1
For things to do in Bazaar development, see
3
https://bugs.launchpad.net/bzr/
4
https://blueprints.launchpad.net/bzr/
1
.. -*- mode: indented-text; compile-command: "make -C doc" -*-
9
See also various low-level TODOs in the source code. Try looking in
10
the list archive or on gmane.org for previous discussion of these
13
These are classified by approximate size: an hour or less, a day or
14
less, and several days or more.
20
* Add of a file that was present in the base revision should put back
23
* Handle diff of files which do not have a trailing newline; probably
24
requires patching difflib to get it exactly right, or otherwise
25
calling out to GNU diff.
27
* Import ElementTree update patch.
29
* Plugins that provide commands. By just installing a file into some
30
directory (e.g. ``/usr/share/bzr/plugins``) it should be possible to
31
create new top-level commands (``bzr frob``). Extensions can be
32
written in either Python (in which case they use the bzrlib API) or
33
in a separate process (in sh, C, whatever). It should be possible
34
to get help for plugin commands.
36
* Smart rewrap text in help messages to fit in $COLUMNS (or equivalent
39
* -r option should take a revision-id as well as a revno.
41
* ``bzr info`` could show space used by working tree, versioned files,
42
unknown and ignored files.
44
* ``bzr info`` should count only people with distinct email addresses as
45
different committers. (Or perhaps only distinct userids?)
47
* On Windows, command-line arguments should be `glob-expanded`__,
48
because the shell doesn't do this. However, there are probably some
49
commands where this shouldn't be done, such as 'bzr ignore', because
50
we want to accept globs.
52
* ``bzr ignore`` command that just adds a line to the ``.bzrignore`` file
53
and makes it versioned. Fix this to break symlinks.
55
* Any useful sanity checks in 'bzr ignore'? Perhaps give a warning if
56
they try to add a single file which is already versioned, or if they
57
add a pattern which already exists, or if it looks like they gave an
60
__ http://mail.python.org/pipermail/python-list/2001-April/037847.html
62
* Separate read and write version checks?
64
* ``bzr status DIR`` should give status on all files under that
67
* ``bzr log DIR`` should give changes to any files within DIR.
69
* Check all commands have decent help.
71
* ``bzr inventory -r REV`` and perhaps unify this with ``bzr ls``,
72
giving options to display ids, types, etc.
74
* Split BzrError into various more specific subclasses for different
75
errors people might want to catch.
77
* If the export destination ends in '.tar', '.tar.gz', etc then create
78
a tarball instead of a directory. (Need to actually make a
79
temporary directory and then tar that up.)
81
http://www.gelato.unsw.edu.au/archives/git/0504/2194.html
83
* RemoteBranch could maintain a cache either in memory or on disk. We
84
know more than an external cache might about which files are
85
immutable and which can vary. On the other hand, it's much simpler
86
to just use an external proxy cache.
88
Perhaps ~/.bzr/http-cache. Baz has a fairly simple cache under
89
~/.arch-cache, containing revision information encoded almost as a
90
bunch of archives. Perhaps we could simply store full paths.
92
* Maybe also store directories in the statcache so that we can quickly
93
identify that they still exist.
95
* Diff should show timestamps; for files from the working directory we
96
can use the file itself; for files from a revision we should use the
97
commit time of the revision.
99
* Perhaps split command infrastructure from the actual command
102
* Cleaner support for negative boolean options like --no-recurse.
104
* Statcache should possibly map all file paths to / separators
106
* quotefn doubles all backslashes on Windows; this is probably not the
107
best thing to do. What would be a better way to safely represent
108
filenames? Perhaps we could doublequote things containing spaces,
109
on the principle that filenames containing quotes are unlikely?
110
Nice for humans; less good for machine parsing.
112
* Patches should probably use only forward slashes, even on Windows,
113
otherwise Unix patch can't apply them. (?)
115
* Branch.update_revisions() inefficiently fetches revisions from the
116
remote server twice; once to find out what text and inventory they
117
need and then again to actually get the thing. This is a bit
120
One complicating factor here is that we don't really want to have
121
revisions present in the revision-store until all their constituent
122
parts are also stored.
124
The basic problem is that RemoteBranch.get_revision() and similar
125
methods return object, but what we really want is the raw XML, which
126
can be popped into our own store. That needs to be refactored.
128
* ``bzr status FOO`` where foo is ignored should say so.
130
* ``bzr mkdir A...`` should just create and add A.
137
* Merge revert patch.
139
* ``bzr mv`` that does either rename or move as in Unix.
141
* More efficient diff of only selected files. We should be able to
142
just get the id for the selected files, look up their location and
143
diff just those files. No need to traverse the entire inventories.
145
* ``bzr status DIR`` or ``bzr diff DIR`` should report on all changes
146
under that directory.
148
* Fix up Inventory objects to represent root object as an entry.
150
* Don't convert entire entry from ElementTree to an object when it is
151
read in, but rather wait until the program actually wants to know
154
* Extract changes from one revision to the next to a text form
155
suitable for transmission over email.
159
- Selected-file commit
161
- Impossible selected-file commit: adding things in non-versioned
162
directories, crossing renames, etc.
164
* Write a reproducible benchmark, perhaps importing various kernel versions.
166
* Directly import diffs! It seems a bit redundant to need to rescan
167
the directory to work out what files diff added/deleted/changed when
168
all the information is there in the diff in the first place.
169
Getting the exact behaviour for added/deleted subdirectories etc
172
At the very least we could run diffstat over the diff, or perhaps
173
read the status output from patch. Just knowing which files might
174
be modified would be enough to guide the add and commit.
176
Given this we might be able to import patches at 1/second or better.
178
* Get branch over http.
180
* Pull pure updates over http.
