3
3
.. and automatically test them.
5
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
6
.. http://bazaar.canonical.com/IntroductionToBzr but has been moved into the
7
.. source tree so it can be kept in sync with
8
8
.. the source and possibly automatically checked.
14
Current for bzr-0.7pre, 2006-01-06.
18
If you are already familiar with decentralized version control, then
19
please feel free to skip ahead to "Introducing Yourself to Bazaar". If,
20
on the other hand, you are familiar with version control but not
21
decentralized version control, then please start at "How DVCS is
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
22
24
different." Otherwise, get some coffee or tea, get comfortable and get
25
The purpose of version control
26
==============================
27
The Purposes of Revision Control
28
================================
28
30
Odds are that you have worked on some sort of textual data -- the sources
29
to a program, web sites or the config files that Unix system
30
administrators have to deal with in /etc. The chances are also good that
31
you have made some sort of mistake that you deeply regretted. Perhaps you
32
deleted the configuration file for your mailserver or perhaps mauled the
33
source code for a pet project. Whatever happened, you have just deleted
34
important information that you would desperately like to get back. If this
35
has ever happened to you, then you are probably ready for Bazaar.
37
Version control systems (which I'll henceforth call VCS) such as
38
Bazaar give you the ability to track changes for a directory by turning
39
it into something slightly more complicated than a directory that we call
40
a **branch**. The branch not only stores how the directory looks right
41
now, but also how it looked at various points in the past. Then, when you
42
do something you wish you hadn't, you can restore the directory to the way
43
it looked at some point in the past.
45
Version control systems give users the ability to save changes to a
46
branch by "committing a **revision**". The revision created is essentially
47
a summary of the changes that were made since the last time the tree was
31
to a program, web sites or the config files that Unix system administrators
32
have to deal with in /etc. The chances are also good that you have made
33
some sort of mistake that you deeply regretted. Perhaps you deleted the
34
configuration file for your mailserver or perhaps mauled the source code
35
for a pet project. Whatever happened, you have just deleted important
36
information that you would desperately like to get back. If this has ever
37
happened to you, then you are probably ready for Bazaar-NG.
39
Revision control systems (which I'll henceforth call RCS) such as Bazaar-NG
40
give you the ability to track changes for a directory by turning it into
41
something slightly more complicated than a directory that we call a
42
**branch**. The branch not only stores how the directory looks right now,
43
but also how it looked at various points in the past. Then, when you do
44
something you wish you hadn't, you can restore the directory to the way it
45
looked at some point in the past.
47
Revision control systems give users the ability to save changes to a branch
48
by "committing a **revision**". The revision created is essentially a
49
summary of the changes that were made since the last time the tree was
50
52
These revisions have other uses as well. For example, one can comment
51
53
revisions to record what the recent set of changes meant by providing an
52
optional log message. Real life log messages include things like "Fixed
53
the web template to close the table" and "Added sftp suppport. Fixes #595"
54
optional log message. Real life log messages include things like "Fixed the
55
web template to close the table" and "Added sftp suppport. Fixes #595"
55
57
We keep these logs so that if later there is some sort of problem with
56
sftp, we can figure out when the problem probably happened.
61
Many Version Control Systems (VCS) are stored on servers. If one wants to
62
work on the code stored within a VCS, then one needs to connect to the
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
63
65
server and "checkout" the code. Doing so gives one a directory in which a
64
person can make changes and then commit. The VCS client then connects to
65
the VCS server and stores the changes. This method is known as the
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
68
The centralized model can have some drawbacks. A centralized VCS requires
70
The centralized model can have some drawbacks. A centralized RCS requires
69
71
that one is able to connect to the server whenever one wants to do version
70
control work. This can be a bit of a problem if your server is on some other
71
machine on the internet and you are not. Or, worse yet, you **are** on the
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
72
74
internet but the server is missing!
