hgbook

annotate en/ch08-branch.xml @ 647:d0160b0b1a9e

Merge with http://hg.serpentine.com/mercurial/book
author Dongsheng Song <dongsheng.song@gmail.com>
date Wed Mar 18 20:32:37 2009 +0800 (2009-03-18)
parents cfdb601a3c8b 8366882f67f2
children
rev   line source
bos@559 1 <!-- vim: set filetype=docbkxml shiftwidth=2 autoindent expandtab tw=77 : -->
bos@559 2
dongsheng@625 3 <chapter id="chap.branch">
bos@572 4 <?dbhtml filename="managing-releases-and-branchy-development.html"?>
bos@559 5 <title>Managing releases and branchy development</title>
bos@559 6
bos@559 7 <para>Mercurial provides several mechanisms for you to manage a
bos@559 8 project that is making progress on multiple fronts at once. To
bos@559 9 understand these mechanisms, let's first take a brief look at a
bos@559 10 fairly normal software project structure.</para>
bos@559 11
bos@559 12 <para>Many software projects issue periodic <quote>major</quote>
bos@559 13 releases that contain substantial new features. In parallel, they
bos@559 14 may issue <quote>minor</quote> releases. These are usually
bos@559 15 identical to the major releases off which they're based, but with
bos@559 16 a few bugs fixed.</para>
bos@559 17
bos@559 18 <para>In this chapter, we'll start by talking about how to keep
bos@559 19 records of project milestones such as releases. We'll then
bos@559 20 continue on to talk about the flow of work between different
bos@559 21 phases of a project, and how Mercurial can help you to isolate and
bos@559 22 manage this work.</para>
bos@559 23
bos@559 24 <sect1>
bos@559 25 <title>Giving a persistent name to a revision</title>
bos@559 26
bos@559 27 <para>Once you decide that you'd like to call a particular
bos@559 28 revision a <quote>release</quote>, it's a good idea to record
bos@559 29 the identity of that revision. This will let you reproduce that
bos@559 30 release at a later date, for whatever purpose you might need at
bos@559 31 the time (reproducing a bug, porting to a new platform, etc).
bos@567 32 &interaction.tag.init;</para>
bos@559 33
bos@559 34 <para>Mercurial lets you give a permanent name to any revision
bos@559 35 using the <command role="hg-cmd">hg tag</command> command. Not
bos@567 36 surprisingly, these names are called <quote>tags</quote>.</para>
bos@567 37
bos@567 38 &interaction.tag.tag;
bos@559 39
bos@559 40 <para>A tag is nothing more than a <quote>symbolic name</quote>
bos@559 41 for a revision. Tags exist purely for your convenience, so that
bos@559 42 you have a handy permanent way to refer to a revision; Mercurial
bos@559 43 doesn't interpret the tag names you use in any way. Neither
bos@559 44 does Mercurial place any restrictions on the name of a tag,
bos@559 45 beyond a few that are necessary to ensure that a tag can be
bos@559 46 parsed unambiguously. A tag name cannot contain any of the
bos@559 47 following characters:</para>
bos@559 48 <itemizedlist>
bos@559 49 <listitem><para>Colon (ASCII 58,
bos@559 50 <quote><literal>:</literal></quote>)</para>
bos@559 51 </listitem>
bos@559 52 <listitem><para>Carriage return (ASCII 13,
bos@559 53 <quote><literal>\r</literal></quote>)</para>
bos@559 54 </listitem>
bos@559 55 <listitem><para>Newline (ASCII 10,
bos@559 56 <quote><literal>\n</literal></quote>)</para>
bos@559 57 </listitem></itemizedlist>
bos@559 58
bos@559 59 <para>You can use the <command role="hg-cmd">hg tags</command>
bos@559 60 command to display the tags present in your repository. In the
bos@559 61 output, each tagged revision is identified first by its name,
bos@559 62 then by revision number, and finally by the unique hash of the
bos@567 63 revision.</para>
bos@567 64
bos@567 65 &interaction.tag.tags;
bos@567 66
bos@567 67 <para>Notice that <literal>tip</literal> is listed in the output
bos@567 68 of <command role="hg-cmd">hg tags</command>. The
bos@567 69 <literal>tip</literal> tag is a special <quote>floating</quote>
bos@567 70 tag, which always identifies the newest revision in the
bos@567 71 repository.</para>
bos@559 72
bos@559 73 <para>In the output of the <command role="hg-cmd">hg
bos@559 74 tags</command> command, tags are listed in reverse order, by
bos@559 75 revision number. This usually means that recent tags are listed
bos@559 76 before older tags. It also means that <literal>tip</literal> is
bos@559 77 always going to be the first tag listed in the output of
bos@559 78 <command role="hg-cmd">hg tags</command>.</para>
bos@559 79
bos@559 80 <para>When you run <command role="hg-cmd">hg log</command>, if it
bos@559 81 displays a revision that has tags associated with it, it will
bos@567 82 print those tags.</para>
bos@567 83
bos@567 84 &interaction.tag.log;
bos@559 85
bos@559 86 <para>Any time you need to provide a revision ID to a Mercurial
bos@559 87 command, the command will accept a tag name in its place.
