hgbook

annotate fr/ch12-mq.xml @ 1030:2ac31ea48a3d

merge with trunk
author Romain PELISSE <belaran@gmail.com>
date Thu Apr 22 12:39:32 2010 +0200 (2010-04-22)
parents 102f89d20be1 6f8c48362758
children
rev   line source
bos@558 1 <!-- vim: set filetype=docbkxml shiftwidth=2 autoindent expandtab tw=77 : -->
bos@558 2
bos@558 3 <chapter id="chap:mq">
bos@572 4 <?dbhtml filename="managing-change-with-mercurial-queues.html"?>
bos@558 5 <title>Managing change with Mercurial Queues</title>
bos@558 6
bos@558 7 <sect1 id="sec:mq:patch-mgmt">
bos@558 8 <title>The patch management problem</title>
bos@558 9
bos@584 10 <para id="x_3ac">Here is a common scenario: you need to install a software
bos@558 11 package from source, but you find a bug that you must fix in the
bos@558 12 source before you can start using the package. You make your
bos@558 13 changes, forget about the package for a while, and a few months
bos@558 14 later you need to upgrade to a newer version of the package. If
bos@558 15 the newer version of the package still has the bug, you must
bos@558 16 extract your fix from the older source tree and apply it against
bos@558 17 the newer version. This is a tedious task, and it's easy to
bos@558 18 make mistakes.</para>
bos@558 19
bos@584 20 <para id="x_3ad">This is a simple case of the <quote>patch management</quote>
bos@558 21 problem. You have an <quote>upstream</quote> source tree that
bos@558 22 you can't change; you need to make some local changes on top of
bos@558 23 the upstream tree; and you'd like to be able to keep those
bos@558 24 changes separate, so that you can apply them to newer versions
bos@558 25 of the upstream source.</para>
bos@558 26
bos@584 27 <para id="x_3ae">The patch management problem arises in many situations.
bos@558 28 Probably the most visible is that a user of an open source
bos@558 29 software project will contribute a bug fix or new feature to the
bos@558 30 project's maintainers in the form of a patch.</para>
bos@558 31
bos@584 32 <para id="x_3af">Distributors of operating systems that include open source
bos@558 33 software often need to make changes to the packages they
bos@558 34 distribute so that they will build properly in their
bos@558 35 environments.</para>
bos@558 36
bos@584 37 <para id="x_3b0">When you have few changes to maintain, it is easy to manage
bos@558 38 a single patch using the standard <command>diff</command> and
bos@592 39 <command>patch</command> programs (see <xref
bos@559 40 linkend="sec:mq:patch"/> for a discussion of these
bos@559 41 tools). Once the number of changes grows, it starts to make
bos@559 42 sense to maintain patches as discrete <quote>chunks of
bos@559 43 work,</quote> so that for example a single patch will contain
bos@559 44 only one bug fix (the patch might modify several files, but it's
bos@559 45 doing <quote>only one thing</quote>), and you may have a number
bos@559 46 of such patches for different bugs you need fixed and local
bos@559 47 changes you require. In this situation, if you submit a bug fix
bos@559 48 patch to the upstream maintainers of a package and they include
bos@559 49 your fix in a subsequent release, you can simply drop that
bos@559 50 single patch when you're updating to the newer release.</para>
bos@558 51
bos@584 52 <para id="x_3b1">Maintaining a single patch against an upstream tree is a
bos@558 53 little tedious and error-prone, but not difficult. However, the
bos@558 54 complexity of the problem grows rapidly as the number of patches
bos@558 55 you have to maintain increases. With more than a tiny number of
bos@558 56 patches in hand, understanding which ones you have applied and
bos@558 57 maintaining them moves from messy to overwhelming.</para>
bos@558 58
bos@584 59 <para id="x_3b2">Fortunately, Mercurial includes a powerful extension,
bos@558 60 Mercurial Queues (or simply <quote>MQ</quote>), that massively
bos@558 61 simplifies the patch management problem.</para>
bos@558 62
bos@558 63 </sect1>
bos@558 64 <sect1 id="sec:mq:history">
bos@558 65 <title>The prehistory of Mercurial Queues</title>
bos@558 66
bos@584 67 <para id="x_3b3">During the late 1990s, several Linux kernel developers
bos@558 68 started to maintain <quote>patch series</quote> that modified
bos@672 69 the behavior of the Linux kernel. Some of these series were
bos@558 70 focused on stability, some on feature coverage, and others were
bos@558 71 more speculative.</para>
bos@558 72
bos@584 73 <para id="x_3b4">The sizes of these patch series grew rapidly. In 2002,
bos@558 74 Andrew Morton published some shell scripts he had been using to
bos@558 75 automate the task of managing his patch queues. Andrew was
bos@558 76 successfully using these scripts to manage hundreds (sometimes
bos@558 77 thousands) of patches on top of the Linux kernel.</para>
bos@558 78
bos@558 79 <sect2 id="sec:mq:quilt">
bos@558 80 <title>A patchwork quilt</title>
bos@558 81
bos@584 82 <para id="x_3b5">In early 2003, Andreas Gruenbacher and Martin Quinson
bos@558 83 borrowed the approach of Andrew's scripts and published a tool
bos@558 84 called <quote>patchwork quilt</quote>
bos@558 85 <citation>web:quilt</citation>, or simply <quote>quilt</quote>
bos@558 86 (see <citation>gruenbacher:2005</citation> for a paper
bos@558 87 describing it). Because quilt substantially automated patch
bos@558 88 management, it rapidly gained a large following among open
bos@558 89 source software developers.</para>
bos@558 90
bos@584 91 <para id="x_3b6">Quilt manages a <emphasis>stack of patches</emphasis> on
bos@558 92 top of a directory tree. To begin, you tell quilt to manage a
bos@558 93 directory tree, and tell it which files you want to manage; it
bos@558 94 stores away the names and contents of those files. To fix a
bos@558 95 bug, you create a new patch (using a single command), edit the
bos@558 96 files you need to fix, then <quote>refresh</quote> the
bos@558 97 patch.</para>
bos@558 98
bos@584 99 <para id="x_3b7">The refresh step causes quilt to scan the directory tree;
bos@558 100 it updates the patch with all of the changes you have made.
bos@558 101 You can create another patch on top of the first, which will
bos@558 102 track the changes required to modify the tree from <quote>tree
bos@558 103 with one patch applied</quote> to <quote>tree with two
bos@558 104 patches applied</quote>.</para>
bos@558 105
bos@584 106 <para id="x_3b8">You can <emphasis>change</emphasis> which patches are
bos@558 107 applied to the tree. If you <quote>pop</quote> a patch, the
bos@558 108 changes made by that patch will vanish from the directory
bos@558 109 tree. Quilt remembers which patches you have popped, though,
bos@558 110 so you can <quote>push</quote> a popped patch again, and the
bos@558 111 directory tree will be restored to contain the modifications
bos@558 112 in the patch. Most importantly, you can run the
bos@558 113 <quote>refresh</quote> command at any time, and the topmost
bos@558 114 applied patch will be updated. This means that you can, at
bos@558 115 any time, change both which patches are applied and what
bos@558 116 modifications those patches make.</para>
bos@558 117
bos@584 118 <para id="x_3b9">Quilt knows nothing about revision control tools, so it
bos@558 119 works equally well on top of an unpacked tarball or a
bos@558 120 Subversion working copy.</para>
bos@682 121 </sect2>
bos@682 122
bos@558 123 <sect2 id="sec:mq:quilt-mq">
bos@558 124 <title>From patchwork quilt to Mercurial Queues</title>
bos@558 125
bos@584 126 <para id="x_3ba">In mid-2005, Chris Mason took the features of quilt and
bos@558 127 wrote an extension that he called Mercurial Queues, which
bos@672 128 added quilt-like behavior to Mercurial.</para>
bos@558 129
bos@584 130 <para id="x_3bb">The key difference between quilt and MQ is that quilt
bos@558 131 knows nothing about revision control systems, while MQ is
bos@558 132 <emphasis>integrated</emphasis> into Mercurial. Each patch
bos@558 133 that you push is represented as a Mercurial changeset. Pop a
bos@558 134 patch, and the changeset goes away.</para>
bos@558 135
bos@584 136 <para id="x_3bc">Because quilt does not care about revision control tools,
bos@558 137 it is still a tremendously useful piece of software to know
bos@558 138 about for situations where you cannot use Mercurial and
bos@558 139 MQ.</para>
bos@558 140
bos@558 141 </sect2>
bos@558 142 </sect1>
bos@558 143 <sect1>
bos@558 144 <title>The huge advantage of MQ</title>
bos@558 145
bos@584 146 <para id="x_3bd">I cannot overstate the value that MQ offers through the
bos@558 147 unification of patches and revision control.</para>
bos@558 148
bos@584 149 <para id="x_3be">A major reason that patches have persisted in the free
bos@559 150 software and open source world&emdash;in spite of the
bos@559 151 availability of increasingly capable revision control tools over
bos@559 152 the years&emdash;is the <emphasis>agility</emphasis> they
bos@559 153 offer.</para>
bos@558 154
bos@584 155 <para id="x_3bf">Traditional revision control tools make a permanent,
bos@558 156 irreversible record of everything that you do. While this has
bos@558 157 great value, it's also somewhat stifling. If you want to
bos@558 158 perform a wild-eyed experiment, you have to be careful in how
bos@558 159 you go about it, or you risk leaving unneeded&emdash;or worse,
bos@559 160 misleading or destabilising&emdash;traces of your missteps and
bos@559 161 errors in the permanent revision record.</para>
bos@558 162
bos@584 163 <para id="x_3c0">By contrast, MQ's marriage of distributed revision control
bos@558 164 with patches makes it much easier to isolate your work. Your
bos@558 165 patches live on top of normal revision history, and you can make
bos@558 166 them disappear or reappear at will. If you don't like a patch,
bos@558 167 you can drop it. If a patch isn't quite as you want it to be,
bos@559 168 simply fix it&emdash;as many times as you need to, until you
bos@559 169 have refined it into the form you desire.</para>
bos@558 170
bos@584 171 <para id="x_3c1">As an example, the integration of patches with revision
bos@558 172 control makes understanding patches and debugging their
bos@558 173 effects&emdash;and their interplay with the code they're based
bos@559 174 on&emdash;<emphasis>enormously</emphasis> easier. Since every
bos@559 175 applied patch has an associated changeset, you can give <command
bos@559 176 role="hg-cmd">hg log</command> a file name to see which
bos@559 177 changesets and patches affected the file. You can use the
bos@559 178 <command role="hg-cmd">hg bisect</command> command to
bos@559 179 binary-search through all changesets and applied patches to see
bos@559 180 where a bug got introduced or fixed. You can use the <command
bos@558 181 role="hg-cmd">hg annotate</command> command to see which
bos@558 182 changeset or patch modified a particular line of a source file.