74
Decentralized Version Control Systems (which I'll call DVCS after this
76
Decentralized Revision Control Systems (which I'll call DRCS after this
75
77
point) deal with this problem by keeping branches on the same machine as
76
the client. In Bazaar's case, the branch is kept in the same place as
77
the code that is being version controlled. This allows the user to save
78
his changes (**commit**) whenever he wants -- even if he is offline. The
79
user only needs internet access when he wants to access the changes in
80
someone else's branch that are somewhere else.
83
A common requirement that many people have is the need to keep track of
84
the changes for a directory such as file and subdirectory changes.
85
Performing this tracking by hand is a awkward process that over time
86
becomes unwieldy. That is, until one considers version control tools such
87
as Bazaar. These tools automate the process of storing data by creating
88
a **revision** of the directory tree whenever the user asks.
90
Version control software such as Bazaar can do much more than just
91
storage and performing undo. For example, with Bazaar a developer can
92
take the modifications in one branch of software and apply them to a
93
related branch -- even if those changes exist in a branch owned by
94
somebody else. This allows developers to cooperate without giving
95
write access to the repository.
97
Bazaar remembers the ''ancestry'' of a revision: the previous revisions
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 his
80
changes (**commit**) whenever he wants -- even if he is offline. The user
81
only needs internet access when he wants to access the changes in someone
82
else's branch that are somewhere else.
85
A common requirement that many people have is the need to keep track of the
86
changes for a directory such as file and subdirectory changes. Performing
87
this tracking by hand is a awkward process that over time becomes unwieldy.
88
That is, until one considers version control tools such as Bazaar-NG. These
89
tools automate the process of storing data by creating a **revision** of
90
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 another,
95
related, branch -- even if those changes exist in a branch owned by
96
somebody else. This allows developers to cooperate without giving write
99
Bazaar-NG remembers the ''ancestry'' of a revision: the previous revisions
98
100
that it is based upon. A single revision may have more than one direct
99
101
descendant, each with different changes, representing a divergence in the
100
evolution of the tree. By branching, Bazaar allows multiple people to
102
evolution of the tree. By branching, Bazaar-NG allows multiple people to
101
103
cooperate on the evolution of a project, without all needing to work in
102
104
strict lock-step. Branching can be useful even for a single developer.
104
Introducing yourself to Bazaar
105
==============================
106
Introducing yourself to Bazaar-NG
107
=================================
107
Bazaar installs a single new command, **bzr**. Everything else is a
108
subcommand of this. You can get some help with ``bzr help``. Some arguments
109
are grouped in topics: ``bzr help topics`` to see which topics are available.
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
111
113
One function of a version control system is to keep track of who changed
112
114
what. In a decentralized system, that requires an identifier for each
113
115
author that is globally unique. Most people already have one of these: an
114
email address. Bazaar is smart enough to automatically generate an email
116
email address. Bzr is smart enough to automatically generate an email
115
117
address by looking up your username and hostname. If you don't like the
116
guess that Bazaar makes, then three options exist:
118
1. Set an email address via ``bzr whoami``. This is the simplest way.
120
To set a global identity, use::
122
% bzr whoami "Your Name <email@example.com>"
124
If you'd like to use a different address for a specific branch, enter
125
the branch folder and use::
127
% bzr whoami --branch "Your Name <email@example.com>"
129
#. Setting the email address in the ``~/.bazaar/bazaar.conf`` [1]_ by
130
adding the following lines. Please note that ``[DEFAULT]`` is case
134
email=Your Name <email@isp.com>
136
As above, you can override this settings on a branch by branch basis
137
by creating a branch section in ``~/.bazaar/locations.conf`` and
138
adding the following lines::
140
[/the/path/to/the/branch]
141
email=Your Name <email@isp.com>
144
#. Overriding the two previous options by setting the global environment
145
variable ``$BZR_EMAIL`` or ``$EMAIL`` (``$BZR_EMAIL`` will take
146
precedence) to your full email address.
148
.. [1] On Windows, the users configuration files can be found in the
149
application data directory. So instead of ``~/.bazaar/branch.conf``
150
the configuration file can be found as:
151
``C:\Documents and Settings\<username>\Application Data\Bazaar\2.0\branch.conf``.
