1
=============================
2
Bazaar Architectural Overview
3
=============================
5
This document describes the key classes and concepts within Bazaar. It is
6
intended to be useful to people working on the Bazaar codebase, or to
7
people writing plugins. People writing plugins may also like to read the
8
guide to `Integrating with Bazaar <integrating.html>`_ for some specific
11
If you have any questions, or if something seems to be incorrect, unclear
12
or missing, please talk to us in ``irc://irc.freenode.net/#bzr``, or write
13
to the Bazaar mailing list.
22
The ``Transport`` layer handles access to local or remote directories.
23
Each Transport object acts as a logical connection to a particular
24
directory, and it allows various operations on files within it. You can
25
*clone* a transport to get a new Transport connected to a subdirectory or
28
Transports are not used for access to the working tree. At present
29
working trees are always local and they are accessed through the regular
30
Python file I/O mechanisms.
35
Transports work in terms of URLs. Take note that URLs are by definition
36
only ASCII - the decision of how to encode a Unicode string into a URL
37
must be taken at a higher level, typically in the Store. (Note that
38
Stores also escape filenames which cannot be safely stored on all
39
filesystems, but this is a different level.)
41
The main reason for this is that it's not possible to safely roundtrip a
42
URL into Unicode and then back into the same URL. The URL standard
43
gives a way to represent non-ASCII bytes in ASCII (as %-escapes), but
44
doesn't say how those bytes represent non-ASCII characters. (They're not
45
guaranteed to be UTF-8 -- that is common but doesn't happen everywhere.)
47
For example, if the user enters the URL ``http://example/%e0``, there's no
48
way to tell whether that character represents "latin small letter a with
49
grave" in iso-8859-1, or "latin small letter r with acute" in iso-8859-2,
50
or malformed UTF-8. So we can't convert the URL to Unicode reliably.
52
Equally problematic is if we're given a URL-like string containing
53
(unescaped) non-ASCII characters (such as the accented a). We can't be
54
sure how to convert that to a valid (i.e. ASCII-only) URL, because we
55
don't know what encoding the server expects for those characters.
56
(Although it is not totally reliable, we might still accept these and
57
assume that they should be put into UTF-8.)
59
A similar edge case is that the URL ``http://foo/sweet%2Fsour`` contains
60
one directory component whose name is "sweet/sour". The escaped slash is
61
not a directory separator, but if we try to convert the URL to a regular
62
Unicode path, this information will be lost.
64
This implies that Transports must natively deal with URLs. For simplicity
65
they *only* deal with URLs; conversion of other strings to URLs is done
66
elsewhere. Information that Transports return, such as from ``list_dir``,
67
is also in the form of URL components.
73
A workingtree is a special type of Tree that's associated with a working
74
directory on disk, where the user can directly modify the files.
78
* Maintaining a WorkingTree on disk at a file path.
79
* Maintaining the basis inventory (the inventory of the last commit done)
80
* Maintaining the working inventory.
81
* Maintaining the pending merges list.
82
* Maintaining the stat cache.
83
* Maintaining the last revision the working tree was updated to.
84
* Knows where its Branch is located.
90
* local access to the working tree
95
A Branch is a key user concept - its a single line of history that one or
96
more people have been committing to.
98
A Branch is responsible for:
100
* Holding user preferences that are set in a Branch.
101
* Holding the 'tip': the last revision to be committed to this Branch. (And the revno of that revision.)
102
* Knowing how to open the Repository that holds its history.
103
* Allowing write locks to be taken out to prevent concurrent alterations to the branch.
106
* URL access to its base directory.
107
* A Transport to access its files.
108
* A Repository to hold its history.
113
Repositories store committed history: file texts, revisions, inventories,
114
and graph relationships between them. A repository holds a bag of
115
revision data that can be pointed to by various branches:
117
* Maintains storage of various history data at a URL:
118
* Revisions (Must have a matching inventory)
120
* Inventories for each Revision. (Must have all the file texts available).
123
* Synchronizes concurrent access to the repository by different
124
processes. (Most repository implementations use a physical
125
mutex only for a short period, and effectively support multiple readers
131
A repository can be configured to refer to a list of "fallback"
132
repositories. If a particular revision is not present in the original
133
repository, it refers the query to the fallbacks.
135
Compression deltas don't span physical repository boundaries. So the
136
first commit to a new, empty repository with fallback repositories will
137
store a full text of the inventory, and of every new file text.
139
At runtime, repository stacking is actually configured by the branch, not
140
the repository. So doing ``a_bzrdir.open_repository()``
141
gets you just the single physical repository, while
142
``a_bzrdir.open_branch().repository`` gets one configured with a stacking.
143
Therefore, to permanently change the fallback repository stored on disk,
144
you must use ``Branch.set_stacked_on_url``.
146
Changing away from an existing stacked-on URL will copy across any
147
necessary history so that the repository remains usable.
149
A repository opened from an HPSS server is never stacked on the server
150
side, because this could cause complexity or security problems with the
151
server acting as a proxy for the client. Instead, the branch on the
152
server exposes the stacked-on URL and the client can open that.