass a regexp and have the
82 existing code always look for that regexp backwards. A more
83 complex one would involve a callback function, but I do not
84 know if that kind of complexity is worth it.
92 * Subproject Porcelain.
97 From: Junio C Hamano <junkio@cox.net>
98 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
100 * make merge-recursive and read-tree -u more robust when D/F
101 conflict is involved.
103 From: Junio C Hamano <junkio@cox.net>
104 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
106 * Use blame machinery to track a single file (not path) in a finer
109 From: Linus Torvalds <torvalds@linux-foundation.org>
110 Message-ID: <alpine.LFD.0.98.0704201554550.9964@woody.linux-foundation.org>
112 [jc: I have a fixed-up one parked in 'pu' and also outlined what
113 other things I think are needed in my response:
115 Message-ID: <7vwt06wqv8.fsf@assigned-by-dhcp.cox.net>
121 * add 'tree' entries to the index.
123 From: Junio C Hamano <junkio@cox.net>
124 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
126 * "pure" clones, that does not know about where it was cloned
127 from. Specifically, no [remote "origin"] in .git/config, nor
130 From: Junio C Hamano <junkio@cox.net>
131 Message-ID: <7vr6pac86g.fsf@assigned-by-dhcp.cox.net>
133 * upload-pack support to start fetching from any valid point on
134 the history, not just published refs. (Erik W. Biederman
135 <m164jc9ekx.fsf@ebiederm.dsl.xmission.com>)
137 * daemon --strict-symlink.
139 * Maybe grok PGP signed text/plain in applymbox as well.
141 * "git fetch" should be able to use foreign SCM import backends
142 such as svnimport and cvsimport.
144 * "git clone" should be a thin wrapper around init/remote/fetch/checkout
146 From: Junio C Hamano <junkio@cox.net>
147 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
152 * Change the "first line of commit message is special" rule to
153 "first paragraph" and then wrap it.
155 From: Junio C Hamano <junkio@cox.net>
156 Message-ID: <7vsla5pkug.fsf@assigned-by-dhcp.cox.net>
158 This is slightly related, but I have been wondering about the
159 interaction with "single-liner summary, empty line and then the
160 rest" convention and various commands in the log family.
162 Currently, --pretty=oneline and --pretty=email (hence format-patch)
163 take and use only the first line. I think we could change it to:
165 - take the first paragraph, where the definition of the first
166 paragraph is "skip all blank lines from the beginning, and
167 then grab everything up to the next empty line".
169 - replace all line breaks with a whitespace.
171 This change would not affect well-behaved commit messages that
172 adhere to the convention, as their first paragraph always
173 consist of a single line. On the other hand, people from
174 different culture can get frustrated by their commit message
175 chomped at the first linebreak in the middle of sentence right
176 now, which would be helped by this change.
178 Their Subject: and --pretty=oneline output would become very
179 long and unsightly, but their commit messages are already
180 ugly anyway, and such a change at least avoid the loss of
183 If we were to do this, Subject: line would most likely use
184 RFC2822 line folding at the places where line breaks were in the
185 original, but that goes without saying.
187 What do people think?
189 * Give --stdin to git-log, similar to git-rev-list
191 From: "Marco Costalba" <mcostalba@gmail.com>
192 Message-ID: <e5bfff550705110413q28aef3d8k3aeb0d342eeb2016@mail.gmail.com>
194 * Update the lockfile protocol so that closing and renaming are
195 done inside lockfile commit time. Some filesystems do not
196 like an open file renamed and then closed. Come up with a
197 patch and pass Alex for an Ack.
199 * Mbx (not mbox) support for git-mailsplit.
201 * git-clone fail .git/refs/foo (Yann Dirson <ydirson@altern.org>)
202 <20060610225040.GA7766@nowhere.earth>
204 * git-proxy should be spawned with sh -c 'command' $1 $2.
206 [jc: should it? -- deciding if it should may not be "trivial",
207 but if it turns out to be the right thing to do, the change
210 * Maybe a true git-proxy command that reads the first request
211 pkt-line, and redirects the request to its real destination.
213 * test scripts for the relative directory path stuff.