bos@558 183 And so on.</para>
bos@682 184 </sect1>
bos@682 185
bos@558 186 <sect1 id="sec:mq:patch">
bos@558 187 <title>Understanding patches</title>
bos@558 188
bos@584 189 <para id="x_3c2">Because MQ doesn't hide its patch-oriented nature, it is
bos@558 190 helpful to understand what patches are, and a little about the
bos@558 191 tools that work with them.</para>
bos@558 192
bos@584 193 <para id="x_3c3">The traditional Unix <command>diff</command> command
bos@558 194 compares two files, and prints a list of differences between
bos@558 195 them. The <command>patch</command> command understands these
bos@558 196 differences as <emphasis>modifications</emphasis> to make to a
bos@558 197 file. Take a look below for a simple example of these commands
bos@558 198 in action.</para>
bos@558 199
bos@689 200 &interaction.mq.dodiff.diff;
bos@558 201
bos@584 202 <para id="x_3c4">The type of file that <command>diff</command> generates (and
bos@558 203 <command>patch</command> takes as input) is called a
bos@558 204 <quote>patch</quote> or a <quote>diff</quote>; there is no
bos@558 205 difference between a patch and a diff. (We'll use the term
bos@558 206 <quote>patch</quote>, since it's more commonly used.)</para>
bos@558 207
bos@584 208 <para id="x_3c5">A patch file can start with arbitrary text; the
bos@558 209 <command>patch</command> command ignores this text, but MQ uses
bos@558 210 it as the commit message when creating changesets. To find the
bos@558 211 beginning of the patch content, <command>patch</command>
bos@558 212 searches for the first line that starts with the string
bos@558 213 <quote><literal>diff -</literal></quote>.</para>
bos@558 214
bos@584 215 <para id="x_3c6">MQ works with <emphasis>unified</emphasis> diffs
bos@558 216 (<command>patch</command> can accept several other diff formats,
bos@558 217 but MQ doesn't). A unified diff contains two kinds of header.
bos@558 218 The <emphasis>file header</emphasis> describes the file being
bos@558 219 modified; it contains the name of the file to modify. When
bos@558 220 <command>patch</command> sees a new file header, it looks for a
bos@558 221 file with that name to start modifying.</para>
bos@558 222
bos@584 223 <para id="x_3c7">After the file header comes a series of
bos@558 224 <emphasis>hunks</emphasis>. Each hunk starts with a header;
bos@558 225 this identifies the range of line numbers within the file that
bos@558 226 the hunk should modify. Following the header, a hunk starts and
bos@558 227 ends with a few (usually three) lines of text from the
bos@558 228 unmodified file; these are called the
bos@558 229 <emphasis>context</emphasis> for the hunk. If there's only a
bos@558 230 small amount of context between successive hunks,
bos@558 231 <command>diff</command> doesn't print a new hunk header; it just
bos@558 232 runs the hunks together, with a few lines of context between
bos@558 233 modifications.</para>
bos@558 234
bos@584 235 <para id="x_3c8">Each line of context begins with a space character. Within
bos@558 236 the hunk, a line that begins with
bos@558 237 <quote><literal>-</literal></quote> means <quote>remove this
bos@558 238 line,</quote> while a line that begins with
bos@558 239 <quote><literal>+</literal></quote> means <quote>insert this
bos@558 240 line.</quote> For example, a line that is modified is
bos@558 241 represented by one deletion and one insertion.</para>
bos@558 242
bos@584 243 <para id="x_3c9">We will return to some of the more subtle aspects of patches
bos@592 244 later (in <xref linkend="sec:mq:adv-patch"/>), but you
bos@559 245 should have
bos@559 246 enough information now to use MQ.</para>
bos@682 247 </sect1>
bos@682 248
bos@558 249 <sect1 id="sec:mq:start">
bos@558 250 <title>Getting started with Mercurial Queues</title>
bos@558 251
bos@584 252 <para id="x_3ca">Because MQ is implemented as an extension, you must
bos@558 253 explicitly enable before you can use it. (You don't need to
bos@558 254 download anything; MQ ships with the standard Mercurial
bos@559 255 distribution.) To enable MQ, edit your <filename
bos@559 256 role="home">~/.hgrc</filename> file, and add the lines
bos@559 257 below.</para>
bos@558 258
bos@580 259 <programlisting>[extensions]
bos@580 260 hgext.mq =</programlisting>
bos@558 261
bos@584 262 <para id="x_3cb">Once the extension is enabled, it will make a number of new
bos@558 263 commands available. To verify that the extension is working,
bos@558 264 you can use <command role="hg-cmd">hg help</command> to see if
bos@559 265 the <command role="hg-ext-mq">qinit</command> command is now
bos@558 266 available.</para>
bos@558 267
bos@682 268 &interaction.mq.qinit-help.help;
bos@558 269
bos@584 270 <para id="x_3cc">You can use MQ with <emphasis>any</emphasis> Mercurial
bos@558 271 repository, and its commands only operate within that
bos@558 272 repository. To get started, simply prepare the repository using
bos@559 273 the <command role="hg-ext-mq">qinit</command> command.</para>
bos@559 274
bos@682 275 &interaction.mq.tutorial.qinit;
bos@559 276
bos@584 277 <para id="x_3cd">This command creates an empty directory called <filename
bos@558 278 role="special" class="directory">.hg/patches</filename>, where
bos@558 279 MQ will keep its metadata. As with many Mercurial commands, the
bos@559 280 <command role="hg-ext-mq">qinit</command> command prints nothing
bos@558 281 if it succeeds.</para>
bos@558 282
bos@558 283 <sect2>
bos@558 284 <title>Creating a new patch</title>
bos@558 285
bos@584 286 <para id="x_3ce">To begin work on a new patch, use the <command
bos@559 287 role="hg-ext-mq">qnew</command> command. This command takes
bos@559 288 one argument, the name of the patch to create.</para>
bos@559 289
bos@584 290 <para id="x_3cf">MQ will use this as the name of an actual file in the
bos@559 291 <filename role="special"
bos@559 292 class="directory">.hg/patches</filename> directory, as you
bos@559 293 can see below.</para>
bos@559 294
bos@682 295 &interaction.mq.tutorial.qnew;
bos@558 296
bos@584 297 <para id="x_3d0">Also newly present in the <filename role="special"
bos@558 298 class="directory">.hg/patches</filename> directory are two
bos@558 299 other files, <filename role="special">series</filename> and
bos@558 300 <filename role="special">status</filename>. The <filename
bos@558 301 role="special">series</filename> file lists all of the
bos@558 302 patches that MQ knows about for this repository, with one
bos@558 303 patch per line. Mercurial uses the <filename
bos@558 304 role="special">status</filename> file for internal
bos@558 305 book-keeping; it tracks all of the patches that MQ has
bos@558 306 <emphasis>applied</emphasis> in this repository.</para>
bos@558 307
bos@558 308 <note>
bos@584 309 <para id="x_3d1"> You may sometimes want to edit the <filename
bos@558 310 role="special">series</filename> file by hand; for
bos@558 311 example, to change the sequence in which some patches are
bos@558 312 applied. However, manually editing the <filename
bos@558 313 role="special">status</filename> file is almost always a
bos@558 314 bad idea, as it's easy to corrupt MQ's idea of what is
bos@558 315 happening.</para>
bos@558 316 </note>
bos@558 317
bos@584 318 <para id="x_3d2">Once you have created your new patch, you can edit files
bos@558 319 in the working directory as you usually would. All of the
bos@558 320 normal Mercurial commands, such as <command role="hg-cmd">hg
bos@558 321 diff</command> and <command role="hg-cmd">hg
bos@558 322 annotate</command>, work exactly as they did before.</para>
bos@682 323 </sect2>
bos@682 324
bos@558 325 <sect2>
bos@558 326 <title>Refreshing a patch</title>
bos@558 327
bos@584 328 <para id="x_3d3">When you reach a point where you want to save your work,
bos@559 329 use the <command role="hg-ext-mq">qrefresh</command> command
bos@559 330 to update the patch you are working on.</para>
bos@558 331
bos@682 332 &interaction.mq.tutorial.qrefresh;
bos@558 333
bos@584 334 <para id="x_3d4">This command folds the changes you have made in the
bos@559 335 working directory into your patch, and updates its
bos@559 336 corresponding changeset to contain those changes.</para>
bos@559 337
bos@584 338 <para id="x_3d5">You can run <command role="hg-ext-mq">qrefresh</command>