152
The same is true for ``locations.conf``, ``ignore``, and the
153
``plugins`` directory.
118
guess that Bazaar-NG makes, then three options exist:
120
1. (**Bazaar-NG 0.6 and later**). Setting the email address in the
121
``~/.bazaar/bazaar.conf`` by adding the following lines. Please note that
122
``[DEFAULT]`` is case sensitive::
125
email= Your Name <email@isp.com>
127
1. (**Bazaar-NG 0.6 and later**) Override the previous setting on a
128
branch by branch basis by creating a branch section in
129
``~/.bazaar/branches.conf`` by adding the following lines::
131
[/the/directory/to/the/branch]
132
email=Your Name <email@isp.com>
134
1. Overriding the two previous options by setting the global environment
135
variable ``$BZREMAIL`` or ``$EMAIL`` (``$BZREMAIL`` will take precedence)
136
to your full email address.
155
138
Creating a branch
156
139
=================
158
History is by default stored in the .bzr directory of the branch. In a
159
future version of Bazaar, there will be a facility to store it in a
160
separate repository, which may be remote.
162
We create a new branch by running ``bzr init`` in an existing directory::
141
History is by default stored in the .bzr directory of the branch. There
142
will be a facility to store it in a separate repository, which may be
143
remote. We create a new branch by running **bzr init** in an existing
176
As with CVS, there are three classes of file: unknown, ignored, and
177
versioned. The **add** command makes a file versioned: that is, changes
178
to it will be recorded by the system::
158
As for CVS, there are three classes of file: unknown, ignored, and
159
versioned. The **add** command makes a file versioned: that is, changes to
160
it will be recorded by the system::
180
162
% echo 'hello world' > hello.txt
184
168
% bzr add hello.txt
191
If you add the wrong file, simply use ``bzr remove`` to make it
192
unversioned again. This does not delete the working copy in this case,
193
though it may in others [2]_.
195
.. [2] ``bzr remove`` will remove the working copy if it is currently
196
versioned, but has no changes from the last committed version. You
197
can force the file to always be kept with the ``--keep`` option to
198
``bzr remove``, or force it to always be deleted with ``--force``.
173
If you add the wrong file, simply use **bzr remove** to make it unversioned
174
again. This does not delete the working copy.
203
179
All history is stored in a branch, which is just an on-disk directory
204
containing control files. By default there is no separate repository or
205
database as used in svn or svk. You can choose to create a repository if
206
you want to (see the ``bzr init-repo`` command). You may wish to do this
207
if you have very large branches, or many branches of a moderately sized
180
containing control files. There is no repository or database as used in
210
183
You'll usually refer to branches on your computer's filesystem just by
211
184
giving the name of the directory containing the branch. bzr also supports
212
accessing branches over http and sftp, for example::
214
% bzr log http://bazaar-vcs.org/bzr/bzr.dev/
215
% bzr log sftp://bazaar-vcs.org/bzr/bzr.dev/
217
By installing bzr plugins you can also access branches using the rsync
220
See the `Publishing your branch`_ section for more about how to put your
221
branch at a given location.
185
accessing branches over http, for example::
187
% bzr log http://bazaar-ng.org/bzr/bzr.dev/
189
By installing bzr plugins you can also access branches over the sftp or
223
192
Reviewing changes
224
193
=================
226
195
Once you have completed some work, you will want to **commit** it to the
227
version history. It is good to commit fairly often: whenever you get a
228
new feature working, fix a bug, or improve some code or documentation.
229
It's also a good practice to make sure that the code compiles and passes
230
its test suite before committing, to make sure that every revision is a
196
version history. It is good to commit fairly often: whenever you get a new
197
feature working, fix a bug, or improve some code or documentation. It's
198
also a good practice to make sure that the code compiles and passes its
199
test suite before committing, to make sure that every revision is a
231
200
known-good state. You can also review your changes, to make sure you're
232
201
committing what you intend to, and as a chance to rethink your work before
233
you permanently record it.
202
you permanently record it.
235
Two bzr commands are particularly useful here: **status** and **diff**.
