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-0.8, 2006-04
20
If you are already familiar with decentralized revision control, then
21
please feel free to skip ahead to "Introducing Yourself to Bazaar-NG". 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-NG.
39
Revision control systems (which I'll henceforth call RCS) such as
40
Bazaar-NG 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 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-NG'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-NG. 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-NG can do much more than just
93
storage and performing undo. For example, with Bazaar-NG developer can
94
take the modifications in one branch of software and apply them to
95
another, related, branch -- even if those changes exist in a branch owned
96
by somebody else. This allows developers to cooperate without giving write
99
Bazaar-NG 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-NG 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-NG
107
=================================
109
Bazaar-NG installs a single new command, **bzr**. Everything else is a
110
subcommand of this. You can get some help with `bzr help`. There will be
113
One function of a version control system is to keep track of who changed
114
what. In a decentralized system, that requires an identifier for each
115
author that is globally unique. Most people already have one of these: an
116
email address. Bzr is smart enough to automatically generate an email
117
address by looking up your username and hostname. If you don't like the
118
guess that Bazaar-NG makes, then three options exist:
120
1. Set an email address via ``bzr whoami``. This is the simplest way.
121
To set a global identity, use::
123
% bzr whoami 'Your Name <email@example.com>'
125
If you'd like to use a different address for a specific branch, enter
126
the branch folder and use::
128
% bzr whoami --branch 'Your Name <email@example.com>'
130
1. Setting the email address in the
131
``~/.bazaar/bazaar.conf`` by adding the following lines. Please note that
132
``[DEFAULT]`` is case sensitive::
135
email= Your Name <email@isp.com>
137
As above, you can override this settings on a branch by branch basis by
138
creating a branch section in ``~/.bazaar/locations.conf`` and adding the
141
[/the/directory/to/the/branch]
142
email=Your Name <email@isp.com>
144
1. Overriding the two previous options by setting the global environment
145
variable ``$BZREMAIL`` or ``$EMAIL`` (``$BZREMAIL`` will take precedence)
146
to your full email address.
151
History is by default stored in the .bzr directory of the branch. There
152
will be a facility to store it in a separate repository, which may be
153
remote. We create a new branch by running **bzr init** in an existing
161
/home/mbp/work/bzr.test/tutorial
168
As for CVS, there are three classes of file: unknown, ignored, and
169
versioned. The **add** command makes a file versioned: that is, changes
170
to it will be recorded by the system::
172
% echo 'hello world' > hello.txt
183
If you add the wrong file, simply use **bzr remove** to make it
184
unversioned again. This does not delete the working copy.
189
All history is stored in a branch, which is just an on-disk directory
190
containing control files. By default there is no separate repository or
191
database as used in svn or svk. You can choose to create a repository if
192
you want to (see the **bzr init-repo** command). You may wish to do this
193
if you have very large branches, or many branches of a moderate sized
196
You'll usually refer to branches on your computer's filesystem just by
197
giving the name of the directory containing the branch. bzr also supports
198
accessing branches over http, for example::
200
% bzr log http://bazaar-ng.org/bzr/bzr.dev/
202
By installing bzr plugins you can also access branches over the sftp or
208
Once you have completed some work, you will want to **commit** it to the
209
version history. It is good to commit fairly often: whenever you get a
210
new feature working, fix a bug, or improve some code or documentation.
211
It's also a good practice to make sure that the code compiles and passes
212
its test suite before committing, to make sure that every revision is a
213
known-good state. You can also review your changes, to make sure you're
214
committing what you intend to, and as a chance to rethink your work before
215
you permanently record it.
217
Two bzr commands are particularly useful here: **status** and **diff**.
222
The **status** command tells you what changes have been made to the
223
working directory since the last revision::
229
By default **bzr status** hides "boring" files that are either unchanged
230
or ignored. To see them too, use the --all option. The status command
231
can optionally be given the name of some files or directories to check.
236
The **diff** command shows the full text of changes to all files as a
237
standard unified diff. This can be piped through many programs such as
238
''patch'', ''diffstat'', ''filterdiff'' and ''colordiff''::
241
*** added file 'hello.txt'
248
With the ''-r'' option, the tree is compared to an earlier revision, or
249
the differences between two versions are shown::
251
% bzr diff -r 1000.. # everything since r1000
252
% bzr diff -r 1000..1100 # changes from 1000 to 1100
254
The --diff-options option causes bzr to run the external diff program,
255
passing options. For example::
257
% bzr diff --diff-options --side-by-side foo
259
Some projects prefer patches to show a prefix at the start of the path for
260
old and new files. The --prefix option can be used to provide such a prefix.
261
As a shortcut, ``bzr diff -p1`` produces a form that works with the
262
command ``patch -p1``.
267
When the working tree state is satisfactory, it can be **committed** to
268
the branch, creating a new revision holding a snapshot of that state.
273
The **commit** command takes a message describing the changes in the
274
revision. It also records your userid, the current time and timezone, and
275
the inventory and contents of the tree. The commit message is specified
276
by the ''-m'' or ''--message'' option. You can enter a multi-line commit
277
message; in most shells you can enter this just by leaving the quotes open
278
at the end of the line.
282
% bzr commit -m "added my first file"
284
You can also use the -F option to take the message from a file. Some
285
people like to make notes for a commit message while they work, then
286
review the diff to make sure they did what they said they did. (This file
287
can also be useful when you pick up your work after a break.)
