5
current for bzr 0.0.6pre, July 2005
8
*NOTE* For a more current and user-editable version of this
9
document, see the wiki at http://bazaar.canonical.com/IntroductionToBzr
1
.. This file is in Python ReStructuredText format - it can be formatted
2
.. into HTML or text. In the future we plan to extract the example commands
3
.. and automatically test them.
5
.. This text was previously on the wiki at
6
.. http://bazaar.canonical.com/IntroductionToBzr
7
.. but has been moved into the source tree so it can be kept in sync with
8
.. the source and possibly automatically checked.
14
Current for bzr-0.16, 2007-05
15
Bazaar-NG is a version control tool. It manages trees of files and subdirectories. In particular, it records *revisions* of trees, representing their state at a particular point in time, and information about those revisions and their relationships. Recording and retrieving tree revisions is useful in several ways if you are writing software or documents or doing similar creative work.
17
* Keeping previous revisions lets you go back if you make a mistake or want to check your work. It acts as a high-level unlimited undo.
19
* By recording comments on every revision, you produce an annotated history of the project, describing what, who, why, and when.
21
* Using a version control tool can be an aid to thinking about a project: getting to a stable state at regular intervals and then writing a description of what you did is an easy way to stay organized and on track.
23
Bazaar-NG remembers the *ancestry* of a revision: the previous revisions that it is based upon. A single revision may have more than one direct descendant, each with different changes, representing a divergence in the evolution of the tree.
24
By branching, Bazaar-NG allows multiple people to cooperate on the evolution of a project, without all needing to work in strict
25
lock-step. Branching can be useful even for a single developer.
27
Bazaar-NG installs a single new command,
28
*bzr*. Everything else is a subcommand of this. You can get
29
some help with ``bzr help``. There will be more in the future.
33
Introducing yourself to bzr
34
===========================
36
One function of a version control system is to keep track of who
37
changed what. In a distributed system that requires an identifier for
38
each author that is globally unique. Most people already have one of
39
these: an email address.
41
[after 0.0.4] To tell bzr which email address to use, put it in the file
42
``$HOME/.bzr.conf/email``, or the environment variable ``$BZREMAIL``.
43
If neither of these are set, bzr will use the ``$EMAIL``
44
variable, or use your username and hostname.
46
To check this has taken effect, or if you forget your own name, use
47
the ``whoami`` ("who am i?") command::
51
Some people want to avoid sharing their email address so as not to
52
get spam. bzr will never
53
disclose your email address unless you tell it to by publishing an
54
archive or transmitting a changeset. It's recommended that you do use
55
a real address, so that people can contact you about your work, but
56
it's not required. You can use an address which is obfuscated, which
57
bounces, or which goes through an anti-spam service such as spamgourmet.com.
20
If you are already familiar with decentralized revision control, then
21
please feel free to skip ahead to "Introducing Yourself to Bazaar". If,
22
on the other hand, you are familiar with revision control but not
23
decentralized revision control, then please start at "How DRCS is
24
different." Otherwise, get some coffee or tea, get comfortable and get
27
The Purposes of Revision Control
28
================================
30
Odds are that you have worked on some sort of textual data -- the sources
31
to a program, web sites or the config files that Unix system
32
administrators have to deal with in /etc. The chances are also good that
33
you have made some sort of mistake that you deeply regretted. Perhaps you
34
deleted the configuration file for your mailserver or perhaps mauled the
35
source code for a pet project. Whatever happened, you have just deleted
36
important information that you would desperately like to get back. If this
37
has ever happened to you, then you are probably ready for Bazaar.
39
Revision control systems (which I'll henceforth call RCS) such as
40
Bazaar give you the ability to track changes for a directory by turning
41
it into something slightly more complicated than a directory that we call
42
a **branch**. The branch not only stores how the directory looks right
43
now, but also how it looked at various points in the past. Then, when you
44
do something you wish you hadn't, you can restore the directory to the way
45
it looked at some point in the past.
