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
* Fix tests so that import errors caused by modules don't produce false reports
21
that the tests themselves don't exist.
23
* Fix tests so that one test failure doesn't prevent other tests from running
25
* print a message at the end of running the tests telling them that the
26
test log and output exists but can be removed
28
* tests for running the commit editor, and fix problem of not passing in
31
* Merging add of a new file clashing with an existing file doesn't
32
work; add gets an error that it's already versioned and the merge
35
* Merge should ignore the destination's working directory, otherwise
36
we get an error about the statcache when pulling from a remote
39
* Add of a file that was present in the base revision should put back
42
* Not sure I'm happy with needing to pass a root id to EmptyTree;
43
comparing anything against an EmptyTree with no root should have the
46
* Handle diff of files which do not have a trailing newline; probably
47
requires patching difflib to get it exactly right, or otherwise
48
calling out to GNU diff.
50
* Should be able to copy files between branches to preserve their
51
file-id (and perhaps eventually parentage.)
53
* -r option should take a revision-id as well as a revno.
55
* ``bzr info`` should count only people with distinct email addresses as
56
different committers. (Or perhaps only distinct userids?)
58
* On Windows, command-line arguments should be `glob-expanded`__,
59
because the shell doesn't do this. However, there are probably some
60
commands where this shouldn't be done, such as 'bzr ignore', because
61
we want to accept globs.
63
* ``bzr ignore`` command that just adds a line to the ``.bzrignore`` file
64
and makes it versioned. Fix this to break symlinks.
66
* Any useful sanity checks in 'bzr ignore'? Perhaps give a warning if
67
they try to add a single file which is already versioned, or if they
68
add a pattern which already exists, or if it looks like they gave an
71
__ http://mail.python.org/pipermail/python-list/2001-April/037847.html
73
* Separate read and write version checks?
75
* ``bzr status DIR`` should give status on all files under that
78
* ``bzr log DIR`` should give changes to any files within DIR; at the
79
moment it only lists things which modify the specific named file
80
(and not its contents)
82
* ``bzr inventory -r REV`` and perhaps unify this with ``bzr ls``,
83
giving options to display ids, types, etc.
85
* Split BzrError into various more specific subclasses for different
86
errors people might want to catch.
88
* If the export destination ends in '.tar', '.tar.gz', etc then create
89
a tarball instead of a directory. (Need to actually make a
90
temporary directory and then tar that up.)
92
http://www.gelato.unsw.edu.au/archives/git/0504/2194.html
94
* RemoteBranch could maintain a cache either in memory or on disk. We
95
know more than an external cache might about which files are
96
immutable and which can vary. On the other hand, it's much simpler
97
to just use an external proxy cache.
99
Perhaps ~/.bzr/http-cache. Baz has a fairly simple cache under
100
~/.arch-cache, containing revision information encoded almost as a
101
bunch of archives. Perhaps we could simply store full paths.
103
* Maybe also store directories in the statcache so that we can quickly
104
identify that they still exist.
106
* Diff should show timestamps; for files from the working directory we
107
can use the file itself; for files from a revision we should use the
108
commit time of the revision.
110
* Perhaps split command infrastructure from the actual command
113
* Cleaner support for negative boolean options like --no-recurse.
115
* Statcache should possibly map all file paths to / separators
117
* quotefn doubles all backslashes on Windows; this is probably not the
118
best thing to do. What would be a better way to safely represent
119
filenames? Perhaps we could doublequote things containing spaces,
120
on the principle that filenames containing quotes are unlikely?
121
Nice for humans; less good for machine parsing.
123
* Patches should probably use only forward slashes, even on Windows,
124
otherwise Unix patch can't apply them. (?)
126
* Branch.update_revisions() inefficiently fetches revisions from the
127
remote server twice; once to find out what text and inventory they
128
need and then again to actually get the thing. This is a bit
131
One complicating factor here is that we don't really want to have
132
revisions present in the revision-store until all their constituent
133
parts are also stored.
135
The basic problem is that RemoteBranch.get_revision() and similar
136
methods return object, but what we really want is the raw XML, which
137
can be popped into our own store. That needs to be refactored.
139
* Guard against repeatedly merging any particular patch.
141
* More options for diff:
143
- diff two revisions of the same tree
145
- diff two different branches, optionally at different revisions
147
- diff a particular file in another tree against the corresponding
148
version in this tree (which should be the default if the second
149
parameter is a tree root)
151
- diff everything under a particular directory, in any of the above
154
- diff two files inside the same tree, even if they have different
157
- and, of course, tests for all this
159
* Reproducible performance benchmark to measure whether performance is
160
getting better or worse.