bos@559 339 as often as you like, so it's a good way to
bos@559 340 <quote>checkpoint</quote> your work. Refresh your patch at an
bos@559 341 opportune time; try an experiment; and if the experiment
bos@558 342 doesn't work out, <command role="hg-cmd">hg revert</command>
bos@558 343 your modifications back to the last time you refreshed.</para>
bos@558 344
bos@682 345 &interaction.mq.tutorial.qrefresh2;
bos@682 346 </sect2>
bos@682 347
bos@558 348 <sect2>
bos@558 349 <title>Stacking and tracking patches</title>
bos@558 350
bos@584 351 <para id="x_3d6">Once you have finished working on a patch, or need to work
bos@559 352 on another, you can use the <command
bos@559 353 role="hg-ext-mq">qnew</command> command again to create a
bos@559 354 new patch. Mercurial will apply this patch on top of your
bos@559 355 existing patch.</para>
bos@559 356
bos@682 357 &interaction.mq.tutorial.qnew2;
bos@682 358
bos@584 359 <para id="x_3d7">Notice that the patch contains the changes in our prior
bos@559 360 patch as part of its context (you can see this more clearly in
bos@559 361 the output of <command role="hg-cmd">hg
bos@558 362 annotate</command>).</para>
bos@558 363
bos@584 364 <para id="x_3d8">So far, with the exception of <command
bos@559 365 role="hg-ext-mq">qnew</command> and <command
bos@559 366 role="hg-ext-mq">qrefresh</command>, we've been careful to
bos@558 367 only use regular Mercurial commands. However, MQ provides
bos@558 368 many commands that are easier to use when you are thinking
bos@559 369 about patches, as illustrated below.</para>
bos@559 370
bos@682 371 &interaction.mq.tutorial.qseries;
bos@558 372
bos@558 373 <itemizedlist>
bos@584 374 <listitem><para id="x_3d9">The <command
bos@559 375 role="hg-ext-mq">qseries</command> command lists every
bos@558 376 patch that MQ knows about in this repository, from oldest
bos@558 377 to newest (most recently
bos@559 378 <emphasis>created</emphasis>).</para>
bos@559 379 </listitem>
bos@584 380 <listitem><para id="x_3da">The <command
bos@559 381 role="hg-ext-mq">qapplied</command> command lists every
bos@558 382 patch that MQ has <emphasis>applied</emphasis> in this
bos@558 383 repository, again from oldest to newest (most recently
bos@559 384 applied).</para>
bos@559 385 </listitem></itemizedlist>
bos@682 386 </sect2>
bos@682 387
bos@558 388 <sect2>
bos@558 389 <title>Manipulating the patch stack</title>
bos@558 390
bos@584 391 <para id="x_3db">The previous discussion implied that there must be a
bos@558 392 difference between <quote>known</quote> and
bos@558 393 <quote>applied</quote> patches, and there is. MQ can manage a
bos@558 394 patch without it being applied in the repository.</para>
bos@558 395
bos@584 396 <para id="x_3dc">An <emphasis>applied</emphasis> patch has a corresponding
bos@558 397 changeset in the repository, and the effects of the patch and
bos@558 398 changeset are visible in the working directory. You can undo
bos@558 399 the application of a patch using the <command
bos@559 400 role="hg-ext-mq">qpop</command> command. MQ still
bos@558 401 <emphasis>knows about</emphasis>, or manages, a popped patch,
bos@558 402 but the patch no longer has a corresponding changeset in the
bos@558 403 repository, and the working directory does not contain the
bos@592 404 changes made by the patch. <xref
bos@559 405 linkend="fig:mq:stack"/> illustrates
bos@559 406 the difference between applied and tracked patches.</para>
bos@558 407
bos@592 408 <figure id="fig:mq:stack">
bos@592 409 <title>Applied and unapplied patches in the MQ patch
bos@592 410 stack</title>
bos@592 411 <mediaobject>
bos@594 412 <imageobject><imagedata fileref="figs/mq-stack.png"/></imageobject>
bos@592 413 <textobject><phrase>XXX add text</phrase></textobject>
bos@592 414 </mediaobject>
bos@592 415 </figure>
bos@558 416
bos@584 417 <para id="x_3de">You can reapply an unapplied, or popped, patch using the
bos@559 418 <command role="hg-ext-mq">qpush</command> command. This
bos@558 419 creates a new changeset to correspond to the patch, and the
bos@558 420 patch's changes once again become present in the working
bos@559 421 directory. See below for examples of <command
bos@559 422 role="hg-ext-mq">qpop</command> and <command
bos@559 423 role="hg-ext-mq">qpush</command> in action.</para>
bos@682 424
bos@682 425 &interaction.mq.tutorial.qpop;
bos@558 426
bos@584 427 <para id="x_3df">Notice that once we have popped a patch or two patches,
bos@559 428 the output of <command role="hg-ext-mq">qseries</command>
bos@559 429 remains the same, while that of <command
bos@559 430 role="hg-ext-mq">qapplied</command> has changed.</para>
bos@559 431
bos@682 432 </sect2>
bos@682 433
bos@558 434 <sect2>
bos@558 435 <title>Pushing and popping many patches</title>
bos@558 436
bos@584 437 <para id="x_3e0">While <command role="hg-ext-mq">qpush</command> and
bos@559 438 <command role="hg-ext-mq">qpop</command> each operate on a
bos@558 439 single patch at a time by default, you can push and pop many
bos@558 440 patches in one go. The <option
steve@872 441 role="hg-ext-mq-cmd-qpush-opt">-a</option> option to
bos@559 442 <command role="hg-ext-mq">qpush</command> causes it to push
bos@558 443 all unapplied patches, while the <option
bos@558 444 role="hg-ext-mq-cmd-qpop-opt">-a</option> option to <command
bos@559 445 role="hg-ext-mq">qpop</command> causes it to pop all applied
bos@558 446 patches. (For some more ways to push and pop many patches,
bos@592 447 see <xref linkend="sec:mq:perf"/> below.)</para>
bos@558 448
bos@682 449 &interaction.mq.tutorial.qpush-a;
bos@682 450 </sect2>
bos@682 451
bos@558 452 <sect2>
bos@558 453 <title>Safety checks, and overriding them</title>
bos@558 454
bos@584 455 <para id="x_3e1">Several MQ commands check the working directory before
bos@558 456 they do anything, and fail if they find any modifications.
bos@558 457 They do this to ensure that you won't lose any changes that
bos@559 458 you have made, but not yet incorporated into a patch. The
bos@559 459 example below illustrates this; the <command
bos@559 460 role="hg-ext-mq">qnew</command> command will not create a
bos@558 461 new patch if there are outstanding changes, caused in this
bos@558 462 case by the <command role="hg-cmd">hg add</command> of
bos@558 463 <filename>file3</filename>.</para>
bos@558 464
bos@682 465 &interaction.mq.tutorial.add;
bos@558 466
bos@584 467 <para id="x_3e2">Commands that check the working directory all take an
bos@558 468 <quote>I know what I'm doing</quote> option, which is always
bos@558 469 named <option>-f</option>. The exact meaning of
bos@558 470 <option>-f</option> depends on the command. For example,
bos@558 471 <command role="hg-cmd">hg qnew <option
steve@872 472 role="hg-ext-mq-cmd-qnew-opt">-f</option></command>
bos@559 473 will incorporate any outstanding changes into the new patch it
bos@558 474 creates, but <command role="hg-cmd">hg qpop <option
steve@872 475 role="hg-ext-mq-cmd-qpop-opt">-f</option></command>
bos@559 476 will revert modifications to any files affected by the patch
bos@559 477 that it is popping. Be sure to read the documentation for a
bos@558 478 command's <option>-f</option> option before you use it!</para>
bos@682 479 </sect2>
bos@682 480
bos@558 481 <sect2>
bos@558 482 <title>Working on several patches at once</title>
bos@558 483
bos@584 484 <para id="x_3e3">The <command role="hg-ext-mq">qrefresh</command> command
bos@558 485 always refreshes the <emphasis>topmost</emphasis> applied
bos@558 486 patch. This means that you can suspend work on one patch (by
bos@558 487 refreshing it), pop or push to make a different patch the top,
bos@558 488 and work on <emphasis>that</emphasis> patch for a
bos@558 489 while.</para>
bos@558 490
bos@584 491 <para id="x_3e4">Here's an example that illustrates how you can use this
bos@558 492 ability. Let's say you're developing a new feature as two
bos@558 493 patches. The first is a change to the core of your software,
bos@559 494 and the second&emdash;layered on top of the
bos@559 495 first&emdash;changes the user interface to use the code you
bos@559 496 just added to the core. If you notice a bug in the core while
bos@559 497 you're working on the UI patch, it's easy to fix the core.