47
Revision control systems give users the ability to save changes to a
48
branch by "committing a **revision**". The revision created is essentially
49
a summary of the changes that were made since the last time the tree was
52
These revisions have other uses as well. For example, one can comment
53
revisions to record what the recent set of changes meant by providing an
54
optional log message. Real life log messages include things like "Fixed
55
the web template to close the table" and "Added sftp suppport. Fixes #595"
57
We keep these logs so that if later there is some sort of problem with
58
sftp, we can figure out when the problem probably happened.
63
Many Revision Control Systems (RCS) are stored on servers. If one wants to
64
work on the code stored within an RCS, then one needs to connect to the
65
server and "checkout" the code. Doing so gives one a directory in which a
66
person can make changes and then commit. The RCS client then connects to
67
the RCS server and stores the changes. This method is known as the
70
The centralized model can have some drawbacks. A centralized RCS requires
71
that one is able to connect to the server whenever one wants to do version
72
control work. This can be a bit of a problem if your server is on some other
73
machine on the internet and you are not. Or, worse yet, you **are** on the
74
internet but the server is missing!
76
Decentralized Revision Control Systems (which I'll call DRCS after this
77
point) deal with this problem by keeping branches on the same machine as
78
the client. In Bazaar's case, the branch is kept in the same place as
79
the code that is being version controlled. This allows the user to save
80
his changes (**commit**) whenever he wants -- even if he is offline. The
81
user only needs internet access when he wants to access the changes in
82
someone else's branch that are somewhere else.
85
A common requirement that many people have is the need to keep track of
86
the changes for a directory such as file and subdirectory changes.
87
Performing this tracking by hand is a awkward process that over time
88
becomes unwieldy. That is, until one considers version control tools such
89
as Bazaar. These tools automate the process of storing data by creating
90
a **revision** of the directory tree whenever the user asks.
92
Revision control software such as Bazaar can do much more than just
93
storage and performing undo. For example, with Bazaar a developer can
94
take the modifications in one branch of software and apply them to a
95
related branch -- even if those changes exist in a branch owned by
96
somebody else. This allows developers to cooperate without giving
97
write access to the repository.
99
Bazaar remembers the ''ancestry'' of a revision: the previous revisions
100
that it is based upon. A single revision may have more than one direct
101
descendant, each with different changes, representing a divergence in the
102
evolution of the tree. By branching, Bazaar allows multiple people to
103
cooperate on the evolution of a project, without all needing to work in
104
strict lock-step. Branching can be useful even for a single developer.
106
Introducing yourself to Bazaar
107
==============================
109
Bazaar installs a single new command, **bzr**. Everything else is a
110
subcommand of this. You can get some help with ``bzr help``. Some arguments
111
are grouped in topics: ``bzr help topics`` to see which topics are available.
112
There will be more in the future.
114
One function of a version control system is to keep track of who changed
115
what. In a decentralized system, that requires an identifier for each
116
author that is globally unique. Most people already have one of these: an
117
email address. Bzr is smart enough to automatically generate an email
118
address by looking up your username and hostname. If you don't like the
119
guess that Bazaar makes, then three options exist:
121
1. Set an email address via ``bzr whoami``. This is the simplest way.
123
To set a global identity, use::
125
% bzr whoami "Your Name <email@example.com>"
127
If you'd like to use a different address for a specific branch, enter
128
the branch folder and use::
130
% bzr whoami --branch "Your Name <email@example.com>"
132
#. Setting the email address in the ``~/.bazaar/bazaar.conf`` [1]_ by
133
adding the following lines. Please note that ``[DEFAULT]`` is case
137
email=Your Name <email@isp.com>
139
As above, you can override this settings on a branch by branch basis
140
by creating a branch section in ``~/.bazaar/locations.conf`` and
141
adding the following lines::
143
[/the/path/to/the/branch]
144
email=Your Name <email@isp.com>
147
#. Overriding the two previous options by setting the global environment
148
variable ``$BZR_EMAIL`` or ``$EMAIL`` (``$BZR_EMAIL`` will take
149
precedence) to your full email address.
151
.. [1] On Windows, the users configuration files can be found in the
152
application data directory. So instead of ``~/.bazaar/branch.conf``
153
the configuration file can be found as:
154
``C:\Documents and Settings\<username>\Application Data\Bazaar\2.0\branch.conf``.