162
* ``bzr log -m foo`` should perhaps error if nothing matches?
164
* ``bzr diff -r 30 -r 40 foo.c`` or ``bzr diff -r30..40 foo.c``
166
If diffing between two branches then we probably want two -r
167
options, since the revisions don't form a range that can be
168
evaluated on either one.
170
* bzr diff shouldn't diff binary files
172
* setup.py install when run from a bzr tree should freeze the tree
173
revision-id into the installed bzr.
175
* bzr script should trap ImportError and perhaps give a better error
178
* revert after a merge should possibly remove all the BASE/THIS/OTHER
179
files to get you back to where you were.
181
* files that are added and then deleted are still reported as added
183
* stores should raise KeyError, not IndexError
185
* merging from a remote branch seems to sometimes raise errors not
188
* should be possible to give a related branch when pulling from a
189
remote branch to make things faster
191
* sometimes gives "conflicting add" even when the contents are in fact
194
* BZRDIR should be in branch.py not __init__.py.
199
* merge should add all revision and inventory XML to the local store.
201
* check should give a warning for revisions that are named in the
202
chain but not actually present in the store.
204
* remove anything outside of the branch implementation that directly
207
* More efficient diff of only selected files. We should be able to
208
just get the id for the selected files, look up their location and
209
diff just those files. No need to traverse the entire inventories.
211
* Fix up Inventory objects to represent root object as an entry.
213
* Don't convert entire entry from ElementTree to an object when it is
214
read in, but rather wait until the program actually wants to know
217
* Extract changes from one revision to the next to a text form
218
suitable for transmission over email.
222
- ``missing`` command
224
- Selected-file commit
226
- Impossible selected-file commit: adding things in non-versioned
227
directories, crossing renames, etc.
229
* Write a reproducible benchmark, perhaps importing various kernel versions.
231
* Directly import diffs! It seems a bit redundant to need to rescan
232
the directory to work out what files diff added/deleted/changed when
233
all the information is there in the diff in the first place.
234
Getting the exact behaviour for added/deleted subdirectories etc
237
At the very least we could run diffstat over the diff, or perhaps
238
read the status output from patch. Just knowing which files might
239
be modified would be enough to guide the add and commit.
241
Given this we might be able to import patches at 1/second or better.
243
* revfile compression.
245
* Split inventory into per-directory files?
247
* Fix ignore file parsing:
249
- fnmatch is not the same as unix patterns
251
- perhaps add extended globs from rsh/rsync
253
- perhaps a pattern that matches only directories or non-directories
255
* Consider using Python logging library as well as/instead of
258
* Commands should give some progress indication by default.
260
- But quieten this with ``--silent``.
262
* Change to using gettext message localization.
264
* Make a clearer separation between internal and external bzrlib
265
interfaces. Make internal interfaces use protected names. Write at
266
least some documentation for those APIs, probably as docstrings.
268
Consider using ZopeInterface definitions for the external interface;
269
I think these are already used in PyBaz. They allow automatic
270
checking of the interface but may be unfamiliar to general Python
271
developers, so I'm not really keen.
273
* Commands to dump out all command help into a manpage or HTML file or
276
* Handle symlinks in the working directory; at the very least it
277
should be possible for them to be present and ignored/unknown
278
without causing assertion failures.
280
Eventually symlinks should be versioned.
282
* Allow init in a subdirectory to create a nested repository, but only
283
if the subdirectory is not already versioned. Perhaps also require
284
a ``--nested`` to protect against confusion.
288
* More test framework:
290
- Class that describes the state of a working tree so we can just
293
* There are too many methods on Branch() that really manipulate the
294
WorkingTree. They should be moved across.
296
Also there are some methods which are duplicated on Tree and
297
Inventory objects, and it should be made more clear which ones are
298
proxies and which ones behave differently, and how.
300
* Try using XSLT to add some formatting to REST-generated HTML. Or
301
maybe write a small Python program that specifies a header and foot
302
for the pages and calls into the docutils libraries.
304
* --format=xml for log, status and other commands.
306
* Attempting to explicitly add a file that's already added should give
307
a warning; however there should be no warning for directories (since
308
we scan for new children) or files encountered in a directory that's
311
* Better handling of possible collisions on case-losing filesystems;
312
make sure a single file does not get added twice under different
315
* Clean up XML inventory:
317
- Use nesting rather than parent_id pointers.
319
- Hold the ElementTree in memory in the Inventory object and work
320
directly on that, rather than converting into Python objects every
321
time it is read in. Probably still expose it through some kind of
322
object interface though, but perhaps that should just be a proxy
325
- Less special cases for the root directory.