204
Two bzr commands are particularly useful here: **status** and **diff**.
302
264
% bzr commit -m "added my first file"
304
You can also use the ``-F`` option to take the message from a file. Some
305
people like to make notes for a commit message while they work, then
306
review the diff to make sure they did what they said they did. (This file
307
can also be useful when you pick up your work after a break.)
266
You can also use the -F option to take the message from a file. Some
267
people like to make notes for a commit message while they work, then review
268
the diff to make sure they did what they said they did. (This file can
269
also be useful when you pick up your work after a break.)
309
271
Message from an editor
310
----------------------
312
If you use neither the ``-m`` nor the ``-F`` option then bzr will open an
313
editor for you to enter a message. The editor to run is controlled by
314
your ``$VISUAL`` or ``$EDITOR`` environment variable, which can be overridden
315
by the ``editor`` setting in ``~/.bazaar/bazaar.conf``; ``$BZR_EDITOR`` will
316
override either of the above mentioned editor options. If you quit the
317
editor without making any changes, the commit will be cancelled.
319
The file that is opened in the editor contains a horizontal line. The part
320
of the file below this line is included for information only, and will not
321
form part of the commit message. Below the separator is shown the list of
322
files that are changed in the commit. You should write your message above
323
the line, and then save the file and exit.
325
If you would like to see the diff that will be committed as you edit the
326
message you can use the ``--show-diff`` option to ``commit``. This will include
327
the diff in the editor when it is opened, below the separator and the
328
information about the files that will be committed. This means that you can
329
read it as you write the message, but the diff itself wont be seen in the
330
commit message when you have finished. If you would like parts to be
331
included in the message you can copy and paste them above the separator.
333
Marking bugs as fixed
334
---------------------
336
Many changes to a project are as a result of fixing bugs. Bazaar can keep
337
metadata about bugs you fixed when you commit them. To do this you use the
338
``--fixes`` option. This option takes an argument that looks like this::
340
% bzr commit --fixes <tracker>:<id>
342
Where ``<tracker>`` is an identifier for a bug tracker and ``<id>`` is an
343
identifier for a bug that is tracked in that bug tracker. ``<id>`` is usually
344
a number. Bazaar already knows about a few popular bug trackers. They are
345
bugs.launchpad.net, bugs.debian.org, and bugzilla.gnome.org. These trackers
346
have their own identifiers: lp, deb, and gnome respectively. For example,
347
if you made a change to fix the bug #1234 on bugs.launchpad.net, you would
348
use the following command to commit your fix::
350
% bzr commit -m "fixed my first bug" --fixes lp:1234
352
For more information on this topic or for information on how to configure
353
other bug trackers please read `Bug Tracker Settings`_.
355
.. _Bug Tracker Settings: ../user-reference/index.html#bug-tracker-settings
272
======================
274
If you use neither the `-m` nor the `-F` option then bzr will open an
275
editor for you to enter a message. The editor to run is controlled by your
276
`$EDITOR` environment variable or (Post Bazaar-NG 0.6) email setting in .
277
If you quit the editor without making any changes, the commit will be
402
321
If a glob contains a slash, it is matched against the whole path from the
403
322
top of the tree; otherwise it is matched against only the filename. So
404
323
the previous example ignores files with extension ``.o`` in all
405
subdirectories, but this example ignores only ``config.h`` at the top level
324
subdirectories, but this example ignores only config.h at the top level
406
325
and HTML files in ``doc/``::
411
To get a list of which files are ignored and what pattern they matched,
412
use ``bzr ignored``::
330
To get a list of which files are ignored and what pattern they matched, use
415
334
config.h ./config.h
418
It is OK to have either an ignore pattern match a versioned file, or to
419
add an ignored file. Ignore patterns have no effect on versioned files;
420
they only determine whether unversioned files are reported as unknown or
337
It is OK to have either an ignore pattern match a versioned file, or to add
338
an ignored file. Ignore patterns have no effect on versioned files; they
339
only determine whether unversioned files are reported as unknown or
423
The ``.bzrignore`` file should normally be versioned, so that new copies
424
of the branch see the same patterns::
342
The ''.bzrignore'' file should normally be versioned, so that new copies of
343
the branch see the same patterns::
426
345
% bzr add .bzrignore
427
346
% bzr commit -m "Add ignore patterns"
433
As an alternative to editing the ``.bzrignore`` file, you can use the
434
``bzr ignore`` command. The ``bzr ignore`` command takes filenames and/or
435
patterns as arguments and then adds them to the ``.bzrignore`` file. If a
436
``.bzrignore`` file does not exist the ``bzr ignore`` command will
437
automatically create one for you, and implicitly add it to be versioned::
444
Just like when editing the ``.bzrignore`` file on your own, you should
445
commit the automatically created ``.bzrignore`` file::
447
% bzr commit -m "Added tags to ignore file"
453
There are some ignored files which are not project specific, but more user
454
specific. Things like editor temporary files, or personal temporary files.