bos@559 88 Internally, Mercurial will translate your tag name into the
bos@567 89 corresponding revision ID, then use that.</para>
bos@567 90
bos@567 91 &interaction.tag.log.v1.0;
bos@559 92
bos@559 93 <para>There's no limit on the number of tags you can have in a
bos@559 94 repository, or on the number of tags that a single revision can
bos@559 95 have. As a practical matter, it's not a great idea to have
bos@559 96 <quote>too many</quote> (a number which will vary from project
bos@559 97 to project), simply because tags are supposed to help you to
bos@559 98 find revisions. If you have lots of tags, the ease of using
bos@559 99 them to identify revisions diminishes rapidly.</para>
bos@559 100
bos@559 101 <para>For example, if your project has milestones as frequent as
bos@559 102 every few days, it's perfectly reasonable to tag each one of
bos@559 103 those. But if you have a continuous build system that makes
bos@559 104 sure every revision can be built cleanly, you'd be introducing a
bos@559 105 lot of noise if you were to tag every clean build. Instead, you
bos@559 106 could tag failed builds (on the assumption that they're rare!),
bos@559 107 or simply not use tags to track buildability.</para>
bos@559 108
bos@559 109 <para>If you want to remove a tag that you no longer want, use
bos@567 110 <command role="hg-cmd">hg tag --remove</command>.</para>
bos@567 111
bos@567 112 &interaction.tag.remove;
bos@567 113
bos@567 114 <para>You can also modify a tag at any time, so that it identifies
bos@567 115 a different revision, by simply issuing a new <command
bos@567 116 role="hg-cmd">hg tag</command> command. You'll have to use the
bos@567 117 <option role="hg-opt-tag">-f</option> option to tell Mercurial
bos@567 118 that you <emphasis>really</emphasis> want to update the
bos@567 119 tag.</para>
bos@567 120
bos@567 121 &interaction.tag.replace;
bos@567 122
bos@567 123 <para>There will still be a permanent record of the previous
bos@567 124 identity of the tag, but Mercurial will no longer use it.