182
* revfile compression.
184
* Split inventory into per-directory files.
186
* Fix ignore file parsing:
188
- fnmatch is not the same as unix patterns
190
- perhaps add extended globs from rsh/rsync
192
- perhaps a pattern that matches only directories or non-directories
194
* Consider using Python logging library as well as/instead of
197
* Commands should give some progress indication by default.
199
- But quieten this with ``--silent``.
201
* Change to using gettext message localization.
203
* Make a clearer separation between internal and external bzrlib
204
interfaces. Make internal interfaces use protected names. Write at
205
least some documentation for those APIs, probably as docstrings.
207
Consider using ZopeInterface definitions for the external interface;
208
I think these are already used in PyBaz. They allow automatic
209
checking of the interface but may be unfamiliar to general Python
210
developers, so I'm not really keen.
212
* Commands to dump out all command help into a manpage or HTML file or
215
* Handle symlinks in the working directory; at the very least it
216
should be possible for them to be present and ignored/unknown
217
without causing assertion failures.
219
Eventually symlinks should be versioned.
221
* Allow init in a subdirectory to create a nested repository, but only
222
if the subdirectory is not already versioned. Perhaps also require
223
a ``--nested`` to protect against confusion.
227
* More test framework:
229
- Class that describes the state of a working tree so we can just
232
* There are too many methods on Branch() that really manipulate the
233
WorkingTree. They should be moved across.
235
Also there are some methods which are duplicated on Tree and
236
Inventory objects, and it should be made more clear which ones are
237
proxies and which ones behave differently, and how.
239
* Try using XSLT to add some formatting to REST-generated HTML. Or
240
maybe write a small Python program that specifies a header and foot
241
for the pages and calls into the docutils libraries.
243
* --format=xml for log, status and other commands.
245
* Attempting to explicitly add a file that's already added should give
246
a warning; however there should be no warning for directories (since
247
we scan for new children) or files encountered in a directory that's
250
* Better handling of possible collisions on case-losing filesystems;
251
make sure a single file does not get added twice under different
254
* Clean up XML inventory:
256
- Use nesting rather than parent_id pointers.
258
- Hold the ElementTree in memory in the Inventory object and work
259
directly on that, rather than converting into Python objects every
260
time it is read in. Probably still exposoe it through some kind of
261
object interface though, but perhaps that should just be a proxy
264
- Less special cases for the root directory.
266
* Perhaps inventories should remember the revision in which each file
267
was last changed, as well as its current state? This is a bit
268
redundant but might often be interested to know.
270
* stat cache should perhaps only stat files as necessary, rather than
271
doing them all up-front. On the other hand, that disallows the
272
opimization of stating them in inode order.
274
* It'd be nice to pipeline multiple HTTP requests. Often we can
275
predict what will be wanted in future: all revisions, or all texts
276
in a particular revision, etc.
278
urlgrabber's docs say they are working on batched downloads; we
279
could perhaps ride on that or just create a background thread (ew).
281
* Paranoid mode where we never trust SHA-1 matches.
283
* Don't commit if there are no changes unless forced.
285
* --dry-run mode for commit? (Or maybe just run with
286
check-command=false?)
288
* Generally, be a bit more verbose unless --silent is specified.
290
* Function that finds all changes to files under a given directory;
291
perhaps log should use this if a directory is given.
293
* XML attributes might have trouble with filenames containing \n and
294
\r. Do we really want to support this? I think perhaps not.
296
* Remember execute bits, so that exports will work OK.
298
* Unify smart_add and plain Branch.add(); perhaps smart_add should
299
just build a list of files to add and pass that to the regular add
302
* Function to list a directory, saying in which revision each file was
303
last modified. Useful for web and gui interfaces, and slow to
304
compute one file at a time.
306
* unittest is standard, but the results are kind of ugly; would be
307
nice to make it cleaner.
309
* Check locking is correct during merge-related operations.
311
* Perhaps attempts to get locks should timeout after some period of
312
time, or at least display a progress message.
314
* Split out upgrade functionality from check command into a separate
321
* Generate annotations from current file relative to previous
324
- Is it necessary to store any kind of annotation where data was
327
* Update revfile_ format and make it active:
329
- Texts should be identified by something keyed on the revision, not
330
an individual text-id. This is much more useful for annotate I
331
think; we want to map back to the revision that last changed it.
333
- Access revfile revisions through the Tree/Store classes.
335
- Check them from check commands.
339
.. _revfile: revfile.html
341
* Hooks for pre-commit, post-commit, etc.
343
Consider the security implications; probably should not enable hooks
344
for remotely-fetched branches by default.
346
* Pre-commit check. If this hook is defined, it needs to be handled
347
specially: create a temporary directory containing the tree as it
348
will be after the commit. This means excluding any ignored/unknown
349
files, and respecting selective commits. Run the pre-commit check
350
(e.g. compile and run test suite) in there.
352
Possibly this should be done by splitting the commit function into
353
several parts (under a single interface). It is already rather
354
large. Decomposition:
356
- find tree modifications and prepare in-memory inventory
358
- export that inventory to a temporary directory
360
- run the test in that temporary directory
362
- if that succeeded, continue to actually finish the commit
364
What should be done with the text of modified files while this is
365
underway? I don't think we want to count on holding them in memory
366
and we can't trust the working files to stay in one place so I
367
suppose we need to move them into the text store, or otherwise into
368
a temporary directory.
370
If the commit does not actually complete, we would rather the
371
content was not left behind in the stores.
375
* GUI (maybe in Python GTK+?)
377
* C library interface
379
* Expansion of $Id$ keywords within working files. Perhaps do this in
380
exports first as a simpler case because then we don't need to deal
381
with removing the tags on the way back in.
added
removed
TODO