bos@559 498 Simply <command role="hg-ext-mq">qrefresh</command> the UI
bos@559 499 patch to save your in-progress changes, and <command
bos@559 500 role="hg-ext-mq">qpop</command> down to the core patch. Fix
bos@559 501 the core bug, <command role="hg-ext-mq">qrefresh</command> the
bos@559 502 core patch, and <command role="hg-ext-mq">qpush</command> back
bos@558 503 to the UI patch to continue where you left off.</para>
bos@682 504 </sect2>
bos@682 505 </sect1>
bos@682 506
bos@558 507 <sect1 id="sec:mq:adv-patch">
bos@558 508 <title>More about patches</title>
bos@558 509
bos@584 510 <para id="x_3e5">MQ uses the GNU <command>patch</command> command to apply
bos@558 511 patches, so it's helpful to know a few more detailed aspects of
bos@558 512 how <command>patch</command> works, and about patches
bos@558 513 themselves.</para>
bos@558 514
bos@558 515 <sect2>
bos@558 516 <title>The strip count</title>
bos@558 517
bos@584 518 <para id="x_3e6">If you look at the file headers in a patch, you will
bos@558 519 notice that the pathnames usually have an extra component on
bos@558 520 the front that isn't present in the actual path name. This is
bos@558 521 a holdover from the way that people used to generate patches
bos@558 522 (people still do this, but it's somewhat rare with modern
bos@558 523 revision control tools).</para>
bos@558 524
bos@584 525 <para id="x_3e7">Alice would unpack a tarball, edit her files, then decide
bos@558 526 that she wanted to create a patch. So she'd rename her
bos@558 527 working directory, unpack the tarball again (hence the need
bos@558 528 for the rename), and use the <option
bos@558 529 role="cmd-opt-diff">-r</option> and <option
bos@558 530 role="cmd-opt-diff">-N</option> options to
bos@558 531 <command>diff</command> to recursively generate a patch
bos@558 532 between the unmodified directory and the modified one. The
bos@558 533 result would be that the name of the unmodified directory
bos@558 534 would be at the front of the left-hand path in every file
bos@558 535 header, and the name of the modified directory would be at the
bos@558 536 front of the right-hand path.</para>
bos@558 537
bos@584 538 <para id="x_3e8">Since someone receiving a patch from the Alices of the net
bos@558 539 would be unlikely to have unmodified and modified directories
bos@558 540 with exactly the same names, the <command>patch</command>
bos@558 541 command has a <option role="cmd-opt-patch">-p</option> option
bos@558 542 that indicates the number of leading path name components to
bos@558 543 strip when trying to apply a patch. This number is called the
bos@558 544 <emphasis>strip count</emphasis>.</para>
bos@558 545
bos@584 546 <para id="x_3e9">An option of <quote><literal>-p1</literal></quote> means
bos@558 547 <quote>use a strip count of one</quote>. If
bos@558 548 <command>patch</command> sees a file name
bos@558 549 <filename>foo/bar/baz</filename> in a file header, it will
bos@558 550 strip <filename>foo</filename> and try to patch a file named
bos@558 551 <filename>bar/baz</filename>. (Strictly speaking, the strip
bos@558 552 count refers to the number of <emphasis>path
bos@558 553 separators</emphasis> (and the components that go with them
bos@558 554 ) to strip. A strip count of one will turn
bos@558 555 <filename>foo/bar</filename> into <filename>bar</filename>,
bos@558 556 but <filename>/foo/bar</filename> (notice the extra leading
bos@558 557 slash) into <filename>foo/bar</filename>.)</para>
bos@558 558
bos@584 559 <para id="x_3ea">The <quote>standard</quote> strip count for patches is
bos@558 560 one; almost all patches contain one leading path name
bos@558 561 component that needs to be stripped. Mercurial's <command
bos@558 562 role="hg-cmd">hg diff</command> command generates path names
bos@558 563 in this form, and the <command role="hg-cmd">hg
bos@558 564 import</command> command and MQ expect patches to have a
bos@558 565 strip count of one.</para>
bos@558 566
bos@584 567 <para id="x_3eb">If you receive a patch from someone that you want to add
bos@558 568 to your patch queue, and the patch needs a strip count other
bos@558 569 than one, you cannot just <command
bos@559 570 role="hg-ext-mq">qimport</command> the patch, because
bos@559 571 <command role="hg-ext-mq">qimport</command> does not yet have
bos@558 572 a <literal>-p</literal> option (see <ulink role="hg-bug"
bos@558 573 url="http://www.selenic.com/mercurial/bts/issue311">issue
bos@558 574 311</ulink>). Your best bet is to <command
bos@559 575 role="hg-ext-mq">qnew</command> a patch of your own, then
bos@558 576 use <command>patch -pN</command> to apply their patch,
bos@558 577 followed by <command role="hg-cmd">hg addremove</command> to
bos@558 578 pick up any files added or removed by the patch, followed by
bos@559 579 <command role="hg-ext-mq">hg qrefresh</command>. This
bos@559 580 complexity may become unnecessary; see <ulink role="hg-bug"
bos@558 581 url="http://www.selenic.com/mercurial/bts/issue311">issue
bos@559 582 311</ulink> for details.
bos@559 583 </para>
bos@558 584 </sect2>
bos@682 585
bos@558 586 <sect2>
bos@558 587 <title>Strategies for applying a patch</title>
bos@558 588
bos@584 589 <para id="x_3ec">When <command>patch</command> applies a hunk, it tries a
bos@558 590 handful of successively less accurate strategies to try to
bos@558 591 make the hunk apply. This falling-back technique often makes
bos@558 592 it possible to take a patch that was generated against an old
bos@558 593 version of a file, and apply it against a newer version of
bos@558 594 that file.</para>
bos@558 595
bos@584 596 <para id="x_3ed">First, <command>patch</command> tries an exact match,
bos@558 597 where the line numbers, the context, and the text to be
bos@558 598 modified must apply exactly. If it cannot make an exact
bos@558 599 match, it tries to find an exact match for the context,
bos@558 600 without honouring the line numbering information. If this
bos@558 601 succeeds, it prints a line of output saying that the hunk was
bos@558 602 applied, but at some <emphasis>offset</emphasis> from the
bos@558 603 original line number.</para>
bos@558 604
bos@584 605 <para id="x_3ee">If a context-only match fails, <command>patch</command>
bos@558 606 removes the first and last lines of the context, and tries a
bos@558 607 <emphasis>reduced</emphasis> context-only match. If the hunk
bos@558 608 with reduced context succeeds, it prints a message saying that
bos@558 609 it applied the hunk with a <emphasis>fuzz factor</emphasis>
bos@558 610 (the number after the fuzz factor indicates how many lines of
bos@558 611 context <command>patch</command> had to trim before the patch
bos@558 612 applied).</para>
bos@558 613
bos@584 614 <para id="x_3ef">When neither of these techniques works,
bos@558 615 <command>patch</command> prints a message saying that the hunk
bos@558 616 in question was rejected. It saves rejected hunks (also
bos@558 617 simply called <quote>rejects</quote>) to a file with the same
bos@558 618 name, and an added <filename role="special">.rej</filename>
bos@558 619 extension. It also saves an unmodified copy of the file with
bos@558 620 a <filename role="special">.orig</filename> extension; the
bos@558 621 copy of the file without any extensions will contain any
bos@558 622 changes made by hunks that <emphasis>did</emphasis> apply
bos@558 623 cleanly. If you have a patch that modifies
bos@558 624 <filename>foo</filename> with six hunks, and one of them fails
bos@558 625 to apply, you will have: an unmodified
bos@558 626 <filename>foo.orig</filename>, a <filename>foo.rej</filename>
bos@558 627 containing one hunk, and <filename>foo</filename>, containing
bos@558 628 the changes made by the five successful hunks.</para>
bos@682 629 </sect2>
bos@682 630
bos@558 631 <sect2>
bos@558 632 <title>Some quirks of patch representation</title>
bos@558 633
bos@584 634 <para id="x_3f0">There are a few useful things to know about how
bos@558 635 <command>patch</command> works with files.</para>
bos@558 636 <itemizedlist>
bos@584 637 <listitem><para id="x_3f1">This should already be obvious, but
bos@558 638 <command>patch</command> cannot handle binary
bos@559 639 files.</para>
bos@559 640 </listitem>
bos@584 641 <listitem><para id="x_3f2">Neither does it care about the executable bit;
bos@558 642 it creates new files as readable, but not
bos@559 643 executable.</para>
bos@559 644 </listitem>
bos@584 645 <listitem><para id="x_3f3"><command>patch</command> treats the removal of
bos@558 646 a file as a diff between the file to be removed and the
bos@558 647 empty file. So your idea of <quote>I deleted this
bos@558 648 file</quote> looks like <quote>every line of this file
bos@559 649 was deleted</quote> in a patch.</para>
bos@559 650 </listitem>
bos@584 651 <listitem><para id="x_3f4">It treats the addition of a file as a diff
bos@558 652 between the empty file and the file to be added. So in a
bos@558 653 patch, your idea of <quote>I added this file</quote> looks
bos@558 654 like <quote>every line of this file was
bos@559 655 added</quote>.</para>
bos@559 656 </listitem>
bos@584 657 <listitem><para id="x_3f5">It treats a renamed file as the removal of the
bos@558 658 old name, and the addition of the new name. This means
bos@558 659 that renamed files have a big footprint in patches. (Note
bos@558 660 also that Mercurial does not currently try to infer when
bos@559 661 files have been renamed or copied in a patch.)</para>
bos@559 662 </listitem>
bos@584 663 <listitem><para id="x_3f6"><command>patch</command> cannot represent
bos@558 664 empty files, so you cannot use a patch to represent the
bos@558 665 notion <quote>I added this empty file to the
bos@559 666 tree</quote>.</para>
bos@559 667 </listitem></itemizedlist>
bos@559 668 </sect2>
bos@682 669
bos@558 670 <sect2>
bos@558 671 <title>Beware the fuzz</title>
bos@558 672
bos@584 673 <para id="x_3f7">While applying a hunk at an offset, or with a fuzz factor,
bos@558 674 will often be completely successful, these inexact techniques
bos@558 675 naturally leave open the possibility of corrupting the patched
bos@558 676 file. The most common cases typically involve applying a