bos@567 125 There's thus no penalty to tagging the wrong revision; all you
bos@567 126 have to do is turn around and tag the correct revision once you
bos@567 127 discover your error.</para>
bos@559 128
bos@559 129 <para>Mercurial stores tags in a normal revision-controlled file
bos@559 130 in your repository. If you've created any tags, you'll find
bos@559 131 them in a file named <filename
bos@559 132 role="special">.hgtags</filename>. When you run the <command
bos@559 133 role="hg-cmd">hg tag</command> command, Mercurial modifies
bos@559 134 this file, then automatically commits the change to it. This
bos@559 135 means that every time you run <command role="hg-cmd">hg
bos@559 136 tag</command>, you'll see a corresponding changeset in the
bos@567 137 output of <command role="hg-cmd">hg log</command>.</para>
bos@567 138
bos@567 139 &interaction.tag.tip;
bos@559 140
bos@559 141 <sect2>
bos@559 142 <title>Handling tag conflicts during a merge</title>
bos@559 143
bos@559 144 <para>You won't often need to care about the <filename
bos@559 145 role="special">.hgtags</filename> file, but it sometimes
bos@559 146 makes its presence known during a merge. The format of the
bos@559 147 file is simple: it consists of a series of lines. Each line
bos@559 148 starts with a changeset hash, followed by a space, followed by
bos@559 149 the name of a tag.</para>
bos@559 150
bos@559 151 <para>If you're resolving a conflict in the <filename
bos@559 152 role="special">.hgtags</filename> file during a merge,
bos@559 153 there's one twist to modifying the <filename
bos@559 154 role="special">.hgtags</filename> file: when Mercurial is
bos@559 155 parsing the tags in a repository, it
bos@559 156 <emphasis>never</emphasis> reads the working copy of the
bos@559 157 <filename role="special">.hgtags</filename> file. Instead, it
bos@559 158 reads the <emphasis>most recently committed</emphasis>
bos@559 159 revision of the file.</para>
bos@559 160
bos@559 161 <para>An unfortunate consequence of this design is that you
bos@559 162 can't actually verify that your merged <filename
bos@559 163 role="special">.hgtags</filename> file is correct until
bos@559 164 <emphasis>after</emphasis> you've committed a change. So if
bos@559 165 you find yourself resolving a conflict on <filename
bos@559 166 role="special">.hgtags</filename> during a merge, be sure to
bos@559 167 run <command role="hg-cmd">hg tags</command> after you commit.
bos@559 168 If it finds an error in the <filename
bos@559 169 role="special">.hgtags</filename> file, it will report the
bos@559 170 location of the error, which you can then fix and commit. You
bos@559 171 should then run <command role="hg-cmd">hg tags</command>
bos@559 172 again, just to be sure that your fix is correct.</para>
bos@559 173
bos@559 174 </sect2>
bos@559 175 <sect2>
bos@559 176 <title>Tags and cloning</title>
bos@559 177
bos@559 178 <para>You may have noticed that the <command role="hg-cmd">hg
bos@559 179 clone</command> command has a <option
bos@559 180 role="hg-opt-clone">-r</option> option that lets you clone
bos@559 181 an exact copy of the repository as of a particular changeset.
bos@559 182 The new clone will not contain any project history that comes
bos@559 183 after the revision you specified. This has an interaction
bos@559 184 with tags that can surprise the unwary.</para>
bos@559 185
bos@559 186 <para>Recall that a tag is stored as a revision to the <filename
bos@559 187 role="special">.hgtags</filename> file, so that when you
bos@559 188 create a tag, the changeset in which it's recorded necessarily
bos@559 189 refers to an older changeset. When you run <command
bos@559 190 role="hg-cmd">hg clone -r foo</command> to clone a
bos@559 191 repository as of tag <literal>foo</literal>, the new clone
bos@559 192 <emphasis>will not contain the history that created the
bos@559 193 tag</emphasis> that you used to clone the repository. The
bos@559 194 result is that you'll get exactly the right subset of the
bos@559 195 project's history in the new repository, but
bos@559 196 <emphasis>not</emphasis> the tag you might have
bos@559 197 expected.</para>
bos@559 198
bos@559 199 </sect2>
bos@559 200 <sect2>
bos@559 201 <title>When permanent tags are too much</title>
bos@559 202
bos@559 203 <para>Since Mercurial's tags are revision controlled and carried
bos@559 204 around with a project's history, everyone you work with will
bos@559 205 see the tags you create. But giving names to revisions has
bos@559 206 uses beyond simply noting that revision
bos@559 207 <literal>4237e45506ee</literal> is really
bos@559 208 <literal>v2.0.2</literal>. If you're trying to track down a
bos@559 209 subtle bug, you might want a tag to remind you of something
bos@559 210 like <quote>Anne saw the symptoms with this
bos@559 211 revision</quote>.</para>
bos@559 212
bos@559 213 <para>For cases like this, what you might want to use are
bos@559 214 <emphasis>local</emphasis> tags. You can create a local tag
bos@559 215 with the <option role="hg-opt-tag">-l</option> option to the
bos@559 216 <command role="hg-cmd">hg tag</command> command. This will
bos@559 217 store the tag in a file called <filename
bos@559 218 role="special">.hg/localtags</filename>. Unlike <filename
bos@559 219 role="special">.hgtags</filename>, <filename
bos@559 220 role="special">.hg/localtags</filename> is not revision
bos@559 221 controlled. Any tags you create using <option
bos@559 222 role="hg-opt-tag">-l</option> remain strictly local to the
bos@559 223 repository you're currently working in.</para>
bos@559 224
bos@559 225 </sect2>
bos@559 226 </sect1>
bos@559 227 <sect1>
bos@559 228 <title>The flow of changes&emdash;big picture vs. little</title>
bos@559 229
bos@559 230 <para>To return to the outline I sketched at the beginning of a
bos@559 231 chapter, let's think about a project that has multiple
bos@559 232 concurrent pieces of work under development at once.</para>
bos@559 233
bos@559 234 <para>There might be a push for a new <quote>main</quote> release;
bos@559 235 a new minor bugfix release to the last main release; and an
bos@559 236 unexpected <quote>hot fix</quote> to an old release that is now
bos@559 237 in maintenance mode.</para>
bos@559 238
bos@559 239 <para>The usual way people refer to these different concurrent
bos@559 240 directions of development is as <quote>branches</quote>.
