GIT 0.99.9
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano committed
Oct 29, 2005 at 17:51 UTC
7b5f821775fbc6925c8aeff1b2643cd62ea19cff
2 files changed
+78
-135
HAVEDONE
+75
-27
@@ -1,48 +1,96 @@
1
-Done in 0.99.8
1
+Done in 0.99.9
2
==============
3
4
+Ports
5
+~~~~~
6
+
7
+* Cygwin port [HPA].
8
+
9
+* OpenBSD build [Merlyn and others].
10
+
11
+
12
+Fixes
13
+~~~~~
14
+
15
+* clone request over git native protocol from a repository with
16
+ too many refs did not work; this has been fixed.
17
+
18
+* git-daemon got safer for kernel.org use [HPA].
19
+
20
+* Extended SHA1 parser was not enforcing uniqueness for
21
+ abbreviated SHA1; this has been fixed.
22
+
23
+* http transport does not barf on funny characters in URL.
24
+
25
+* The ref naming restrictions have been formalized and the
26
+ coreish refuses to create funny refs; we still need to audit
27
+ importers. See git-check-ref-format(1).
28
+
29
+
30
New Features and Commands
31
~~~~~~~~~~~~~~~~~~~~~~~~~
32
7
-* 'git pull' now uses 'git-merge' instead of 'git-resolve', so you can
8
- specify alternative merge strategy to use on its command line.
33
+* .git/config file as a per-repository configuration mechanism,
34
+ and some commands understand it [Linus]. See
35
+ git(7).
36
10
-* 'git pull -s recursive' has been taught about renaming merges,
11
- which may deal with HPA's klibc vs klibc-kbuild situation better.
37
+* The core.filemode configuration item can be used to make us a
38
+ bit more FAT friendly. See git(7).
39
13
-* Symbolic refs -- instead of using symlinks to express .git/HEAD,
14
- you can have a regular file that stores a single line
15
- 'ref: refs/heads/master' in it.
40
+* The extended SHA1 notation acquired Peel-the-onion operator
41
+ ^{type} and ^{}. See git-rev-parse(1).
42
17
- git-update-ref is the preferred way to write into .git/HEAD,
18
- not "echo >.git/HEAD". git-symbolic-ref is the preferred way
19
- to check which underlying ref a symbolic ref .git/HEAD refers
20
- to, not "readlink .git/HEAD".
43
+* SVN importer [Matthias]. See git-svnimport(1).
44
22
-* A couple of new diff options (-l<num> and --name-status).
45
+* .git/objects/[0-9a-f]{2} directories are created on demand,
46
+ and removed when becomes empty after prune-packed [Linus].
47
24
-* Commit walker over http acquired more SSL options.
48
+* Filenames output from various commands without -z option are
49
+ quoted when they embed funny characters (TAB and LF) using
50
+ C-style quoting within double-quotes, to match the proposed
51
+ GNU diff/patch notation [me, but many people contributed in
52
+ the discussion].
53
26
-* 'git clone' checks out the working tree by default.
54
+* git-mv is expected to be a better replacement for git-rename.
55
+ While the latter has two parameter restriction, it acts more
56
+ like the regular 'mv' that can move multiple things to one
57
+ destinatino directory [Josef Weidendorfer].
58
59
+* git-checkout can take filenames to revert the changes to
60
+ them. See git-checkout(1)
61
29
-Fixes
30
-~~~~~
62
+* The new program git-am is a replacement for git-applymbox that
63
+ has saner command line options and a bit easier to use when a
64
+ patch does not apply cleanly.
65
+
66
+* git-ls-remote can show unwrapped onions using ^{} notation, to
67
+ help Cogito to track tags.
68
+
69
+* git-merge-recursive backend can merge unrelated projects.
70
+
71
+* git-clone over native transport leaves the result packed.
72
+
73
+* git-http-fetch issues multiple requests in parallel when
74
+ underlying cURL library supports it [Nick and Daniel].
75
32
-* Removed unused commands (diff-helper, rev-tree, and export).