bos@558 677 patch twice, or at an incorrect location in the file. If
bos@558 678 <command>patch</command> or <command
bos@559 679 role="hg-ext-mq">qpush</command> ever mentions an offset or
bos@558 680 fuzz factor, you should make sure that the modified files are
bos@558 681 correct afterwards.</para>
bos@558 682
bos@584 683 <para id="x_3f8">It's often a good idea to refresh a patch that has applied
bos@558 684 with an offset or fuzz factor; refreshing the patch generates
bos@558 685 new context information that will make it apply cleanly. I
bos@558 686 say <quote>often,</quote> not <quote>always,</quote> because
bos@558 687 sometimes refreshing a patch will make it fail to apply
bos@558 688 against a different revision of the underlying files. In some
bos@558 689 cases, such as when you're maintaining a patch that must sit
bos@558 690 on top of multiple versions of a source tree, it's acceptable
bos@558 691 to have a patch apply with some fuzz, provided you've verified
bos@558 692 the results of the patching process in such cases.</para>
bos@682 693 </sect2>
bos@682 694
bos@558 695 <sect2>
bos@558 696 <title>Handling rejection</title>
bos@558 697
bos@584 698 <para id="x_3f9">If <command role="hg-ext-mq">qpush</command> fails to
bos@558 699 apply a patch, it will print an error message and exit. If it
bos@558 700 has left <filename role="special">.rej</filename> files
bos@558 701 behind, it is usually best to fix up the rejected hunks before
bos@558 702 you push more patches or do any further work.</para>
bos@558 703
bos@584 704 <para id="x_3fa">If your patch <emphasis>used to</emphasis> apply cleanly,
bos@558 705 and no longer does because you've changed the underlying code
bos@558 706 that your patches are based on, Mercurial Queues can help; see
bos@592 707 <xref linkend="sec:mq:merge"/> for details.</para>
bos@558 708
bos@584 709 <para id="x_3fb">Unfortunately, there aren't any great techniques for
bos@558 710 dealing with rejected hunks. Most often, you'll need to view
bos@558 711 the <filename role="special">.rej</filename> file and edit the
bos@558 712 target file, applying the rejected hunks by hand.</para>
bos@558 713
bos@682 714 <para id="x_3fd">A Linux kernel hacker, Chris Mason (the author
bos@682 715 of Mercurial Queues), wrote a tool called
bos@682 716 <command>mpatch</command> (<ulink
bos@682 717 url="http://oss.oracle.com/~mason/mpatch/">http://oss.oracle.com/~mason/mpatch/</ulink>),
bos@558 718 which takes a simple approach to automating the application of
bos@558 719 hunks rejected by <command>patch</command>. The
bos@558 720 <command>mpatch</command> command can help with four common
bos@558 721 reasons that a hunk may be rejected:</para>
bos@558 722
bos@558 723 <itemizedlist>
bos@584 724 <listitem><para id="x_3fe">The context in the middle of a hunk has
bos@559 725 changed.</para>
bos@559 726 </listitem>
bos@584 727 <listitem><para id="x_3ff">A hunk is missing some context at the
bos@559 728 beginning or end.</para>
bos@559 729 </listitem>
bos@584 730 <listitem><para id="x_400">A large hunk might apply better&emdash;either
bos@559 731 entirely or in part&emdash;if it was broken up into
bos@559 732 smaller hunks.</para>
bos@559 733 </listitem>
bos@584 734 <listitem><para id="x_401">A hunk removes lines with slightly different
bos@559 735 content than those currently present in the file.</para>
bos@559 736 </listitem></itemizedlist>
bos@558 737
bos@682 738 <para id="x_402">If you use <command>mpatch</command>, you
bos@682 739 should be doubly careful to check your results when you're
bos@682 740 done. In fact, <command>mpatch</command> enforces this method
bos@682 741 of double-checking the tool's output, by automatically
bos@682 742 dropping you into a merge program when it has done its job, so
bos@682 743 that you can verify its work and finish off any remaining
bos@558 744 merges.</para>
bos@682 745 </sect2>
bos@682 746 </sect1>
bos@682 747
bos@682 748 <sect1>
bos@682 749 <title>More on patch management</title>
bos@682 750
bos@684 751 <para id="x_6db">As you grow familiar with MQ, you will find yourself wanting
bos@682 752 to perform other kinds of patch management operations.</para>
bos@682 753
bos@682 754 <sect2>
bos@682 755 <title>Deleting unwanted patches</title>
bos@682 756
bos@684 757 <para id="x_6dc">If you want to get rid of a patch, use the <command
bos@682 758 role="hg-ext-mq">hg qdelete</command> command to delete the
bos@682 759 patch file and remove its entry from the patch series. If you
bos@682 760 try to delete a patch that is still applied, <command
bos@682 761 role="hg-ext-mq">hg qdelete</command> will refuse.</para>
bos@682 762
bos@682 763 &interaction.ch11-qdelete.go;
bos@682 764 </sect2>
bos@682 765
bos@682 766 <sect2>
bos@682 767 <title>Converting to and from permanent revisions</title>
bos@682 768
bos@689 769 <para id="x_6dd">Once you're done working on a patch and want to
bos@689 770 turn it into a permanent changeset, use the <command
bos@689 771 role="hg-ext-mq">hg qfinish</command> command. Pass a revision
bos@689 772 to the command to identify the patch that you want to turn into
bos@689 773 a regular changeset; this patch must already be applied.</para>
bos@682 774
bos@682 775 &interaction.ch11-qdelete.convert;
bos@682 776
bos@689 777 <para id="x_6e0">The <command role="hg-ext-mq">hg qfinish</command> command
bos@689 778 accepts an <option>--all</option> or <option>-a</option>
bos@689 779 option, which turns all applied patches into regular
bos@689 780 changesets.</para>
bos@689 781
bos@684 782 <para id="x_6de">It is also possible to turn an existing changeset into a
bos@682 783 patch, by passing the <option>-r</option> option to <command
bos@682 784 role="hg-ext-mq">hg qimport</command>.</para>
bos@682 785
bos@682 786 &interaction.ch11-qdelete.import;
bos@682 787
bos@684 788 <para id="x_6df">Note that it only makes sense to convert a changeset into
bos@682 789 a patch if you have not propagated that changeset into any
bos@682 790 other repositories. The imported changeset's ID will change
bos@682 791 every time you refresh the patch, which will make Mercurial
bos@682 792 treat it as unrelated to the original changeset if you have
bos@682 793 pushed it somewhere else.</para>
bos@682 794 </sect2>
bos@682 795 </sect1>
bos@682 796
bos@558 797 <sect1 id="sec:mq:perf">
bos@558 798 <title>Getting the best performance out of MQ</title>
bos@558 799
bos@683 800 <para id="x_403">MQ is very efficient at handling a large number
bos@683 801 of patches. I ran some performance experiments in mid-2006 for a
bos@683 802 talk that I gave at the 2006 EuroPython conference (on modern
bos@683 803 hardware, you should expect better performance than you'll see
bos@683 804 below). I used as my data set the Linux 2.6.17-mm1 patch
bos@683 805 series, which consists of 1,738 patches. I applied these on top
bos@683 806 of a Linux kernel repository containing all 27,472 revisions
bos@683 807 between Linux 2.6.12-rc2 and Linux 2.6.17.</para>
bos@558 808
bos@584 809 <para id="x_404">On my old, slow laptop, I was able to <command
bos@558 810 role="hg-cmd">hg qpush <option
steve@872 811 role="hg-ext-mq-cmd-qpush-opt">-a</option></command> all
bos@558 812 1,738 patches in 3.5 minutes, and <command role="hg-cmd">hg qpop
steve@872 813 <option role="hg-ext-mq-cmd-qpop-opt">-a</option></command>
bos@558 814 them all in 30 seconds. (On a newer laptop, the time to push
bos@558 815 all patches dropped to two minutes.) I could <command
bos@559 816 role="hg-ext-mq">qrefresh</command> one of the biggest patches
bos@558 817 (which made 22,779 lines of changes to 287 files) in 6.6
bos@558 818 seconds.</para>
bos@558 819
bos@584 820 <para id="x_405">Clearly, MQ is well suited to working in large trees, but
bos@558 821 there are a few tricks you can use to get the best performance
bos@558 822 of it.</para>
bos@558 823
bos@584 824 <para id="x_406">First of all, try to <quote>batch</quote> operations
bos@558 825 together. Every time you run <command
bos@559 826 role="hg-ext-mq">qpush</command> or <command
bos@559 827 role="hg-ext-mq">qpop</command>, these commands scan the
bos@558 828 working directory once to make sure you haven't made some
bos@558 829 changes and then forgotten to run <command
bos@559 830 role="hg-ext-mq">qrefresh</command>. On a small tree, the
bos@558 831 time that this scan takes is unnoticeable. However, on a
bos@558 832 medium-sized tree (containing tens of thousands of files), it
bos@558 833 can take a second or more.</para>
bos@558 834
bos@584 835 <para id="x_407">The <command role="hg-ext-mq">qpush</command> and <command
bos@559 836 role="hg-ext-mq">qpop</command> commands allow you to push and
bos@558 837 pop multiple patches at a time. You can identify the
bos@558 838 <quote>destination patch</quote> that you want to end up at.
bos@559 839 When you <command role="hg-ext-mq">qpush</command> with a
bos@558 840 destination specified, it will push patches until that patch is
bos@558 841 at the top of the applied stack. When you <command
bos@559 842 role="hg-ext-mq">qpop</command> to a destination, MQ will pop
bos@558 843 patches until the destination patch is at the top.</para>
bos@558 844
bos@584 845 <para id="x_408">You can identify a destination patch using either the name
bos@558 846 of the patch, or by number. If you use numeric addressing,
bos@558 847 patches are counted from zero; this means that the first patch
bos@558 848 is zero, the second is one, and so on.</para>
bos@682 849 </sect1>
bos@682 850
bos@558 851 <sect1 id="sec:mq:merge">
bos@558 852 <title>Updating your patches when the underlying code
bos@558 853 changes</title>
bos@558 854
bos@584 855 <para id="x_409">It's common to have a stack of patches on top of an
bos@558 856 underlying repository that you don't modify directly. If you're
bos@558 857 working on changes to third-party code, or on a feature that is
bos@558 858 taking longer to develop than the rate of change of the code
bos@558 859 beneath, you will often need to sync up with the underlying
bos@558 860 code, and fix up any hunks in your patches that no longer apply.
