Documentation/git-tag.txt

   1 git-tag(1)
   2 ==========
   3
   4 NAME
   5 ----
   6 git-tag - Create, list, delete or verify a tag object signed with GPG
   7
   8
   9 SYNOPSIS
  10 --------
  11 [verse]
  12 'git tag' [-a | -s | -u <key-id>] [-f] [-m <msg> | -F <file>]
  13         <name> [<commit> | <object>]
  14 'git tag' -d <name>...
  15 'git tag' [-n[<num>]] -l [--contains <commit>] [<pattern>]
  16 'git tag' -v <name>...
  17
  18 DESCRIPTION
  19 -----------
  20 Adds a 'tag' reference in `.git/refs/tags/`
  21
  22 Unless `-f` is given, the tag must not yet exist in
  23 `.git/refs/tags/` directory.
  24
  25 If one of `-a`, `-s`, or `-u <key-id>` is passed, the command
  26 creates a 'tag' object, and requires the tag message.  Unless
  27 `-m <msg>` or `-F <file>` is given, an editor is started for the user to type
  28 in the tag message.
  29
  30 If `-m <msg>` or `-F <file>` is given and `-a`, `-s`, and `-u <key-id>`
  31 are absent, `-a` is implied.
  32
  33 Otherwise just the SHA1 object name of the commit object is
  34 written (i.e. a lightweight tag).
  35
  36 A GnuPG signed tag object will be created when `-s` or `-u
  37 <key-id>` is used.  When `-u <key-id>` is not used, the
  38 committer identity for the current user is used to find the
  39 GnuPG key for signing.
  40
  41 OPTIONS
  42 -------
  43 -a::
  44         Make an unsigned, annotated tag object
  45
  46 -s::
  47         Make a GPG-signed tag, using the default e-mail address's key
  48
  49 -u <key-id>::
  50         Make a GPG-signed tag, using the given key
  51
  52 -f::
  53         Replace an existing tag with the given name (instead of failing)
  54
  55 -d::
  56         Delete existing tags with the given names.
  57
  58 -v::
  59         Verify the gpg signature of the given tag names.
  60
  61 -n<num>::
  62         <num> specifies how many lines from the annotation, if any,
  63         are printed when using -l.
  64         The default is not to print any annotation lines.
  65         If no number is given to `-n`, only the first line is printed.
  66         If the tag is not annotated, the commit message is displayed instead.
  67
  68 -l <pattern>::
  69         List tags with names that match the given pattern (or all if no pattern is given).
  70         Typing "git tag" without arguments, also lists all tags.
  71
  72 --contains <commit>::
  73         Only list tags which contain the specified commit.
  74
  75 -m <msg>::
  76         Use the given tag message (instead of prompting).
  77         If multiple `-m` options are given, their values are
  78         concatenated as separate paragraphs.
  79         Implies `-a` if none of `-a`, `-s`, or `-u <key-id>`
  80         is given.
  81
  82 -F <file>::
  83         Take the tag message from the given file.  Use '-' to
  84         read the message from the standard input.
  85         Implies `-a` if none of `-a`, `-s`, or `-u <key-id>`
  86         is given.
  87
  88 CONFIGURATION
  89 -------------
  90 By default, 'git-tag' in sign-with-default mode (-s) will use your
  91 committer identity (of the form "Your Name <your@email.address>") to
  92 find a key.  If you want to use a different default key, you can specify
  93 it in the repository configuration as follows:
  94
  95 -------------------------------------
  96 [user]
  97     signingkey = <gpg-key-id>
  98 -------------------------------------
  99
 100
 101 DISCUSSION
 102 ----------
 103
 104 On Re-tagging
 105 ~~~~~~~~~~~~~
 106
 107 What should you do when you tag a wrong commit and you would
 108 want to re-tag?
 109
 110 If you never pushed anything out, just re-tag it. Use "-f" to
 111 replace the old one. And you're done.
 112
 113 But if you have pushed things out (or others could just read
 114 your repository directly), then others will have already seen
 115 the old tag. In that case you can do one of two things:
 116
 117 . The sane thing.
 118 Just admit you screwed up, and use a different name. Others have
 119 already seen one tag-name, and if you keep the same name, you
 120 may be in the situation that two people both have "version X",
 121 but they actually have 'different' "X"'s.  So just call it "X.1"
 122 and be done with it.
 123
 124 . The insane thing.
 125 You really want to call the new version "X" too, 'even though'
 126 others have already seen the old one. So just use 'git-tag -f'
 127 again, as if you hadn't already published the old one.
 128
 129 However, Git does *not* (and it should not) change tags behind
 130 users back. So if somebody already got the old tag, doing a
 131 'git-pull' on your tree shouldn't just make them overwrite the old
 132 one.
