Typo
[rsync/rsync.git] / TODO
... / ...
CommitLineData
1-*- indented-text -*-
2
3URGENT ---------------------------------------------------------------
4
5
6IMPORTANT ------------------------------------------------------------
7
8Cross-test versions
9
10 Part of the regression suite should be making sure that we don't
11 break backwards compatibility: old clients vs new servers and so
12 on. Ideally we would test the cross product of versions.
13
14 It might be sufficient to test downloads from well-known public
15 rsync servers running different versions of rsync. This will give
16 some testing and also be the most common case for having different
17 versions and not being able to upgrade.
18
19use chroot
20
21 If the platform doesn't support it, then don't even try.
22
23 If running as non-root, then don't fail, just give a warning.
24 (There was a thread about this a while ago?)
25
26 http://lists.samba.org/pipermail/rsync/2001-August/thread.html
27 http://lists.samba.org/pipermail/rsync/2001-September/thread.html
28
29--files-from
30
31 Avoids traversal. Better option than a pile of --include statements
32 for people who want to generate the file list using a find(1)
33 command or a script.
34
35Performance
36
37 Traverse just one directory at a time. Tridge says it's possible.
38
39 Can possibly also be smarter about memory use while looking for hard
40 links by reducing the refcount as we find alternative names. In
41 fact at the moment the code seems to make a whole second copy of the
42 file list, which seems unnecessary.
43
44IPv6
45
46 Implement suggestions from http://www.kame.net/newsletter/19980604/
47 and ftp://ftp.iij.ad.jp/pub/RFC/rfc2553.txt
48
49 If a host has multiple addresses, then listen try to connect to all
50 in order until we get through. (getaddrinfo may return multiple
51 addresses.) This is kind of implemented already.
52
53 Possibly also when starting as a server we may need to listen on
54 multiple passive addresses. This might be a bit harder, because we
55 may need to select on all of them. Hm.
56
57 Define a syntax for IPv6 literal addresses. Since they include
58 colons, they tend to break most naming systems, including ours.
59 Based on the HTTP IPv6 syntax, I think we should use
60
61 rsync://[::1]/foo/bar
62 [::1]::bar
63
64 which should just take a small change to the parser code.
65
66Errors
67
68 If we hang or get SIGINT, then explain where we were up to. Perhaps
69 have a static buffer that contains the current function name, or
70 some kind of description of what we were trying to do. This is a
71 little easier on people than needing to run strace/truss.
72
73 "The dungeon collapses! You are killed." Rather than "unexpected
74 eof" give a message that is more detailed if possible and also more
75 helpful.
76
77File attributes
78
79 Device major/minor numbers should be at least 32 bits each. See
80 http://lists.samba.org/pipermail/rsync/2001-November/005357.html
81
82 Transfer ACLs. Need to think of a standard representation.
83 Probably better not to even try to convert between NT and POSIX.
84 Possibly can share some code with Samba.
85
86Empty directories
87
88 With the current common --include '*/' --exclude '*' pattern, people
89 can end up with many empty directories. We might avoid this by
90 lazily creating such directories.
91
92zlib
93
94 Perhaps don't use our own zlib. Will we actually be incompatible,
95 or just be slightly less efficient?
96
97logging
98
99 Perhaps flush stdout after each filename, so that people trying to
100 monitor progress in a log file can do so more easily. See
101 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=48108
102
103rsyncd over ssh
104
105 There are already some patches to do this.
106
107PLATFORMS ------------------------------------------------------------
108
109Win32
110
111 Don't detach, because this messes up --srvany.
112
113 http://sources.redhat.com/ml/cygwin/2001-08/msg00234.html
114
115 According to "Effective TCP/IP Programming" (??) close() on a socket
116 has incorrect behaviour on Windows -- it sends a RST packet to the
117 other side, which gives a "connection reset by peer" error. On that
118 platform we should probably do shutdown() instead. However, on Unix
119 we are correct to call close(), because shutdown() discards
120 untransmitted data.
121
122DOCUMENTATION --------------------------------------------------------
123
124Update README
125
126BUILD FARM -----------------------------------------------------------
127
128Add machines
129
130 AMDAHL UTS (Dave Dykstra)
131
132 Cygwin (on different versions of Win32?)
133
134 HP-UX variants (via HP?)
135
136 SCO
137
138NICE -----------------------------------------------------------------
139
140SIGHUP
141
142 Re-read config file (just exec() ourselves) rather than exiting.
143
144--no-detach and --no-fork options
145
146 Very useful for debugging. Also good when running under a
147 daemon-monitoring process that tries to restart the service when the
148 parent exits.
149
150hang/timeout friendliness
151
152 On
153
154internationalization
155
156 Change to using gettext(). Probably need to ship this for platforms
157 that don't have it.
158
159 Solicit translations.
160
161 Does anyone care?
162
163rsyncsh
164
165 Write a small emulation of interactive ftp as a Pythonn program
166 that calls rsync. Commands such as "cd", "ls", "ls *.c" etc map
167 fairly directly into rsync commands: it just needs to remember the
168 current host, directory and so on. We can probably even do
169 completion of remote filenames.
170