155
The same is true for ``locations.conf``, ``ignore``, and the
156
``plugins`` directory.
65
History is by default stored in the .bzr directory of the branch.
66
There will be a facility to store it in a separate repository, which
67
may be remote. We create a new branch by running *bzr init* in
68
an existing directory::
161
History is by default stored in the .bzr directory of the branch. In a
162
future version of Bazaar, there will be a facility to store it in a
163
separate repository, which may be remote. We create a new branch by
164
running ``bzr init`` in an existing directory::
76
171
/home/mbp/work/bzr.test/tutorial
83
As for CVS, there are three classes of file: unknown, ignored, and
84
versioned. The *add* command makes a file versioned: that is,
85
changes to it will be recorded by the system::
178
As with CVS, there are three classes of file: unknown, ignored, and
179
versioned. The **add** command makes a file versioned: that is, changes
180
to it will be recorded by the system::
87
182
% echo 'hello world' > hello.txt
92
% bzr add -v hello.txt
97
If you add the wrong file, simply use ``bzr remove`` to make
98
it unversioned again. This does not delete the working copy.
193
If you add the wrong file, simply use ``bzr remove`` to make it
194
unversioned again. This does not delete the working copy in this case,
195
though it may in others [2]_.
197
.. [2] ``bzr remove`` will remove the working copy if it is currently
198
versioned, but has no changes from the last committed version. You
199
can force the file to always be kept with the ``--keep`` option to
200
``bzr remove``, or force it to always be deleted with ``--force``.
205
All history is stored in a branch, which is just an on-disk directory
206
containing control files. By default there is no separate repository or
207
database as used in svn or svk. You can choose to create a repository if
208
you want to (see the ``bzr init-repo`` command). You may wish to do this
209
if you have very large branches, or many branches of a moderately sized
212
You'll usually refer to branches on your computer's filesystem just by
213
giving the name of the directory containing the branch. bzr also supports
214
accessing branches over http and sftp, for example::
216
% bzr log http://bazaar-vcs.org/bzr/bzr.dev/
217
% bzr log sftp://bazaar-vcs.org/bzr/bzr.dev/
219
By installing bzr plugins you can also access branches using the rsync
222
See the `Publishing your branch`_ section for more about how to put your
223
branch at a given location.
101
225
Reviewing changes
102
226
=================
104
Once you have completed some work, you will want to *commit*
105
it to the version history. It is good to commit fairly often:
106
whenever you get a new feature working, fix a bug, or improve some
107
code or documentation. It's also a good practice to make sure that
108
the code compiles and passes its test suite before committing, to make
109
sure that every revision is a known-good state. You can also review
110
your changes, to make sure you're committing what you intend to, and
111
as a chance to rethink your work before you permanently record it.
113
Two bzr commands are particularly useful here: *status* and
114
*diff*. The *status* command
115
shows a listing with one line per file, indicating whether it has been
116
Added, Deleted, Modified, or Renamed in the current revision. Unknown
117
files are shown as '?'. With the ``--all`` option, the status
118
command also shows unmodified versioned files as '.', and ignored
228
Once you have completed some work, you will want to **commit** it to the
229
version history. It is good to commit fairly often: whenever you get a
230
new feature working, fix a bug, or improve some code or documentation.
231
It's also a good practice to make sure that the code compiles and passes
232
its test suite before committing, to make sure that every revision is a
233
known-good state. You can also review your changes, to make sure you're
234
committing what you intend to, and as a chance to rethink your work before
235
you permanently record it.
237
Two bzr commands are particularly useful here: **status** and **diff**.
242
The **status** command tells you what changes have been made to the
243
working directory since the last revision::
124
The *diff* command shows the full text of changes to all
125
files as a standard unified diff. This can be piped through many
126
programs such as ``patch``, ``diffstat``,
127
``filterdiff`` and ``colordiff``::
249
``bzr status`` hides "boring" files that are either unchanged or ignored.
250
The status command can optionally be given the name of some files or
251
directories to check.