bos@559 241 However, we've already seen numerous times that Mercurial treats
bos@559 242 <emphasis>all of history</emphasis> as a series of branches and
bos@559 243 merges. Really, what we have here is two ideas that are
bos@559 244 peripherally related, but which happen to share a name.</para>
bos@559 245 <itemizedlist>
bos@559 246 <listitem><para><quote>Big picture</quote> branches represent
bos@559 247 the sweep of a project's evolution; people give them names,
bos@559 248 and talk about them in conversation.</para>
bos@559 249 </listitem>
bos@559 250 <listitem><para><quote>Little picture</quote> branches are
bos@559 251 artefacts of the day-to-day activity of developing and
bos@559 252 merging changes. They expose the narrative of how the code
bos@559 253 was developed.</para>
bos@559 254 </listitem></itemizedlist>
bos@559 255
bos@559 256 </sect1>
bos@559 257 <sect1>
bos@559 258 <title>Managing big-picture branches in repositories</title>
bos@559 259
bos@559 260 <para>The easiest way to isolate a <quote>big picture</quote>
bos@559 261 branch in Mercurial is in a dedicated repository. If you have
bos@559 262 an existing shared repository&emdash;let's call it
bos@559 263 <literal>myproject</literal>&emdash;that reaches a
bos@559 264 <quote>1.0</quote> milestone, you can start to prepare for
bos@559 265 future maintenance releases on top of version 1.0 by tagging the
bos@567 266 revision from which you prepared the 1.0 release.</para>
bos@567 267
bos@567 268 &interaction.branch-repo.tag;
bos@567 269
bos@567 270 <para>You can then clone a new shared
bos@567 271 <literal>myproject-1.0.1</literal> repository as of that
bos@567 272 tag.</para>
bos@567 273
bos@567 274 &interaction.branch-repo.clone;
bos@559 275
bos@559 276 <para>Afterwards, if someone needs to work on a bug fix that ought
bos@559 277 to go into an upcoming 1.0.1 minor release, they clone the
bos@559 278 <literal>myproject-1.0.1</literal> repository, make their
bos@567 279 changes, and push them back.</para>
bos@567 280
bos@567 281 &interaction.branch-repo.bugfix;
bos@567 282
bos@567 283 <para>Meanwhile, development for
bos@559 284 the next major release can continue, isolated and unabated, in
bos@567 285 the <literal>myproject</literal> repository.</para>
bos@567 286
bos@567 287 &interaction.branch-repo.new;
bos@559 288
bos@559 289 </sect1>
bos@559 290 <sect1>
bos@559 291 <title>Don't repeat yourself: merging across branches</title>
bos@559 292
bos@559 293 <para>In many cases, if you have a bug to fix on a maintenance
bos@559 294 branch, the chances are good that the bug exists on your
bos@559 295 project's main branch (and possibly other maintenance branches,
bos@559 296 too). It's a rare developer who wants to fix the same bug
bos@559 297 multiple times, so let's look at a few ways that Mercurial can
bos@559 298 help you to manage these bugfixes without duplicating your
bos@559 299 work.</para>
bos@559 300
bos@559 301 <para>In the simplest instance, all you need to do is pull changes
bos@559 302 from your maintenance branch into your local clone of the target
bos@567 303 branch.</para>
bos@567 304
bos@567 305 &interaction.branch-repo.pull;
bos@567 306
bos@567 307 <para>You'll then need to merge the heads of the two branches, and
bos@567 308 push back to the main branch.</para>
bos@567 309
bos@567 310 &interaction.branch-repo.merge;
bos@559 311
bos@559 312 </sect1>
bos@559 313 <sect1>