bos@558 861 This is called <emphasis>rebasing</emphasis> your patch
bos@558 862 series.</para>
bos@558 863
bos@584 864 <para id="x_40a">The simplest way to do this is to <command role="hg-cmd">hg
bos@559 865 qpop <option role="hg-ext-mq-cmd-qpop-opt">hg
bos@559 866 -a</option></command> your patches, then <command
bos@559 867 role="hg-cmd">hg pull</command> changes into the underlying
bos@559 868 repository, and finally <command role="hg-cmd">hg qpush <option
steve@872 869 role="hg-ext-mq-cmd-qpop-opt">-a</option></command> your
bos@558 870 patches again. MQ will stop pushing any time it runs across a
bos@558 871 patch that fails to apply during conflicts, allowing you to fix
bos@559 872 your conflicts, <command role="hg-ext-mq">qrefresh</command> the
bos@558 873 affected patch, and continue pushing until you have fixed your
bos@558 874 entire stack.</para>
bos@558 875
bos@584 876 <para id="x_40b">This approach is easy to use and works well if you don't
bos@558 877 expect changes to the underlying code to affect how well your
bos@558 878 patches apply. If your patch stack touches code that is modified
bos@558 879 frequently or invasively in the underlying repository, however,
bos@558 880 fixing up rejected hunks by hand quickly becomes
bos@558 881 tiresome.</para>
bos@558 882
bos@584 883 <para id="x_40c">It's possible to partially automate the rebasing process.
bos@559 884 If your patches apply cleanly against some revision of the
bos@558 885 underlying repo, MQ can use this information to help you to
bos@558 886 resolve conflicts between your patches and a different
bos@558 887 revision.</para>
bos@558 888
bos@584 889 <para id="x_40d">The process is a little involved.</para>
bos@558 890 <orderedlist>
bos@584 891 <listitem><para id="x_40e">To begin, <command role="hg-cmd">hg qpush
bos@558 892 -a</command> all of your patches on top of the revision
bos@559 893 where you know that they apply cleanly.</para>
bos@559 894 </listitem>
bos@584 895 <listitem><para id="x_40f">Save a backup copy of your patch directory using
bos@558 896 <command role="hg-cmd">hg qsave <option
bos@559 897 role="hg-ext-mq-cmd-qsave-opt">hg -e</option> <option
bos@559 898 role="hg-ext-mq-cmd-qsave-opt">hg -c</option></command>.
bos@558 899 This prints the name of the directory that it has saved the
bos@558 900 patches in. It will save the patches to a directory called
bos@558 901 <filename role="special"
bos@558 902 class="directory">.hg/patches.N</filename>, where
bos@558 903 <literal>N</literal> is a small integer. It also commits a
bos@558 904 <quote>save changeset</quote> on top of your applied
bos@558 905 patches; this is for internal book-keeping, and records the
bos@558 906 states of the <filename role="special">series</filename> and
bos@559 907 <filename role="special">status</filename> files.</para>
bos@559 908 </listitem>
bos@584 909 <listitem><para id="x_410">Use <command role="hg-cmd">hg pull</command> to
bos@559 910 bring new changes into the underlying repository. (Don't
bos@559 911 run <command role="hg-cmd">hg pull -u</command>; see below
bos@559 912 for why.)</para>
bos@559 913 </listitem>
bos@584 914 <listitem><para id="x_411">Update to the new tip revision, using <command
bos@558 915 role="hg-cmd">hg update <option
bos@558 916 role="hg-opt-update">-C</option></command> to override
bos@559 917 the patches you have pushed.</para>
bos@559 918 </listitem>
bos@584 919 <listitem><para id="x_412">Merge all patches using <command>hg qpush -m
bos@580 920 -a</command>. The <option
bos@580 921 role="hg-ext-mq-cmd-qpush-opt">-m</option> option to
bos@559 922 <command role="hg-ext-mq">qpush</command> tells MQ to
bos@558 923 perform a three-way merge if the patch fails to
bos@559 924 apply.</para>
bos@559 925 </listitem></orderedlist>
bos@558 926
bos@584 927 <para id="x_413">During the <command role="hg-cmd">hg qpush <option
bos@559 928 role="hg-ext-mq-cmd-qpush-opt">hg -m</option></command>,
bos@559 929 each patch in the <filename role="special">series</filename>
bos@559 930 file is applied normally. If a patch applies with fuzz or
bos@559 931 rejects, MQ looks at the queue you <command
bos@559 932 role="hg-ext-mq">qsave</command>d, and performs a three-way
bos@558 933 merge with the corresponding changeset. This merge uses
bos@558 934 Mercurial's normal merge machinery, so it may pop up a GUI merge
bos@558 935 tool to help you to resolve problems.</para>
bos@558 936
bos@584 937 <para id="x_414">When you finish resolving the effects of a patch, MQ
bos@558 938 refreshes your patch based on the result of the merge.</para>
bos@558 939
bos@584 940 <para id="x_415">At the end of this process, your repository will have one
bos@558 941 extra head from the old patch queue, and a copy of the old patch
bos@558 942 queue will be in <filename role="special"
bos@558 943 class="directory">.hg/patches.N</filename>. You can remove the
bos@559 944 extra head using <command role="hg-cmd">hg qpop -a -n
bos@558 945 patches.N</command> or <command role="hg-cmd">hg
bos@558 946 strip</command>. You can delete <filename role="special"
bos@558 947 class="directory">.hg/patches.N</filename> once you are sure
bos@558 948 that you no longer need it as a backup.</para>
bos@682 949 </sect1>
bos@682 950
bos@558 951 <sect1>
bos@558 952 <title>Identifying patches</title>
bos@558 953
bos@584 954 <para id="x_416">MQ commands that work with patches let you refer to a patch
bos@558 955 either by using its name or by a number. By name is obvious
bos@558 956 enough; pass the name <filename>foo.patch</filename> to <command
bos@559 957 role="hg-ext-mq">qpush</command>, for example, and it will
bos@558 958 push patches until <filename>foo.patch</filename> is
bos@558 959 applied.</para>
bos@558 960
bos@584 961 <para id="x_417">As a shortcut, you can refer to a patch using both a name
bos@558 962 and a numeric offset; <literal>foo.patch-2</literal> means
bos@558 963 <quote>two patches before <literal>foo.patch</literal></quote>,
bos@558 964 while <literal>bar.patch+4</literal> means <quote>four patches
bos@558 965 after <literal>bar.patch</literal></quote>.</para>
bos@558 966
bos@584 967 <para id="x_418">Referring to a patch by index isn't much different. The
bos@558 968 first patch printed in the output of <command
bos@559 969 role="hg-ext-mq">qseries</command> is patch zero (yes, it's
bos@558 970 one of those start-at-zero counting systems); the second is
bos@558 971 patch one; and so on.</para>
bos@558 972
bos@584 973 <para id="x_419">MQ also makes it easy to work with patches when you are
bos@558 974 using normal Mercurial commands. Every command that accepts a
bos@558 975 changeset ID will also accept the name of an applied patch. MQ
bos@558 976 augments the tags normally in the repository with an eponymous
bos@558 977 one for each applied patch. In addition, the special tags
bos@580 978 <literal role="tag">qbase</literal> and
bos@580 979 <literal role="tag">qtip</literal> identify
bos@558 980 the <quote>bottom-most</quote> and topmost applied patches,
bos@558 981 respectively.</para>
bos@558 982
bos@584 983 <para id="x_41a">These additions to Mercurial's normal tagging capabilities
bos@558 984 make dealing with patches even more of a breeze.</para>
bos@558 985 <itemizedlist>
bos@584 986 <listitem><para id="x_41b">Want to patchbomb a mailing list with your
bos@558 987 latest series of changes?</para>
bos@580 988 <programlisting>hg email qbase:qtip</programlisting>
bos@584 989 <para id="x_41c"> (Don't know what <quote>patchbombing</quote> is? See
bos@592 990 <xref linkend="sec:hgext:patchbomb"/>.)</para>
bos@559 991 </listitem>
bos@584 992 <listitem><para id="x_41d">Need to see all of the patches since
bos@558 993 <literal>foo.patch</literal> that have touched files in a
bos@558 994 subdirectory of your tree?</para>
bos@580 995 <programlisting>hg log -r foo.patch:qtip subdir</programlisting>
bos@559 996 </listitem>
bos@558 997 </itemizedlist>
bos@558 998
bos@584 999 <para id="x_41e">Because MQ makes the names of patches available to the rest
bos@558 1000 of Mercurial through its normal internal tag machinery, you
bos@558 1001 don't need to type in the entire name of a patch when you want
bos@558 1002 to identify it by name.</para>
bos@558 1003
bos@584 1004 <para id="x_41f">Another nice consequence of representing patch names as tags
bos@558 1005 is that when you run the <command role="hg-cmd">hg log</command>
bos@558 1006 command, it will display a patch's name as a tag, simply as part
bos@558 1007 of its normal output. This makes it easy to visually
bos@558 1008 distinguish applied patches from underlying
bos@559 1009 <quote>normal</quote> revisions. The following example shows a
bos@559 1010 few normal Mercurial commands in use with applied
bos@559 1011 patches.</para>
bos@559 1012
bos@682 1013 &interaction.mq.id.output;
bos@682 1014 </sect1>
bos@682 1015
bos@558 1016 <sect1>
bos@558 1017 <title>Useful things to know about</title>
bos@558 1018
bos@584 1019 <para id="x_420">There are a number of aspects of MQ usage that don't fit
bos@558 1020 tidily into sections of their own, but that are good to know.