256
The **diff** command shows the full text of changes to all files as a
257
standard unified diff. This can be piped through many programs such as
258
''patch'', ''diffstat'', ''filterdiff'' and ''colordiff''::
130
*** added file 'hello.txt'
261
=== added file 'hello.txt'
262
--- hello.txt 1970-01-01 00:00:00 +0000
263
+++ hello.txt 2005-10-18 14:23:29 +0000
136
With the ``-r`` option, the tree is compared to an earlier
139
[TODO: options to run external diff; to get context diff or other
140
formats; to diff only selected files; to compare two historical
268
With the ``-r`` option, the tree is compared to an earlier revision, or
269
the differences between two versions are shown::
271
% bzr diff -r 1000.. # everything since r1000
272
% bzr diff -r 1000..1100 # changes from 1000 to 1100
274
The ``--diff-options`` option causes bzr to run the external diff program,
275
passing options. For example::
277
% bzr diff --diff-options --side-by-side foo
279
Some projects prefer patches to show a prefix at the start of the path
280
for old and new files. The ``--prefix`` option can be used to provide
282
As a shortcut, ``bzr diff -p1`` produces a form that works with the
283
command ``patch -p1``.
145
285
Committing changes
146
286
==================
148
When the working tree state is satisfactory, it can be
149
*committed* to the branch, creating a new revision holding a
150
snapshot of that state.
152
The ``commit`` command takes a message describing the changes
153
in the revision. It also records your userid, the current time and
154
timezone, and the inventory and contents of the tree. The commit
155
message is specified by the ``-m`` or ``--message`` option.
156
You can enter a multi-line commit message; in most shells you can
157
enter this just by leaving the quotes open at the end of the line. ::
288
When the working tree state is satisfactory, it can be **committed** to
289
the branch, creating a new revision holding a snapshot of that state.
294
The **commit** command takes a message describing the changes in the
295
revision. It also records your userid, the current time and timezone, and
296
the inventory and contents of the tree. The commit message is specified
297
by the ``-m`` or ``--message`` option. You can enter a multi-line commit
298
message; in most shells you can enter this just by leaving the quotes open
299
at the end of the line.
159
303
% bzr commit -m "added my first file"
161
[TODO: commit message interactively, through an editor or from a
164
[TODO: commit only selected files, including renamed/added/deleted
305
You can also use the ``-F`` option to take the message from a file. Some
306
people like to make notes for a commit message while they work, then
307
review the diff to make sure they did what they said they did. (This file
308
can also be useful when you pick up your work after a break.)
310
Message from an editor
311
======================
313
If you use neither the ``-m`` nor the ``-F`` option then bzr will open an
314
editor for you to enter a message. The editor to run is controlled by
315
your ``$VISUAL`` or ``$EDITOR`` environment variable, which can be overridden
316
by the ``editor`` setting in ``~/.bazaar/bazaar.conf``; ``$BZR_EDITOR`` will
317
override either of the above mentioned editor options. If you quit the
318
editor without making any changes, the commit will be cancelled.
323
If you give file or directory names on the commit command line then only
324
the changes to those files will be committed. For example::
326
% bzr commit -m "documentation fix" commit.py
328
By default bzr always commits all changes to the tree, even if run from a
329
subdirectory. To commit from only the current directory down, use::
169
334
Removing uncommitted changes
170
335
============================
172
337
If you've made some changes and don't want to keep them, use the
173
``revert`` command to go back to the previous head version. It's a
174
good idea to use ``bzr diff`` first to see what will be removed.
175
By default the revert command reverts the whole tree; if file or
176
directory names are given then only those ones will be affected.
177
revert also clears the list of pending merges revisions.
338
**revert** command to go back to the previous head version. It's a good
339
idea to use ``bzr diff`` first to see what will be removed. By default the
340
revert command reverts the whole tree; if file or directory names are
341
given then only those ones will be affected. ``bzr revert`` also clears the
342
list of pending merges revisions.
186
Many source trees contain some files that do not need to be
187
versioned, such as editor backups, object or bytecode files, and built
188
programs. You can simply not add them, but then they'll always crop
189
up as unknown files. You can also tell bzr to ignore these files by
190
adding them to a file called ``.bzrignore`` at the top of the
347
Many source trees contain some files that do not need to be versioned,
348
such as editor backups, object or bytecode files, and built programs. You
349
can simply not add them, but then they'll always crop up as unknown files.