bos@559 314 <title>Naming branches within one repository</title>
bos@559 315
bos@559 316 <para>In most instances, isolating branches in repositories is the
bos@559 317 right approach. Its simplicity makes it easy to understand; and
bos@559 318 so it's hard to make mistakes. There's a one-to-one
bos@559 319 relationship between branches you're working in and directories
bos@559 320 on your system. This lets you use normal (non-Mercurial-aware)
bos@559 321 tools to work on files within a branch/repository.</para>
bos@559 322
bos@559 323 <para>If you're more in the <quote>power user</quote> category
bos@559 324 (<emphasis>and</emphasis> your collaborators are too), there is
bos@559 325 an alternative way of handling branches that you can consider.
bos@559 326 I've already mentioned the human-level distinction between
bos@559 327 <quote>small picture</quote> and <quote>big picture</quote>
bos@559 328 branches. While Mercurial works with multiple <quote>small
bos@559 329 picture</quote> branches in a repository all the time (for
bos@559 330 example after you pull changes in, but before you merge them),
bos@559 331 it can <emphasis>also</emphasis> work with multiple <quote>big
bos@559 332 picture</quote> branches.</para>
bos@559 333
bos@559 334 <para>The key to working this way is that Mercurial lets you
bos@559 335 assign a persistent <emphasis>name</emphasis> to a branch.
bos@559 336 There always exists a branch named <literal>default</literal>.
bos@559 337 Even before you start naming branches yourself, you can find
bos@559 338 traces of the <literal>default</literal> branch if you look for
bos@559 339 them.</para>
bos@559 340
bos@559 341 <para>As an example, when you run the <command role="hg-cmd">hg
bos@559 342 commit</command> command, and it pops up your editor so that
bos@559 343 you can enter a commit message, look for a line that contains
bos@559 344 the text <quote><literal>HG: branch default</literal></quote> at
bos@559 345 the bottom. This is telling you that your commit will occur on
bos@559 346 the branch named <literal>default</literal>.</para>
bos@559 347
bos@559 348 <para>To start working with named branches, use the <command
bos@559 349 role="hg-cmd">hg branches</command> command. This command
bos@559 350 lists the named branches already present in your repository,
bos@567 351 telling you which changeset is the tip of each.</para>
bos@567 352
bos@567 353 &interaction.branch-named.branches;
bos@567 354
bos@567 355 <para>Since you haven't created any named branches yet, the only
bos@567 356 one that exists is <literal>default</literal>.</para>
bos@559 357
bos@559 358 <para>To find out what the <quote>current</quote> branch is, run
bos@559 359 the <command role="hg-cmd">hg branch</command> command, giving
bos@559 360 it no arguments. This tells you what branch the parent of the
bos@567 361 current changeset is on.</para>
bos@567 362
bos@567 363 &interaction.branch-named.branch;
bos@559 364
bos@559 365 <para>To create a new branch, run the <command role="hg-cmd">hg
bos@559 366 branch</command> command again. This time, give it one
bos@567 367 argument: the name of the branch you want to create.</para>
bos@567 368
bos@567 369 &interaction.branch-named.create;
bos@559 370
bos@559 371 <para>After you've created a branch, you might wonder what effect
bos@559 372 the <command role="hg-cmd">hg branch</command> command has had.