455
Rather than add these ignores to every project, bzr supports a global
456
ignore file in ``~/.bazaar/ignore`` [1]_. It has the same syntax as the
457
per-project ignore file.
460
349
Examining history
461
350
=================
466
The ``bzr log`` command shows a list of previous revisions. The ``bzr log
467
--forward`` command does the same in chronological order to get most
468
recent revisions printed at last.
355
The **bzr log** command shows a list of previous revisions. The **bzr log
356
--forward** command does the same in chronological order to get most recent
357
revisions printed at last.
470
As with ``bzr diff``, ``bzr log`` supports the ``-r`` argument::
359
As with bzr diff, bzr log supports the -r argument::
472
361
% bzr log -r 1000.. # Revision 1000 and everything after it
473
% bzr log -r ..1000 # Everything up to and including r1000
362
% bzr log -r ..1000 # Everything up to and including revision % 1000
474
363
% bzr log -r 1000..1100 # changes from 1000 to 1100
475
364
% bzr log -r 1000 # The changes in only revision 1000
516
405
% bzr remove -v hello.txt
526
If you remove the wrong file by accident, you can use ``bzr revert`` to
412
If you remove the wrong file by accident, you can use **bzr revert** to restore it.
533
417
Often rather than starting your own project, you will want to submit a
534
change to an existing project. To do this, you'll need to get a copy of
535
the existing branch. Because this new copy is potentially a new branch,
536
the command is called **branch**::
418
change to an existing project. You can get a copy of an existing branch by
419
copying its directory, expanding a tarball, or by a remote copy using
420
something like rsync. You can also use bzr to fetch a copy. Because this
421
new copy is potentially a new branch, the command is called *branch*::
538
% bzr branch http://bazaar-vcs.org/bzr/bzr.dev
423
% bzr branch http://bazaar-ng.org/bzr/bzr.dev
541
426
This copies down the complete history of this branch, so we can do all
542
427
operations on it locally: log, annotate, making and merging branches.
543
428
There will be an option to get only part of the history if you wish.
545
You can also get a copy of an existing branch by copying its directory,
546
expanding a tarball, or by a remote copy using something like rsync.
433
Another thing that gives you files to edit is a checkout. A checkout is
434
always associated with another branch. One of the purposes of checkouts is
435
to get multiple working trees for a branch. Another reason for checkouts is
436
to get access to the files in a branch without keeping a full copy of the
437
branch available locally. Getting a checkout looks like this::
439
% bzr checkout http://bazaar-ng.org/bzr/bzr.dev
442
Checkouts have several advantages and drawbacks. A checkout does not have
443
any RCS history as the history is stored in the branch for which a checkout
444
is assocated with. This behaviour results in two general rules:
446
1. checkouts take much less space than a full branch
447
2. checkouts are much faster if the branch is on a local filesystem.
548
450
Following upstream changes
549
451
==========================
551
You can stay up-to-date with the parent branch by "pulling" in their
453
You can stay up-to-date with the parent branch by "pulling" in their changes
556
After this change, the local directory will be a mirror of the source. This
557
includes the ''revision-history'' - which is a list of the commits done in
558
this branch, rather than merged from other branches.