 133
 134 If somebody got a release tag from you, you cannot just change
 135 the tag for them by updating your own one. This is a big
 136 security issue, in that people MUST be able to trust their
 137 tag-names.  If you really want to do the insane thing, you need
 138 to just fess up to it, and tell people that you messed up. You
 139 can do that by making a very public announcement saying:
 140
 141 ------------
 142 Ok, I messed up, and I pushed out an earlier version tagged as X. I
 143 then fixed something, and retagged the *fixed* tree as X again.
 144
 145 If you got the wrong tag, and want the new one, please delete
 146 the old one and fetch the new one by doing:
 147
 148         git tag -d X
 149         git fetch origin tag X
 150
 151 to get my updated tag.
 152
 153 You can test which tag you have by doing
 154
 155         git rev-parse X
 156
 157 which should return 0123456789abcdef.. if you have the new version.
 158
 159 Sorry for inconvenience.
 160 ------------
 161
 162 Does this seem a bit complicated?  It *should* be. There is no
 163 way that it would be correct to just "fix" it behind peoples
 164 backs. People need to know that their tags might have been
 165 changed.
 166
 167
 168 On Automatic following
 169 ~~~~~~~~~~~~~~~~~~~~~~
 170
 171 If you are following somebody else's tree, you are most likely
 172 using tracking branches (`refs/heads/origin` in traditional
 173 layout, or `refs/remotes/origin/master` in the separate-remote
 174 layout).  You usually want the tags from the other end.
 175
 176 On the other hand, if you are fetching because you would want a
 177 one-shot merge from somebody else, you typically do not want to
 178 get tags from there.  This happens more often for people near
 179 the toplevel but not limited to them.  Mere mortals when pulling
 180 from each other do not necessarily want to automatically get
 181 private anchor point tags from the other person.
 182
 183 You would notice "please pull" messages on the mailing list says
 184 repo URL and branch name alone.  This is designed to be easily
 185 cut&pasted to a 'git-fetch' command line:
 186
 187 ------------
 188 Linus, please pull from
 189
 190         git://git..../proj.git master
 191
 192 to get the following updates...
 193 ------------
 194
 195 becomes:
 196
 197 ------------
 198 $ git pull git://git..../proj.git master
 199 ------------
 200
 201 In such a case, you do not want to automatically follow other's
 202 tags.
 203
 204 One important aspect of git is it is distributed, and being
 205 distributed largely means there is no inherent "upstream" or
 206 "downstream" in the system.  On the face of it, the above
 207 example might seem to indicate that the tag namespace is owned
 208 by upper echelon of people and tags only flow downwards, but
 209 that is not the case.  It only shows that the usage pattern
 210 determines who are interested in whose tags.
 211
 212 A one-shot pull is a sign that a commit history is now crossing
 213 the boundary between one circle of people (e.g. "people who are
 214 primarily interested in the networking part of the kernel") who may
 215 have their own set of tags (e.g. "this is the third release
 216 candidate from the networking group to be proposed for general
 217 consumption with 2.6.21 release") to another circle of people
 218 (e.g. "people who integrate various subsystem improvements").
 219 The latter are usually not interested in the detailed tags used
 220 internally in the former group (that is what "internal" means).
 221 That is why it is desirable not to follow tags automatically in
 222 this case.
 223
 224 It may well be that among networking people, they may want to
 225 exchange the tags internal to their group, but in that workflow
 226 they are most likely tracking with each other's progress by
 227 having tracking branches.  Again, the heuristic to automatically
 228 follow such tags is a good thing.
 229
 230
 231 On Backdating Tags
 232 ~~~~~~~~~~~~~~~~~~
 233
 234 If you have imported some changes from another VCS and would like
 235 to add tags for major releases of your work, it is useful to be able
 236 to specify the date to embed inside of the tag object.  The data in
 237 the tag object affects, for example, the ordering of tags in the
 238 gitweb interface.
 239
 240 To set the date used in future tag objects, set the environment
 241 variable GIT_COMMITTER_DATE to one or more of the date and time.  The
 242 date and time can be specified in a number of ways; the most common
 243 is "YYYY-MM-DD HH:MM".
 244
 245 An example follows.
 246
 247 ------------
 248 $ GIT_COMMITTER_DATE="2006-10-02 10:31" git tag -s v1.0.1
 249 ------------
 250
 251
 252 Author
 253 ------
 254 Written by Linus Torvalds <torvalds@osdl.org>,
 255 Junio C Hamano <gitster@pobox.com> and Chris Wright <chrisw@osdl.org>.
 256
 257 Documentation
 258 --------------
 259 Documentation by David Greaves, Junio C Hamano and the git-list <git@vger.kernel.org>.
 260
 261 GIT
 262 ---
 263 Part of the linkgit:git[1] suite