350
You can also tell bzr to ignore these files by adding them to a file
351
called ``.bzrignore`` at the top of the tree.
193
This file contains a list of file wildcards (or "globs"), one
194
per line. Typical contents are like this::
353
This file contains a list of file wildcards (or "globs"), one per line.
354
Typical contents are like this::
201
If a glob contains a slash, it is matched against the whole path
202
from the top of the tree; otherwise it is matched against only the
203
filename. So the previous example ignores ``*.o`` in all
204
subdirectories, but this example ignores only config.h at the top
205
level and HTML files in ``doc/``::
361
If a glob contains a slash, it is matched against the whole path from the
362
top of the tree; otherwise it is matched against only the filename. So
363
the previous example ignores files with extension ``.o`` in all
364
subdirectories, but this example ignores only ``config.h`` at the top level
365
and HTML files in ``doc/``::
210
To get a list of which files are ignored and what pattern they matched, use ``bzr ignored``::
370
To get a list of which files are ignored and what pattern they matched,
371
use ``bzr ignored``::
216
It is OK to have an ignore pattern match a versioned file, or to
217
add an ignored file. Ignore patterns have no effect on versioned
218
files; they only determine whether unversioned files are reported as
221
The ``.bzrignore`` file should normally be versioned, so that new
222
copies of the branch see the same patterns::
377
It is OK to have either an ignore pattern match a versioned file, or to
378
add an ignored file. Ignore patterns have no effect on versioned files;
379
they only determine whether unversioned files are reported as unknown or
382
The ``.bzrignore`` file should normally be versioned, so that new copies
383
of the branch see the same patterns::
224
385
% bzr add .bzrignore
225
386
% bzr commit -m "Add ignore patterns"
392
There are some ignored files which are not project specific, but more user
393
specific. Things like editor temporary files, or personal temporary files.
394
Rather than add these ignores to every project, bzr supports a global
395
ignore file in ``~/.bazaar/ignore`` [1]_. It has the same syntax as the
396
per-project ignore file.
228
399
Examining history
229
400
=================
234
The ``log`` command shows a list of previous revisions.
405
The ``bzr log`` command shows a list of previous revisions. The ``bzr log
406
--forward`` command does the same in chronological order to get most
407
recent revisions printed at last.
409
As with ``bzr diff``, ``bzr log`` supports the ``-r`` argument::
411
% bzr log -r 1000.. # Revision 1000 and everything after it
412
% bzr log -r ..1000 # Everything up to and including r1000
413
% bzr log -r 1000..1100 # changes from 1000 to 1100
414
% bzr log -r 1000 # The changes in only revision 1000
237
417
Branch statistics
238
418
=================
240
The ``bzr info`` command shows some summary information about
241
the working tree and the branch history.
420
The ``bzr info`` command shows some summary information about the working
421
tree and the branch history.
244
424
Versioning directories
245
425
======================
247
bzr versions files and directories in a way that can keep track of
248
renames and intelligently merge them::
427
bzr versions files and directories in a way that can keep track of renames
428
and intelligently merge them::
251
431
% echo 'int main() {}' > src/simple.c
256
% bzr add src/simple.c
262
441
Deleting and removing files
263
442
===========================
265
You can delete files or directories by just deleting them from the
266
working directory. This is a bit different to CVS, which requires
267
that you also do *cvs remove*.
269
*bzr remove* makes the file un-versioned, but does not
270
delete the working copy. This is useful when you add the wrong file,
271
or decide that a file should actually not be versioned. ::
444
You can delete files or directories by just deleting them from the working
445
directory. This is a bit different to CVS, which requires that you also
448
``bzr remove`` makes the file un-versioned, but may or may not delete
449
the working copy [2]_. This is useful when you add the wrong file,
450
or decide that a file should actually not be versioned.