327
* Perhaps inventories should remember the revision in which each file
328
was last changed, as well as its current state? This is a bit
329
redundant but might often be interested to know.
331
* stat cache should perhaps only stat files as necessary, rather than
332
doing them all up-front. On the other hand, that disallows the
333
optimization of stating them in inode order.
335
* It'd be nice to pipeline multiple HTTP requests. Often we can
336
predict what will be wanted in future: all revisions, or all texts
337
in a particular revision, etc.
339
urlgrabber's docs say they are working on batched downloads; we
340
could perhaps ride on that or just create a background thread (ew).
342
* Paranoid mode where we never trust SHA-1 matches.
344
* Don't commit if there are no changes unless forced.
346
* --dry-run mode for commit? (Or maybe just run with
347
check-command=false?)
349
* Generally, be a bit more verbose unless --silent is specified.
351
* Function that finds all changes to files under a given directory;
352
perhaps log should use this if a directory is given.
354
* XML attributes might have trouble with filenames containing \n and
355
\r. Do we really want to support this? I think perhaps not.
357
* Remember execute bits, so that exports will work OK.
359
* Unify smart_add and plain Branch.add(); perhaps smart_add should
360
just build a list of files to add and pass that to the regular add
363
* Function to list a directory, saying in which revision each file was
364
last modified. Useful for web and GUI interfaces, and slow to
365
compute one file at a time.
367
This will be done when we track file texts by referring to the
368
version that created them.
370
* Check locking is correct during merge-related operations.
372
* Perhaps attempts to get locks should timeout after some period of
373
time, or at least display a progress message.
375
* Split out upgrade functionality from check command into a separate
378
* Don't pass around command classes but rather pass objects. This'd
379
make it cleaner to construct objects wrapping external commands.
381
* Track all merged-in revisions in a versioned add-only metafile.
383
* ``pull --clobber`` should discard any local changes not present
384
remotely. Not generally what you want, but possibly useful when
385
you're just mirroring another branch and want to keep tracking it
386
even when they e.g. uncommit or make similar non-forward movements.
387
Also for push I suppose. Clobber may not be the best name, maybe
390
* ``uncommit`` command that removes a revision from the end of the
391
revision-history; just doing this is enough to remove the commit,
392
and a new commit will automatically be made against the
393
predecessor. This can be repeated.
395
It only makes sense to delete from the tail of history, not from the
398
The revision, its inventory and texts remain floating in the store.
399
We should perhaps add the revision to a list of removed-commits, so
400
that it can be restored or at least accounted for when checking
401
consistency. This file would not be versioned, and probably should
402
not propagate when branched.
404
If we track merged revisions then we need to update this list too.
405
If the list is stored in a weave it's easy (implicit): the version
406
of the list can remain but it won't be referenced anymore. It's
407
probably best to just store this list in a weave in the first place
414
* Generate annotations from current file relative to previous
417
- Is it necessary to store any kind of annotation where data was
420
* Update revfile_ format and make it active:
422
- Texts should be identified by something keyed on the revision, not
423
an individual text-id. This is much more useful for annotate I
424
think; we want to map back to the revision that last changed it.
426
- Access revfile revisions through the Tree/Store classes.
428
- Check them from check commands.
432
.. _revfile: revfile.html
434
* Hooks for pre-commit, post-commit, etc.
436
Consider the security implications; probably should not enable hooks
437
for remotely-fetched branches by default.
439
* Pre-commit check. If this hook is defined, it needs to be handled
440
specially: create a temporary directory containing the tree as it
441
will be after the commit. This means excluding any ignored/unknown
442
files, and respecting selective commits. Run the pre-commit check
443
(e.g. compile and run test suite) in there.
445
Possibly this should be done by splitting the commit function into
446
several parts (under a single interface). It is already rather
447
large. Decomposition:
449
- find tree modifications and prepare in-memory inventory
451
- export that inventory to a temporary directory
453
- run the test in that temporary directory
455
- if that succeeded, continue to actually finish the commit
457
What should be done with the text of modified files while this is
458
underway? I don't think we want to count on holding them in memory
459
and we can't trust the working files to stay in one place so I
460
suppose we need to move them into the text store, or otherwise into
461
a temporary directory.
463
If the commit does not actually complete, we would rather the
464
content was not left behind in the stores.
468
* GUI (maybe in Python GTK+?)
470
* C library interface
472
* Expansion of $Id$ keywords within working files. Perhaps do this in
473
exports first as a simpler case because then we don't need to deal
474
with removing the tags on the way back in.
1
For things to do in Bazaar development, see
3
https://bugs.launchpad.net/bzr/
4
https://blueprints.launchpad.net/bzr/
added
removed
TODO