bos@559 373 What do the <command role="hg-cmd">hg status</command> and
bos@567 374 <command role="hg-cmd">hg tip</command> commands report?</para>
bos@567 375
bos@567 376 &interaction.branch-named.status;
bos@567 377
bos@567 378 <para>Nothing has changed in the
bos@559 379 working directory, and there's been no new history created. As
bos@559 380 this suggests, running the <command role="hg-cmd">hg
bos@559 381 branch</command> command has no permanent effect; it only
bos@559 382 tells Mercurial what branch name to use the
bos@559 383 <emphasis>next</emphasis> time you commit a changeset.</para>
bos@559 384
bos@559 385 <para>When you commit a change, Mercurial records the name of the
bos@559 386 branch on which you committed. Once you've switched from the
bos@559 387 <literal>default</literal> branch to another and committed,
bos@559 388 you'll see the name of the new branch show up in the output of
bos@559 389 <command role="hg-cmd">hg log</command>, <command
bos@559 390 role="hg-cmd">hg tip</command>, and other commands that
bos@567 391 display the same kind of output.</para>
bos@567 392
bos@567 393 &interaction.branch-named.commit;
bos@567 394
bos@567 395 <para>The <command role="hg-cmd">hg log</command>-like commands
bos@567 396 will print the branch name of every changeset that's not on the
bos@559 397 <literal>default</literal> branch. As a result, if you never
bos@559 398 use named branches, you'll never see this information.</para>
bos@559 399
bos@559 400 <para>Once you've named a branch and committed a change with that
bos@559 401 name, every subsequent commit that descends from that change
bos@559 402 will inherit the same branch name. You can change the name of a
bos@559 403 branch at any time, using the <command role="hg-cmd">hg
bos@567 404 branch</command> command.</para>
bos@567 405
bos@567 406 &interaction.branch-named.rebranch;
bos@567 407
bos@567 408 <para>In practice, this is something you won't do very often, as
bos@567 409 branch names tend to have fairly long lifetimes. (This isn't a
bos@567 410 rule, just an observation.)</para>
bos@559 411
bos@559 412 </sect1>
bos@559 413 <sect1>
bos@559 414 <title>Dealing with multiple named branches in a
bos@559 415 repository</title>
bos@559 416
bos@559 417 <para>If you have more than one named branch in a repository,
bos@559 418 Mercurial will remember the branch that your working directory
bos@559 419 on when you start a command like <command role="hg-cmd">hg
bos@559 420 update</command> or <command role="hg-cmd">hg pull
bos@559 421 -u</command>. It will update the working directory to the tip
bos@559 422 of this branch, no matter what the <quote>repo-wide</quote> tip
bos@559 423 is. To update to a revision that's on a different named branch,
bos@559 424 you may need to use the <option role="hg-opt-update">-C</option>
bos@559 425 option to <command role="hg-cmd">hg update</command>.</para>
bos@559 426
bos@559 427 <para>This behaviour is a little subtle, so let's see it in
bos@559 428 action. First, let's remind ourselves what branch we're
bos@567 429 currently on, and what branches are in our repository.</para>
bos@567 430
bos@567 431 &interaction.branch-named.parents;
bos@567 432
bos@567 433 <para>We're on the <literal>bar</literal> branch, but there also
bos@567 434 exists an older <command role="hg-cmd">hg foo</command>
bos@567 435 branch.</para>
bos@559 436
bos@559 437 <para>We can <command role="hg-cmd">hg update</command> back and
bos@559 438 forth between the tips of the <literal>foo</literal> and
bos@559 439 <literal>bar</literal> branches without needing to use the
bos@559 440 <option role="hg-opt-update">-C</option> option, because this
bos@559 441 only involves going backwards and forwards linearly through our
bos@567 442 change history.</para>
bos@567 443
bos@567 444 &interaction.branch-named.update-switchy;
bos@559 445
bos@559 446 <para>If we go back to the <literal>foo</literal> branch and then
bos@559 447 run <command role="hg-cmd">hg update</command>, it will keep us
bos@559 448 on <literal>foo</literal>, not move us to the tip of
bos@567 449 <literal>bar</literal>.</para>
bos@567 450
bos@567 451 &interaction.branch-named.update-nothing;
bos@559 452
bos@559 453 <para>Committing a new change on the <literal>foo</literal> branch
bos@567 454 introduces a new head.</para>
bos@567 455
bos@567 456 &interaction.branch-named.foo-commit;
bos@559 457
bos@559 458 </sect1>
bos@559 459 <sect1>
bos@559 460 <title>Branch names and merging</title>
bos@559 461
bos@559 462 <para>As you've probably noticed, merges in Mercurial are not
bos@559 463 symmetrical. Let's say our repository has two heads, 17 and 23.