76
+* git-fetch-pack and git-upload-pack try harder to figure out
77
+ better common commits [Johannes].
78
34
-* Platforms with only Python 2.3 installed can use recursive merge
35
- strategy.
79
+* git-read-tree -u removes a directory when it makes it empty.
80
37
-* Octopus documented.
81
+* git-diff-* records abbreviated SHA1 names of original and
82
+ resulting blob; this sometimes helps to apply otherwise an
83
+ unapplicable patch by falling back to 3-way merge.
84
39
-* Merge is more careful noticing potentially ambiguous situation.
85
+* git-format-patch now takes series of from..to rev ranges and
86
+ with '-m --stdout', writes them out to the standard output.
87
+ This can be piped to 'git-am' to implement cheaper
88
+ cherry-picking.
89
41
-* Git pull does not blindly do Octopus when Pull: lines in remotes
42
- file specifies more than one remote branches.
90
+* git-tag takes '-u' to specify the tag signer identity [Linus].
91
44
-* Commit walker got safer to use after interrupted downloads.
92
+* git-rev-list can take optional pathspecs to skip commits that
93
+ do not touch them (--dense) [Linus].
94
46
-* Commit walker over http can resume partial downloads.
95
+* Comes with new and improved gitk [Paulus and Linus].
96
48
-* More portability fixes for BSD and Solaris.
TODO
+3
-108
@@ -9,15 +9,15 @@ The GIT To-Do File
9
Tool Renames Plan
10
=================
11
12
- - In 0.99.9, the backward compatible symbolic links in
13
- $(bindir) will be removed.
12
+ - Immedately after 0.99.9, the backward compatible symbolic
13
+ links in $(bindir) will be removed.
14
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.
18
19
20
-What to expect after 0.99.8
20
+What to expect after 0.99.9
21
===========================
22
23
This is written in a form of to-do list for me, so if I say
@@ -31,12 +31,6 @@ patch, perhaps ;-).
31
Documentation
32
-------------
33
34
-* Document the ref naming restrictions [DONE].
35
-
36
-* David Ho's report suggests whatchanged documentation should
37
- mention -m as "commonly used options". Steal Linus'
38
- response [DONE].
39
-
34
* Help Jon Loeliger to find place in the documentation to place
35
his drawing.
36
@@ -68,14 +62,6 @@ Documentation
62
Technical (heavier)
63
-------------------
64
71
-* We might want to optimize cloning with GIT native transport
72
- not to explode the pack, and store it in objects/pack instead.
73
- We would need a tool to generate an idx file out of a pack
74
- file for this. Also this itself may turn out to be a bad
75
- idea, making the set of packs in repositories everybody has
76
- different from each other. [DONE; git-index-pack by Sergey,
77
- tweaking clone by me]
78
-
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
@@ -111,20 +97,12 @@ Technical (heavier)
97
[Daniel has his own diff tool almost ready to start
98
integrating and testing; Post 1.0]
99
114
-* Accept patches to fetch multiple objects by HTTP in parallel.
115
- [DONE]
116
-
100
* Plug-in file-level merges [Post 1.0].
101
119
-* Per-repository configuration mechanism [DONE by Linus].
120
-
102
103
Technical (milder)
104
------------------
105
125
-* Merlyn reports trouble with http fetch [Hopefully solved by
126
- Nick's updates].
127
-
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.
@@ -132,46 +110,9 @@ Technical (milder)
110
* Perhaps detect cloning request in upload-pack and cache the
111
result for next cloning request until any of our refs change.
112
135
-* Perhaps accept more "want"s in upload-pack and do something
136
- intelligent about it [Ongoing, by Johannes].
137
-
138
-* Perhaps send less "want"s from fetch-pack [DONE by JS].
139
-
140
-* Look at svn importer Smurf has. There is a small cvsimport
141
- update in his tree as well [DONE].
142
-
143
-* Decide the notation of "peeling the onion" operator, and
144
- implement it in sha1_name.c. Perhaps postfix "^{}" to mean
145
- "peel and expect anything", "^{blob}" to mean "peel and barf
146
- unless blob". The current "^0" becomes shorthand for
147
- "^{commit}". [DONE]
148
-
149
-* Quote the URL so that libcurl's metecharacter mechanism would
150
- not kick in [DONE].