bos@558 1021 Here they are, in one place.</para>
bos@558 1022
bos@558 1023 <itemizedlist>
bos@584 1024 <listitem><para id="x_421">Normally, when you <command
bos@559 1025 role="hg-ext-mq">qpop</command> a patch and <command
bos@559 1026 role="hg-ext-mq">qpush</command> it again, the changeset
bos@558 1027 that represents the patch after the pop/push will have a
bos@558 1028 <emphasis>different identity</emphasis> than the changeset
bos@592 1029 that represented the hash beforehand. See <xref
bos@559 1030 linkend="sec:mqref:cmd:qpush"/> for
bos@559 1031 information as to why this is.</para>
bos@559 1032 </listitem>
bos@584 1033 <listitem><para id="x_422">It's not a good idea to <command
bos@558 1034 role="hg-cmd">hg merge</command> changes from another
bos@558 1035 branch with a patch changeset, at least if you want to
bos@558 1036 maintain the <quote>patchiness</quote> of that changeset and
bos@558 1037 changesets below it on the patch stack. If you try to do
bos@558 1038 this, it will appear to succeed, but MQ will become
bos@559 1039 confused.</para>
bos@559 1040 </listitem></itemizedlist>
bos@682 1041 </sect1>
bos@682 1042
bos@558 1043 <sect1 id="sec:mq:repo">
bos@558 1044 <title>Managing patches in a repository</title>
bos@558 1045
bos@584 1046 <para id="x_423">Because MQ's <filename role="special"
bos@558 1047 class="directory">.hg/patches</filename> directory resides
bos@558 1048 outside a Mercurial repository's working directory, the
bos@558 1049 <quote>underlying</quote> Mercurial repository knows nothing
bos@558 1050 about the management or presence of patches.</para>
bos@558 1051
bos@584 1052 <para id="x_424">This presents the interesting possibility of managing the
bos@558 1053 contents of the patch directory as a Mercurial repository in its
bos@558 1054 own right. This can be a useful way to work. For example, you
bos@558 1055 can work on a patch for a while, <command
bos@559 1056 role="hg-ext-mq">qrefresh</command> it, then <command
bos@558 1057 role="hg-cmd">hg commit</command> the current state of the
bos@558 1058 patch. This lets you <quote>roll back</quote> to that version
bos@558 1059 of the patch later on.</para>
bos@558 1060
bos@584 1061 <para id="x_425">You can then share different versions of the same patch
bos@558 1062 stack among multiple underlying repositories. I use this when I
bos@558 1063 am developing a Linux kernel feature. I have a pristine copy of
bos@558 1064 my kernel sources for each of several CPU architectures, and a
bos@558 1065 cloned repository under each that contains the patches I am
bos@558 1066 working on. When I want to test a change on a different
bos@558 1067 architecture, I push my current patches to the patch repository
bos@558 1068 associated with that kernel tree, pop and push all of my
bos@558 1069 patches, and build and test that kernel.</para>
bos@558 1070
bos@584 1071 <para id="x_426">Managing patches in a repository makes it possible for
bos@558 1072 multiple developers to work on the same patch series without
bos@558 1073 colliding with each other, all on top of an underlying source
bos@558 1074 base that they may or may not control.</para>
bos@558 1075
bos@558 1076 <sect2>
bos@558 1077 <title>MQ support for patch repositories</title>
bos@558 1078
bos@584 1079 <para id="x_427">MQ helps you to work with the <filename role="special"
bos@558 1080 class="directory">.hg/patches</filename> directory as a
bos@558 1081 repository; when you prepare a repository for working with
bos@559 1082 patches using <command role="hg-ext-mq">qinit</command>, you
bos@559 1083 can pass the <option role="hg-ext-mq-cmd-qinit-opt">hg
bos@559 1084 -c</option> option to create the <filename role="special"
bos@558 1085 class="directory">.hg/patches</filename> directory as a
bos@558 1086 Mercurial repository.</para>
bos@558 1087
bos@558 1088 <note>
bos@584 1089 <para id="x_428"> If you forget to use the <option
bos@559 1090 role="hg-ext-mq-cmd-qinit-opt">hg -c</option> option, you
bos@559 1091 can simply go into the <filename role="special"
bos@558 1092 class="directory">.hg/patches</filename> directory at any
bos@559 1093 time and run <command role="hg-cmd">hg init</command>.
bos@559 1094 Don't forget to add an entry for the <filename
bos@558 1095 role="special">status</filename> file to the <filename
bos@558 1096 role="special">.hgignore</filename> file, though</para>
bos@558 1097
bos@584 1098 <para id="x_429"> (<command role="hg-cmd">hg qinit <option
bos@559 1099 role="hg-ext-mq-cmd-qinit-opt">hg -c</option></command>
bos@558 1100 does this for you automatically); you
bos@558 1101 <emphasis>really</emphasis> don't want to manage the
bos@558 1102 <filename role="special">status</filename> file.</para>
bos@558 1103 </note>
bos@558 1104
bos@584 1105 <para id="x_42a">As a convenience, if MQ notices that the <filename
bos@558 1106 class="directory">.hg/patches</filename> directory is a
bos@558 1107 repository, it will automatically <command role="hg-cmd">hg
bos@558 1108 add</command> every patch that you create and import.</para>
bos@558 1109
bos@584 1110 <para id="x_42b">MQ provides a shortcut command, <command
bos@559 1111 role="hg-ext-mq">qcommit</command>, that runs <command
bos@558 1112 role="hg-cmd">hg commit</command> in the <filename
bos@558 1113 role="special" class="directory">.hg/patches</filename>
bos@558 1114 directory. This saves some bothersome typing.</para>
bos@558 1115
bos@584 1116 <para id="x_42c">Finally, as a convenience to manage the patch directory,
bos@558 1117 you can define the alias <command>mq</command> on Unix
bos@558 1118 systems. For example, on Linux systems using the
bos@558 1119 <command>bash</command> shell, you can include the following
bos@559 1120 snippet in your <filename
bos@559 1121 role="home">~/.bashrc</filename>.</para>
bos@559 1122
bos@580 1123 <programlisting>alias mq=`hg -R $(hg root)/.hg/patches'</programlisting>
bos@558 1124
bos@584 1125 <para id="x_42d">You can then issue commands of the form <command>mq
bos@558 1126 pull</command> from the main repository.</para>
bos@682 1127 </sect2>
bos@682 1128
bos@558 1129 <sect2>
bos@558 1130 <title>A few things to watch out for</title>
bos@558 1131
bos@584 1132 <para id="x_42e">MQ's support for working with a repository full of patches
bos@558 1133 is limited in a few small respects.</para>
bos@558 1134
bos@584 1135 <para id="x_42f">MQ cannot automatically detect changes that you make to
bos@558 1136 the patch directory. If you <command role="hg-cmd">hg
bos@558 1137 pull</command>, manually edit, or <command role="hg-cmd">hg
bos@558 1138 update</command> changes to patches or the <filename
bos@558 1139 role="special">series</filename> file, you will have to
bos@558 1140 <command role="hg-cmd">hg qpop <option
steve@872 1141 role="hg-ext-mq-cmd-qpop-opt">-a</option></command> and
bos@558 1142 then <command role="hg-cmd">hg qpush <option
steve@872 1143 role="hg-ext-mq-cmd-qpush-opt">-a</option></command> in
bos@558 1144 the underlying repository to see those changes show up there.
bos@558 1145 If you forget to do this, you can confuse MQ's idea of which
bos@558 1146 patches are applied.</para>
bos@558 1147
bos@558 1148 </sect2>
bos@558 1149 </sect1>
bos@558 1150 <sect1 id="sec:mq:tools">
bos@558 1151 <title>Third party tools for working with patches</title>
bos@558 1152
bos@584 1153 <para id="x_430">Once you've been working with patches for a while, you'll
bos@558 1154 find yourself hungry for tools that will help you to understand
bos@558 1155 and manipulate the patches you're dealing with.</para>
bos@558 1156
bos@584 1157 <para id="x_431">The <command>diffstat</command> command
bos@558 1158 <citation>web:diffstat</citation> generates a histogram of the
bos@558 1159 modifications made to each file in a patch. It provides a good
bos@559 1160 way to <quote>get a sense of</quote> a patch&emdash;which files
bos@559 1161 it affects, and how much change it introduces to each file and
bos@559 1162 as a whole. (I find that it's a good idea to use
bos@558 1163 <command>diffstat</command>'s <option
bos@558 1164 role="cmd-opt-diffstat">-p</option> option as a matter of
bos@558 1165 course, as otherwise it will try to do clever things with
bos@558 1166 prefixes of file names that inevitably confuse at least
bos@558 1167 me.)</para>
bos@558 1168
bos@567 1169 &interaction.mq.tools.tools;
bos@558 1170
bos@584 1171 <para id="x_432">The <literal role="package">patchutils</literal> package
bos@558 1172 <citation>web:patchutils</citation> is invaluable. It provides a
bos@558 1173 set of small utilities that follow the <quote>Unix
bos@558 1174 philosophy;</quote> each does one useful thing with a patch.