289
Message from an editor
290
======================
292
If you use neither the `-m` nor the `-F` option then bzr will open an
293
editor for you to enter a message. The editor to run is controlled by
294
your `$EDITOR` environment variable or
295
add `editor` to ~/.bazaar/bazaar.conf; `$BZR_EDITOR` will override
296
the above mentioned editor options. If you quit the editor without
297
making any changes, the commit will be cancelled.
302
If you give file or directory names on the commit command line then only
303
the changes to those files will be committed. For example::
305
% bzr commit -m "documentation fix" commit.py
307
By default bzr always commits all changes to the tree, even if run from a
308
subdirectory. To commit from only the current directory down, use::
313
Removing uncommitted changes
314
============================
316
If you've made some changes and don't want to keep them, use the
317
**revert** command to go back to the previous head version. It's a good
318
idea to use **bzr diff** first to see what will be removed. By default the
319
revert command reverts the whole tree; if file or directory names are
320
given then only those ones will be affected. **revert** also clears the
321
list of pending merges revisions.
326
Many source trees contain some files that do not need to be versioned,
327
such as editor backups, object or bytecode files, and built programs. You
328
can simply not add them, but then they'll always crop up as unknown files.
329
You can also tell bzr to ignore these files by adding them to a file
330
called ''.bzrignore'' at the top of the tree.
332
This file contains a list of file wildcards (or "globs"), one per line.
333
Typical contents are like this::
340
If a glob contains a slash, it is matched against the whole path from the
341
top of the tree; otherwise it is matched against only the filename. So
342
the previous example ignores files with extension ``.o`` in all
343
subdirectories, but this example ignores only config.h at the top level
344
and HTML files in ``doc/``::
349
To get a list of which files are ignored and what pattern they matched,
350
use ''bzr ignored''::
356
It is OK to have either an ignore pattern match a versioned file, or to
357
add an ignored file. Ignore patterns have no effect on versioned files;
358
they only determine whether unversioned files are reported as unknown or
361
The ''.bzrignore'' file should normally be versioned, so that new copies
362
of the branch see the same patterns::
365
% bzr commit -m "Add ignore patterns"
374
The **bzr log** command shows a list of previous revisions. The **bzr log
375
--forward** command does the same in chronological order to get most
376
recent revisions printed at last.
378
As with bzr diff, bzr log supports the -r argument::
380
% bzr log -r 1000.. # Revision 1000 and everything after it
381
% bzr log -r ..1000 # Everything up to and including r1000
382
% bzr log -r 1000..1100 # changes from 1000 to 1100
383
% bzr log -r 1000 # The changes in only revision 1000
389
The **bzr info** command shows some summary information about the working
390
tree and the branch history.
393
Versioning directories
394
======================
396
bzr versions files and directories in a way that can keep track of renames
397
and intelligently merge them::
400
% echo 'int main() {}' > src/simple.c
410
Deleting and removing files
411
===========================
413
You can delete files or directories by just deleting them from the working
414
directory. This is a bit different to CVS, which requires that you also
417
**bzr remove** makes the file un-versioned, but does not delete
418
the working copy. This is useful when you add the wrong file, or decide
419
that a file should actually not be versioned.
424
% bzr remove -v hello.txt
434
If you remove the wrong file by accident, you can use **bzr revert** to
441
Often rather than starting your own project, you will want to submit a
442
change to an existing project. You can get a copy of an existing branch
443
by copying its directory, expanding a tarball, or by a remote copy using
444
something like rsync. You can also use bzr to fetch a copy. Because this
445
new copy is potentially a new branch, the command is called *branch*::
447
% bzr branch http://bazaar-ng.org/bzr/bzr.dev
450
This copies down the complete history of this branch, so we can do all
451
operations on it locally: log, annotate, making and merging branches.
452
There will be an option to get only part of the history if you wish.
454
Following upstream changes
455
==========================
457
You can stay up-to-date with the parent branch by "pulling" in their
462
After this change, the local directory will be a mirror of the source. This
463
includes the ''revision-history'' - which is a list of the commits done in
464
this branch, rather than merged from other branches.
466
This command only works if your local (destination) branch is either an
467
older copy of the parent branch with no new commits of its own, or if the
468
most recent commit in your local branch has been merged into the parent
471
Merging from related branches
472
=============================
474
If two branches have diverged (both have unique changes) then **bzr
475
merge** is the appropriate command to use. Merge will automatically
476
calculate the changes that exist in the branch you're merging from that
477
are not in your branch and attempt to apply them in your branch.
484
If there is a conflict during a merge, 3 files with the same basename are
485
created. The filename of the common base is appended with .BASE, the
486
filename of the file containing your changes is appended .THIS and the
487
filename with the changes from the other tree is appended .OTHER.
488
Using a program such as kdiff3, you can now comfortably merge them into
489
one file. To commit you have to rename it to the original basename and
490
delete the other two files. As long as there exist files with .BASE, .THIS
491
or .OTHER the commit command will complain.
493
[**TODO**: explain conflict markers within files]
496
Publishing your branch
497
======================
499
You don't need a special server to publish a bzr branch, just a normal web
500
server. Just mirror the files to your server, including the .bzr
501
directory. One can push a branch (or the changes for a branch) by one of
502
the following three methods:
504
* Rsync: rsync -avrz LOCALBRANCH servername.com/this/directory/here
506
(or any other tool for publishing a directory to a web site.)
508
* bzr push sftp://servername.com/this/directory/here
510
(The directory that must already exist)
512
* The rspush plugin that comes with BzrTools
added
removed
doc/tutorial.txt