274
455
% bzr remove -v hello.txt
465
If you remove the wrong file by accident, you can use ``bzr revert`` to
285
Often rather than starting your own project, you will want to
286
submit a change to an existing project. You can get a copy of an
287
existing branch by copying its directory, expanding a tarball, or by a
288
remote copy using something like rsync. You can also use bzr to fetch
289
a copy. Because this new copy is potentially a new branch, the
290
command is called *branch*::
472
Often rather than starting your own project, you will want to submit a
473
change to an existing project. To do this, you'll need to get a copy of
474
the existing branch. Because this new copy is potentially a new branch,
475
the command is called **branch**::
292
% bzr branch http://bazaar-ng.org/bzr/bzr.dev
477
% bzr branch http://bazaar-vcs.org/bzr/bzr.dev
295
This copies down the complete history of this branch, so we can
296
do all operations on it locally: log, annotate, making and merging
297
branches. There will be an option to get only part of the history if
480
This copies down the complete history of this branch, so we can do all
481
operations on it locally: log, annotate, making and merging branches.
482
There will be an option to get only part of the history if you wish.
484
You can also get a copy of an existing branch by copying its directory,
485
expanding a tarball, or by a remote copy using something like rsync.
302
487
Following upstream changes
303
488
==========================
305
You can stay up-to-date with the parent branch by *pulling*
490
You can stay up-to-date with the parent branch by "pulling" in their
310
This only works if your local branch includes only changes from the
311
parent branch. Otherwise, the branches are said to have *diverged*,
312
and they must be merged instead.
495
After this change, the local directory will be a mirror of the source. This
496
includes the ''revision-history'' - which is a list of the commits done in
497
this branch, rather than merged from other branches.
499
This command only works if your local (destination) branch is either an
500
older copy of the parent branch with no new commits of its own, or if the
501
most recent commit in your local branch has been merged into the parent
504
Merging from related branches
505
=============================
507
If two branches have diverged (both have unique changes) then ``bzr
508
merge`` is the appropriate command to use. Merge will automatically
509
calculate the changes that exist in the branch you're merging from that
510
are not in your branch and attempt to apply them in your branch.
517
If there is a conflict during a merge, 3 files with the same basename
518
are created. The filename of the common base is appended with ".BASE",
519
the filename of the file containing your changes is appended with
520
".THIS" and the filename with the changes from the other tree is
521
appended with ".OTHER". Using a program such as kdiff3, you can now
522
comfortably merge them into one file. In order to commit you have to
523
rename the merged file (".THIS") to the original file name. To
524
complete the conflict resolution you must use the resolve command,
525
which will remove the ".OTHER" and ".BASE" files. As long as there
526
exist files with .BASE, .THIS or .OTHER the commit command will
531
% kdiff3 file.BASE file.OTHER file.THIS
535
[**TODO**: explain conflict markers within files]
538
Publishing your branch
539
======================
541
You don't need a special server to publish a bzr branch, just a normal web
542
server. Just mirror the files to your server, including the .bzr
543
directory. One can push a branch (or the changes for a branch) by one of
544
the following three methods:
546
* The best method is to use bzr itself to do it.
550
% bzr push sftp://servername.com/path/to/directory
552
(The destination directory must already exist unless the
553
``--create-prefix`` option is used.)
555
* Another option is the ``rspush`` plugin that comes with BzrTools, which
556
uses rsync to push the changes to the revision history and the working
559
You can also use copy the files around manually, by sending a tarball, or
560
using rsync, or other related file transfer methods. This is usually
561
less safe than using ``push``, but may be faster or easier in some
564
Moving changes between trees
565
============================
567
It happens to the best of us: sometimes you'll make changes in the wrong
568
tree. Maybe because you've accidentally started work in the wrong directory,
569
maybe because as you're working, the change turns out to be bigger than you
570
expected, so you start a new branch for it.
572
To move your changes from one tree to another, use
577
% bzr merge --uncommitted OLDDIR
579
This will apply all of the uncommitted changes you made in OLDDIR to NEWDIR.
580
It will not apply committed changes, even if they could be applied to NEWDIR
581
with a regular merge. The changes will remain in OLDDIR, but you can use ``bzr
582
revert OLDDIR`` to remove them, once you're satisfied with NEWDIR.
584
NEWDIR does not have to be a copy of OLDDIR, but they should be related.
585
The more different they are, the greater the chance of conflicts.
added
removed
doc/tutorial.txt