458
After this change, the local directory will be a mirror of the source.
560
This command only works if your local (destination) branch is either an
561
older copy of the parent branch with no new commits of its own, or if the
562
most recent commit in your local branch has been merged into the parent
460
This command only works if your local (destination) branch includes only
461
changes from the parent branch and no commits of its own. Otherwise, the
462
branches are said to have ''diverged'', and they must be merged instead.
565
464
Merging from related branches
566
465
=============================
568
If two branches have diverged (both have unique changes) then ``bzr
569
merge`` is the appropriate command to use. Merge will automatically
570
calculate the changes that exist in the branch you're merging from that
571
are not in your branch and attempt to apply them in your branch.
467
If two branches have diverged (both have unique changes) then **bzr merge**
468
is the appropriate command to use. Merge will automatically calculate the
469
changes that exist in the branch you're merging from that are not in your
470
branch and attempt to apply them in your branch. ::
578
If there is a conflict during a merge, 3 files with the same basename
579
are created. The filename of the common base is appended with ".BASE",
580
the filename of the file containing your changes is appended with
581
".THIS" and the filename with the changes from the other tree is
582
appended with ".OTHER". Using a program such as kdiff3, you can now
583
comfortably merge them into one file. In order to commit you have to
584
rename the merged file (".THIS") to the original file name. To
585
complete the conflict resolution you must use the resolve command,
586
which will remove the ".OTHER" and ".BASE" files. As long as there
587
exist files with .BASE, .THIS or .OTHER the commit command will
592
% kdiff3 file.BASE file.OTHER file.THIS
596
[**TODO**: explain conflict markers within files]
475
Conflicts during merge
476
======================
478
Sometimes two branches modify the same lines in the same files at the same
479
time. The result of this collision is called a "conflict". These conflicts
480
must be resolved by hand as Bazaar-NG can not tell which change you would
481
prefer to keep. You are, however, provided with some information which
482
makes the job easier for you to deal with.
484
The first thing that Bazaar-NG does is to merge in the parts that it can.
485
The parts that can not be merged are put into the file for you to decide
488
The second thing that Bazaar-NG does is to provide you with three extra
489
files for each file that is conflicted. The names of these files are the
490
same as the original filename with ".BASE", ".THIS" and ".OTHER" appended.
491
Each of these files serves a different purpose:
493
+------------------+-----------------------------------------------------+
494
| filename.BASE | filename as it looked back when filename was |
495
| | identical in both branches. |
496
+-----------------+------------------------------------------------------+
497
| filename.THIS | The version of filename for this branch. All of the |
498
| | parts of the merge with the other branch that can be |
499
| | merged from the other branch without conflicts are |
500
| | accepted. The parts which conflict are thrown away |
501
+-----------------+------------------------------------------------------+
502
| filename.OTHER | The version of filename for the other branch. All of |
503
| | parts of the merge from this branch that can be |
504
| | successfully merged are accepted. The parts of the |
505
| | merge which conflict are thrown away |
506
+-----------------+------------------------------------------------------+
509
For example, imagine that you and another developer were working on on a
510
simple branch that only contained hello.c. Both of you agree that saying
511
"Yo" to users is a little _too_ informal. Both of you also decide to fix
512
the problem at the same time, but take slightly different approaches. You
513
decide upon saying "Hi", while your compatriot decides upon a slightly more
514
formal "Hello". You both make the respective change and commit. Then, you
515
merge his change. The result would look something like this:
517
+----------------------+------------------------+
518
| hello.c.THIS | hello.c.THAT |
519
+----------------------+------------------------+
520
| | #include <stdio.h> | | #include <stdio.h> |
522
| | main() | | main() |
524
| | printf("Hi"); | | printf("Hello"); |
527
+----------------------+------------------------+
528
| hello.c.BASE | hello.c |
529
+----------------------+------------------------+
530
| | #include <stdio.h> | | #include <stdio.h> |
532
| | main() | | main() |
534
| | printf("Yo"); | | <<<<<<< TREE |
535
| | } | | printf("Hi"); |
537
| | | printf("Hello"); |
538
| | | <<<<<<< MERGE-SOURCE |
541
+----------------------+------------------------+
543
Your job as a developer is to decide which better. A multitude of options
544
exist for how to solve the conflict. Two of them are:
546
1. Decide whether you'd like hello.c.THIS, hello.c.THAT or hello.c.BASE.
547
Copy that one to hello.c.