bos@558 1175 The <literal role="package">patchutils</literal> command I use
bos@558 1176 most is <command>filterdiff</command>, which extracts subsets
bos@558 1177 from a patch file. For example, given a patch that modifies
bos@558 1178 hundreds of files across dozens of directories, a single
bos@558 1179 invocation of <command>filterdiff</command> can generate a
bos@558 1180 smaller patch that only touches files whose names match a
bos@592 1181 particular glob pattern. See <xref
bos@558 1182 linkend="mq-collab:tips:interdiff"/> for another
bos@558 1183 example.</para>
bos@558 1184
bos@558 1185 </sect1>
bos@558 1186 <sect1>
bos@558 1187 <title>Good ways to work with patches</title>
bos@558 1188
bos@584 1189 <para id="x_433">Whether you are working on a patch series to submit to a
bos@558 1190 free software or open source project, or a series that you
bos@558 1191 intend to treat as a sequence of regular changesets when you're
bos@558 1192 done, you can use some simple techniques to keep your work well
bos@674 1193 organized.</para>
bos@558 1194
bos@584 1195 <para id="x_434">Give your patches descriptive names. A good name for a
bos@558 1196 patch might be <filename>rework-device-alloc.patch</filename>,
bos@558 1197 because it will immediately give you a hint what the purpose of
bos@558 1198 the patch is. Long names shouldn't be a problem; you won't be
bos@558 1199 typing the names often, but you <emphasis>will</emphasis> be
bos@558 1200 running commands like <command
bos@559 1201 role="hg-ext-mq">qapplied</command> and <command
bos@559 1202 role="hg-ext-mq">qtop</command> over and over. Good naming
bos@558 1203 becomes especially important when you have a number of patches
bos@558 1204 to work with, or if you are juggling a number of different tasks
bos@558 1205 and your patches only get a fraction of your attention.</para>
bos@558 1206
bos@584 1207 <para id="x_435">Be aware of what patch you're working on. Use the <command
bos@559 1208 role="hg-ext-mq">qtop</command> command and skim over the text
bos@558 1209 of your patches frequently&emdash;for example, using <command
bos@558 1210 role="hg-cmd">hg tip <option
bos@559 1211 role="hg-opt-tip">-p</option></command>)&emdash;to be sure
bos@559 1212 of where you stand. I have several times worked on and <command
bos@559 1213 role="hg-ext-mq">qrefresh</command>ed a patch other than the
bos@558 1214 one I intended, and it's often tricky to migrate changes into
bos@558 1215 the right patch after making them in the wrong one.</para>
bos@558 1216
bos@584 1217 <para id="x_436">For this reason, it is very much worth investing a little
bos@558 1218 time to learn how to use some of the third-party tools I
bos@592 1219 described in <xref linkend="sec:mq:tools"/>,
bos@559 1220 particularly
bos@559 1221 <command>diffstat</command> and <command>filterdiff</command>.
bos@559 1222 The former will give you a quick idea of what changes your patch
bos@559 1223 is making, while the latter makes it easy to splice hunks
bos@559 1224 selectively out of one patch and into another.</para>
bos@558 1225
bos@558 1226 </sect1>
bos@558 1227 <sect1>
bos@558 1228 <title>MQ cookbook</title>
bos@558 1229
bos@558 1230 <sect2>
bos@558 1231 <title>Manage <quote>trivial</quote> patches</title>
bos@558 1232
bos@584 1233 <para id="x_437">Because the overhead of dropping files into a new
bos@558 1234 Mercurial repository is so low, it makes a lot of sense to
bos@558 1235 manage patches this way even if you simply want to make a few
bos@558 1236 changes to a source tarball that you downloaded.</para>
bos@558 1237
bos@584 1238 <para id="x_438">Begin by downloading and unpacking the source tarball, and
bos@567 1239 turning it into a Mercurial repository.</para>
bos@567 1240
bos@567 1241 &interaction.mq.tarball.download;
bos@558 1242
bos@584 1243 <para id="x_439">Continue by creating a patch stack and making your
bos@567 1244 changes.</para>
bos@567 1245
bos@567 1246 &interaction.mq.tarball.qinit;
bos@558 1247
bos@584 1248 <para id="x_43a">Let's say a few weeks or months pass, and your package
bos@558 1249 author releases a new version. First, bring their changes
bos@567 1250 into the repository.</para>
bos@567 1251
bos@567 1252 &interaction.mq.tarball.newsource;
bos@567 1253
bos@584 1254 <para id="x_43b">The pipeline starting with <command role="hg-cmd">hg
bos@558 1255 locate</command> above deletes all files in the working
bos@558 1256 directory, so that <command role="hg-cmd">hg
bos@558 1257 commit</command>'s <option
bos@558 1258 role="hg-opt-commit">--addremove</option> option can
bos@558 1259 actually tell which files have really been removed in the
bos@558 1260 newer version of the source.</para>
bos@558 1261
bos@584 1262 <para id="x_43c">Finally, you can apply your patches on top of the new
bos@567 1263 tree.</para>
bos@567 1264
bos@567 1265 &interaction.mq.tarball.repush;
bos@682 1266 </sect2>
bos@682 1267
bos@558 1268 <sect2 id="sec:mq:combine">
bos@558 1269 <title>Combining entire patches</title>
bos@558 1270
bos@584 1271 <para id="x_43d">MQ provides a command, <command
bos@559 1272 role="hg-ext-mq">qfold</command> that lets you combine
bos@558 1273 entire patches. This <quote>folds</quote> the patches you
bos@558 1274 name, in the order you name them, into the topmost applied
bos@558 1275 patch, and concatenates their descriptions onto the end of its
bos@558 1276 description. The patches that you fold must be unapplied
bos@558 1277 before you fold them.</para>
bos@558 1278
bos@584 1279 <para id="x_43e">The order in which you fold patches matters. If your
bos@558 1280 topmost applied patch is <literal>foo</literal>, and you
bos@559 1281 <command role="hg-ext-mq">qfold</command>
bos@558 1282 <literal>bar</literal> and <literal>quux</literal> into it,
bos@558 1283 you will end up with a patch that has the same effect as if
bos@558 1284 you applied first <literal>foo</literal>, then
bos@558 1285 <literal>bar</literal>, followed by
bos@558 1286 <literal>quux</literal>.</para>
bos@682 1287 </sect2>
bos@682 1288
bos@558 1289 <sect2>
bos@558 1290 <title>Merging part of one patch into another</title>
bos@558 1291
bos@584 1292 <para id="x_43f">Merging <emphasis>part</emphasis> of one patch into
bos@558 1293 another is more difficult than combining entire
bos@558 1294 patches.</para>
bos@558 1295
bos@584 1296 <para id="x_440">If you want to move changes to entire files, you can use
bos@558 1297 <command>filterdiff</command>'s <option
bos@558 1298 role="cmd-opt-filterdiff">-i</option> and <option
bos@558 1299 role="cmd-opt-filterdiff">-x</option> options to choose the
bos@558 1300 modifications to snip out of one patch, concatenating its
bos@558 1301 output onto the end of the patch you want to merge into. You
bos@558 1302 usually won't need to modify the patch you've merged the
bos@558 1303 changes from. Instead, MQ will report some rejected hunks
bos@559 1304 when you <command role="hg-ext-mq">qpush</command> it (from
bos@558 1305 the hunks you moved into the other patch), and you can simply
bos@559 1306 <command role="hg-ext-mq">qrefresh</command> the patch to drop
bos@558 1307 the duplicate hunks.</para>
bos@558 1308
bos@584 1309 <para id="x_441">If you have a patch that has multiple hunks modifying a
bos@558 1310 file, and you only want to move a few of those hunks, the job
bos@558 1311 becomes more messy, but you can still partly automate it. Use
bos@558 1312 <command>lsdiff -nvv</command> to print some metadata about
bos@567 1313 the patch.</para>
bos@567 1314
bos@567 1315 &interaction.mq.tools.lsdiff;
bos@558 1316
bos@584 1317 <para id="x_442">This command prints three different kinds of
bos@558 1318 number:</para>
bos@558 1319 <itemizedlist>
bos@584 1320 <listitem><para id="x_443">(in the first column) a <emphasis>file
bos@558 1321 number</emphasis> to identify each file modified in the
bos@559 1322 patch;</para>
bos@559 1323 </listitem>
bos@584 1324 <listitem><para id="x_444">(on the next line, indented) the line number
bos@559 1325 within a modified file where a hunk starts; and</para>
bos@559 1326 </listitem>
bos@584 1327 <listitem><para id="x_445">(on the same line) a <emphasis>hunk
bos@559 1328 number</emphasis> to identify that hunk.</para>
bos@559 1329 </listitem></itemizedlist>
bos@558 1330
bos@584 1331 <para id="x_446">You'll have to use some visual inspection, and reading of
bos@558 1332 the patch, to identify the file and hunk numbers you'll want,
bos@558 1333 but you can then pass them to to
bos@558 1334 <command>filterdiff</command>'s <option
bos@558 1335 role="cmd-opt-filterdiff">--files</option> and <option
bos@558 1336 role="cmd-opt-filterdiff">--hunks</option> options, to
bos@558 1337 select exactly the file and hunk you want to extract.</para>
bos@558 1338
bos@584 1339 <para id="x_447">Once you have this hunk, you can concatenate it onto the
bos@558 1340 end of your destination patch and continue with the remainder
bos@592 1341 of <xref linkend="sec:mq:combine"/>.</para>
bos@558 1342
bos@558 1343 </sect2>
bos@558 1344 </sect1>
bos@558 1345 <sect1>
bos@558 1346 <title>Differences between quilt and MQ</title>
bos@558 1347
bos@584 1348 <para id="x_448">If you are already familiar with quilt, MQ provides a
bos@558 1349 similar command set. There are a few differences in the way
bos@558 1350 that it works.</para>
bos@558 1351
bos@584 1352 <para id="x_449">You will already have noticed that most quilt commands have
bos@558 1353 MQ counterparts that simply begin with a
bos@558 1354 <quote><literal>q</literal></quote>. The exceptions are quilt's
bos@558 1355 <literal>add</literal> and <literal>remove</literal> commands,
bos@558 1356 the counterparts for which are the normal Mercurial <command
bos@558 1357 role="hg-cmd">hg add</command> and <command role="hg-cmd">hg
bos@558 1358 remove</command> commands. There is no MQ equivalent of the
bos@558 1359 quilt <literal>edit</literal> command.</para>
bos@558 1360
bos@558 1361 </sect1>
bos@558 1362 </chapter>
bos@558 1363
bos@558 1364 <!--
bos@558 1365 local variables:
bos@558 1366 sgml-parent-document: ("00book.xml" "book" "chapter")
bos@558 1367 end:
bos@558 1368 -->