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 ----------------------------------------------------------------
19 From: Linus Torvalds <torvalds@linux-foundation.org>
20 Message-ID: <alpine.LFD.0.98.0705051524300.17381@woody.linux-foundation.org>
21 From: Junio C Hamano <junkio@cox.net>
22 Message-ID: <7vabwifl23.fsf@assigned-by-dhcp.cox.net>
24 * Pushing into a non-bare repository more gracefully.
26 When git-push is done to a non-bare repository and updates the
27 branch that is currently checked out, we currently do not do
30 From: Linus Torvalds <torvalds@linux-foundation.org>
31 Message-ID: <Pine.LNX.4.64.0704160931550.5473@woody.linux-foundation.org>
35 From: Franck Bui-Huu <vagabon.xyz@gmail.com>
36 Message-ID: <450EABD0.1040102@innova-card.com>
38 Repeated requests against git-daemon makes it stuck under --syslog
40 [jc: does not reproduce easily for me; has anybody seen it?]
42 * AsciiDoc 8 would break our documentation.
44 From: Stefan Richter <stefanr@s5r6.in-berlin.de>
45 Message-ID: <4523EC14.6070806@s5r6.in-berlin.de>
47 AsciiDoc 8 does not grok documents written for AsciiDoc 7 out of
52 * Delegate gitweb part to somebody else.
54 * Use gitattributes for more things.
56 - 'precious' files that are not tracked but not
57 build-products. Currently people seem to put them in
58 .gitignore, but that is not quite right, as .gitignore is
59 meant for ignoring things that can be lost (build products,
60 editor backup files). "git clean -x" and "git checkout" to
61 another branch that has a file where the current branch has a
62 directory could lose such 'precious' files.
64 - Customized "diff -p" markers per path (Johannes, on #git
67 I think it makes sense to give an extra parameter to xdiff
68 machinery to affect how "diff -p" markers are constructed (as
69 opposed to teach xdiff machinery to read gitattributes -- the
70 code does not have path information at that level). The
71 simplest interface would be to pass a regexp and have the
72 existing code always look for that regexp backwards. A more
73 complex one would involve a callback function, but I do not
74 know if that kind of complexity is worth it.
82 * Subproject Porcelain.
87 From: Junio C Hamano <junkio@cox.net>
88 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
90 * make merge-recursive and read-tree -u more robust when D/F
93 From: Junio C Hamano <junkio@cox.net>
94 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
96 * Use blame machinery to track a single file (not path) in a finer
99 From: Linus Torvalds <torvalds@linux-foundation.org>
100 Message-ID: <alpine.LFD.0.98.0704201554550.9964@woody.linux-foundation.org>
102 [jc: I have a fixed-up one parked in 'pu' and also outlined what
103 other things I think are needed in my response:
105 Message-ID: <7vwt06wqv8.fsf@assigned-by-dhcp.cox.net>
111 * add 'tree' entries to the index.
113 From: Junio C Hamano <junkio@cox.net>
114 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
116 * "pure" clones, that does not know about where it was cloned
117 from. Specifically, no [remote "origin"] in .git/config, nor
120 From: Junio C Hamano <junkio@cox.net>
121 Message-ID: <7vr6pac86g.fsf@assigned-by-dhcp.cox.net>
123 * upload-pack support to start fetching from any valid point on
124 the history, not just published refs. (Erik W. Biederman
125 <m164jc9ekx.fsf@ebiederm.dsl.xmission.com>)
127 * daemon --strict-symlink.
129 * Maybe grok PGP signed text/plain in applymbox as well.
131 * "git fetch" should be able to use foreign SCM import backends
132 such as svnimport and cvsimport.
134 * "git clone" should be a thin wrapper around init/remote/fetch/checkout
136 From: Junio C Hamano <junkio@cox.net>
137 Message-ID: <11793556371774-git-send-email-junkio@cox.net>
142 * Change the "first line of commit message is special" rule to
143 "first paragraph" and then wrap it.
145 From: Junio C Hamano <junkio@cox.net>
146 Message-ID: <7vsla5pkug.fsf@assigned-by-dhcp.cox.net>
148 This is slightly related, but I have been wondering about the
149 interaction with "single-liner summary, empty line and then the
150 rest" convention and various commands in the log family.
152 Currently, --pretty=oneline and --pretty=email (hence format-patch)
153 take and use only the first line. I think we could change it to:
155 - take the first paragraph, where the definition of the first
156 paragraph is "skip all blank lines from the beginning, and
157 then grab everything up to the next empty line".
159 - replace all line breaks with a whitespace.
161 This change would not affect well-behaved commit messages that
162 adhere to the convention, as their first paragraph always
163 consist of a single line. On the other hand, people from
164 different culture can get frustrated by their commit message
165 chomped at the first linebreak in the middle of sentence right
166 now, which would be helped by this change.
168 Their Subject: and --pretty=oneline output would become very
169 long and unsightly, but their commit messages are already
170 ugly anyway, and such a change at least avoid the loss of
173 If we were to do this, Subject: line would most likely use
174 RFC2822 line folding at the places where line breaks were in the
175 original, but that goes without saying.
177 What do people think?
179 * Give --stdin to git-log, similar to git-rev-list
181 From: "Marco Costalba" <mcostalba@gmail.com>
182 Message-ID: <e5bfff550705110413q28aef3d8k3aeb0d342eeb2016@mail.gmail.com>
184 * Update the lockfile protocol so that closing and renaming are
185 done inside lockfile commit time. Some filesystems do not
186 like an open file renamed and then closed. Come up with a
187 patch and pass Alex for an Ack.
189 * Mbx (not mbox) support for git-mailsplit.
191 * git-clone fail .git/refs/foo (Yann Dirson <ydirson@altern.org>)
192 <20060610225040.GA7766@nowhere.earth>
194 * git-proxy should be spawned with sh -c 'command' $1 $2.
196 [jc: should it? -- deciding if it should may not be "trivial",
197 but if it turns out to be the right thing to do, the change
200 * Maybe a true git-proxy command that reads the first request
201 pkt-line, and redirects the request to its real destination.
203 * test scripts for the relative directory path stuff.