bos@559 464 If I <command role="hg-cmd">hg update</command> to 17 and then
bos@559 465 <command role="hg-cmd">hg merge</command> with 23, Mercurial
bos@559 466 records 17 as the first parent of the merge, and 23 as the
bos@559 467 second. Whereas if I <command role="hg-cmd">hg update</command>
bos@559 468 to 23 and then <command role="hg-cmd">hg merge</command> with
bos@559 469 17, it records 23 as the first parent, and 17 as the
bos@559 470 second.</para>
bos@559 471
bos@559 472 <para>This affects Mercurial's choice of branch name when you
bos@559 473 merge. After a merge, Mercurial will retain the branch name of
bos@559 474 the first parent when you commit the result of the merge. If
bos@559 475 your first parent's branch name is <literal>foo</literal>, and
bos@559 476 you merge with <literal>bar</literal>, the branch name will
bos@559 477 still be <literal>foo</literal> after you merge.</para>
bos@559 478
bos@559 479 <para>It's not unusual for a repository to contain multiple heads,
bos@559 480 each with the same branch name. Let's say I'm working on the
bos@559 481 <literal>foo</literal> branch, and so are you. We commit
bos@559 482 different changes; I pull your changes; I now have two heads,
bos@559 483 each claiming to be on the <literal>foo</literal> branch. The
bos@559 484 result of a merge will be a single head on the
bos@559 485 <literal>foo</literal> branch, as you might hope.</para>
bos@559 486
bos@559 487 <para>But if I'm working on the <literal>bar</literal> branch, and
bos@559 488 I merge work from the <literal>foo</literal> branch, the result
bos@567 489 will remain on the <literal>bar</literal> branch.</para>
bos@567 490
bos@567 491 &interaction.branch-named.merge;
bos@559 492
bos@559 493 <para>To give a more concrete example, if I'm working on the
bos@559 494 <literal>bleeding-edge</literal> branch, and I want to bring in
bos@559 495 the latest fixes from the <literal>stable</literal> branch,
bos@559 496 Mercurial will choose the <quote>right</quote>
bos@559 497 (<literal>bleeding-edge</literal>) branch name when I pull and
bos@559 498 merge from <literal>stable</literal>.</para>
bos@559 499
bos@559 500 </sect1>
bos@559 501 <sect1>
bos@559 502 <title>Branch naming is generally useful</title>
bos@559 503
bos@559 504 <para>You shouldn't think of named branches as applicable only to
bos@559 505 situations where you have multiple long-lived branches
bos@559 506 cohabiting in a single repository. They're very useful even in
bos@559 507 the one-branch-per-repository case.</para>
bos@559 508
bos@559 509 <para>In the simplest case, giving a name to each branch gives you
bos@559 510 a permanent record of which branch a changeset originated on.
bos@559 511 This gives you more context when you're trying to follow the
bos@559 512 history of a long-lived branchy project.</para>
bos@559 513
bos@559 514 <para>If you're working with shared repositories, you can set up a
bos@559 515 <literal role="hook">pretxnchangegroup</literal> hook on each
bos@559 516 that will block incoming changes that have the
bos@559 517 <quote>wrong</quote> branch name. This provides a simple, but
bos@559 518 effective, defence against people accidentally pushing changes
bos@559 519 from a <quote>bleeding edge</quote> branch to a
bos@559 520 <quote>stable</quote> branch. Such a hook might look like this
bos@559 521 inside the shared repo's <filename role="special">
bos@559 522 /.hgrc</filename>.</para>
bos@580 523 <programlisting>[hooks]
bos@580 524 pretxnchangegroup.branch = hg heads --template '{branches} ' | grep mybranch</programlisting>
bos@559 525
bos@559 526 </sect1>
bos@559 527 </chapter>
bos@559 528
bos@559 529 <!--
bos@559 530 local variables:
bos@559 531 sgml-parent-document: ("00book.xml" "book" "chapter")
bos@559 532 end:
bos@559 533 -->