549
1. edit hello.c and decide which part you wish to keep. Your solution
550
will be between the lines that say "<<<<<<< TREE" and "=======". His
551
solution will be between "=======" and "<<<<<<< MERGE-SOURCE"). Do not
552
forget to delete the markers too!
554
Bazaar also tracks which files conflicted during merge to prevent you from
555
accidentally committing the conflict markers to your branch. To tell
556
Bazaar-NG that you have fixed one of the conflicted files, run::
558
% bzr resolve hello.c
560
Using a program such as kdiff3, you can now comfortably merge them into one
561
file. To commit you have to rename it to the original basename and delete
562
the other two files. As long as there exist files with .BASE, .THIS or
563
.OTHER the commit command will complain.
599
566
Publishing your branch
600
567
======================
602
You don't need a special server to publish a bzr branch, just a normal web
603
server. Just mirror the files to your server, including the .bzr
604
directory. One can push a branch (or the changes for a branch) by one of
605
the following three methods:
607
* The best method is to use bzr itself to do it.
611
% bzr push sftp://servername.com/path/to/directory
613
(The destination directory must already exist unless the
614
``--create-prefix`` option is used.)
616
* Another option is the ``rspush`` plugin that comes with BzrTools, which
617
uses rsync to push the changes to the revision history and the working
620
* You can also copy the files around manually, by sending a tarball, or using
621
rsync, or other related file transfer methods. This is usually less safe
622
than using ``push``, but may be faster or easier in some situations.
625
Moving changes between trees
626
============================
628
It happens to the best of us: sometimes you'll make changes in the wrong
629
tree. Maybe because you've accidentally started work in the wrong directory,
630
maybe because as you're working, the change turns out to be bigger than you
631
expected, so you start a new branch for it.
633
To move your changes from one tree to another, use
638
% bzr merge --uncommitted OLDDIR
640
This will apply all of the uncommitted changes you made in OLDDIR to NEWDIR.
641
It will not apply committed changes, even if they could be applied to NEWDIR
642
with a regular merge. The changes will remain in OLDDIR, but you can use ``bzr
643
revert OLDDIR`` to remove them, once you're satisfied with NEWDIR.
645
NEWDIR does not have to be a copy of OLDDIR, but they should be related.
646
The more different they are, the greater the chance of conflicts.
569
You do not need a special server to publish a bzr branch. Generally
570
speaking, one can push a branch via sftp to a directory that is served via
571
apache. Bazaar-NG supports a variety of methods for reading and writing
574
Bazaar-NG can push branches via these methods::
575
1. sftp via ssh (the most common)
577
3. Any filesystem mounted via remote filesystem tools like NFS and Samba
578
4. The Canonical.com branch hosting service (via scoming soon)
579
5. Rsync (requires bzrtools plugin from http://bazaar-vcs.org/BzrTools)
581
Bazaar-NG can also read branches via a variety of methods::
583
1. Standard http, (E.g. Apache)
588
The most common approach is to push a branch via sftp to a place that is
589
served via SFTP. This usually looks something like::
591
* You have a branch in ~/somebranch
592
* You have a directory on /home/yourname/public_html on BigServer that
593
shows up as http://BigServer/~yourname
595
In cases such as this, you'll usually perform the following commands to
596
initially push your branch via rsync (which requires the bzrtools plugin)::
599
% bzr push yourname@BigServer:public_html/somebranch/
601
Other people can then access for your branch by doing the following::
604
% bzr merge http://BigServer/~yourname/somebranch
added
removed
doc/tutorial.txt