151
-
152
-* Review the Makefile variables and exporting rules for them,
153
- while looking at prefix passing by Kai Ruemmler [DONE].
154
-
155
-* Review the 'sparse object database' change by Linus and move
156
- the first phase of it to the "master" branch [DONE].
157
-
158
-* Decide on mmap(). I am inclined to just stick to mmap
159
- replacement by Johannes Schindelin and do nothing else right
160
- now, except perhaps drop the writing-back support [DONE].
161
-
162
-* Revisit Santi's patch to move commit temorary files out of the
163
- working tree toplevel [DONE].
164
-
165
-* More generally, review the use of temporary files again.
166
- Assuming writable $GIT_DIR is more acceptable, but the
167
- working tree toplevel may not be in a rare usage pattern.
168
-
113
* Encourage concrete proposals to commit log message templates
114
we discussed some time ago.
115
172
-* Accept patches to cause "read-tree -u" delete a directory when
173
- it makes it empty [DONE].
174
-
116
* Perhaps accept patches to do undo/redo.
117
118
* Perhaps accept patch to optionally allow '--fuzz' in
@@ -181,22 +122,8 @@ Technical (milder)
122
to say '\No newline' if both input ends with incomplete
123
lines.
124
184
-* What to do with TABs and LFs in pathnames without breaking GNU
185
- patch? [DONE -- go with GNU patch extension proposed by Paul].
186
-
187
-* Adjust apply.c to proposed GNU patch extension that quotes \n
188
- and \t in C style, inside "". [DONE]
189
-
190
-* Adjust diff.c to the same. [DONE]
191
-
125
* Maybe grok PGP signed text/plain in applymbox as well.
126
194
-* Perhaps a tool to revert a single file to pre-modification
195
- state? People with BK background know this operation as
196
- 'clean'. 'git checkout [-f] ent [path...]' was suggested by
197
- Matthias Urlichs which sounds a natural extention to what the
198
- command currently does [DONE with a prodding by Linus].
199
-
127
* Enhance "git repack" to not always use --all; this would be
128
handy if the repository contains wagging heads like "pu" in
129
git.git repository.
@@ -214,42 +141,10 @@ Technical (milder)
141
external spec sent out with a quick hack. Will know if that
142
is what he needs hopefully soon].
143
217
-* An mechanism to ignore filesystem mode bits altogether [DONE].
218
-
144
145
Technical (trivial)
146
-------------------
147
223
-* Peter Hagervall's sparse fix [DONE].
224
-
225
-* Alex Riesen reported that hooks are in effect in tests.
226
- Should fix [DONE].
227
-
228
-* Disallow [\001-\040\177] byte values from ref names. Also we
229
- need to disallow ':' (used in refspec), '^' and '~' (postfix
230
- "peel the onion" operators), and '..' ("ref1..ref2" notation
231
- becomes ambiguous otherwise) [DONE].
232
-
233
-* Update fetch-pack and clone-pack to ignore funny refs from the
234
- other end, while making sure peek-remote does not discard them.
235
- [DONE]
236
-
237
-* Update upload-pack to send tag^{}. This would hopefully help
238
- Pasky's automated tag tracking, and also Martin's findtags.
239
- [DONE]
240
-
241
-* Adjust update-index to quoted --index-info [DONE].
242
-
243
-* Prepare apply.c changes for maint branch (0.99.8e) [DONE].
244
-
245
-* Perhaps show ^{commit}, ^{tree} instead of ^{} from ls-remote? [NO]
246
-
247
-* Re-adjust maint branch for the above if we did so. [NOT NEEDED]
248
-
249
-* short SHA1 naming is not enforcing uniqueness. Should fix [DONE].
250
-
251
-* 'git repack' can be DOSed. Should fix [DONE].
252
-
148
* Stop installing the old-name symlinks [IN 0.99.9].
149
150
* 'git merge-projects'?