aboutsummaryrefslogtreecommitdiffstats
path: root/libbe
Commit message (Collapse)AuthorAgeFilesLines
* Closed a few old bugs that seem outdated/dealt with.W. Trevor King2008-12-041-1/+1
|
* Fixed missing directory error when merging comments onto a bare bug.W. Trevor King2008-12-041-0/+2
| | | | | | | | | If bug_A had no comments (and so, no comment directory), changing comment settings before saving raised missing directory errors. save_settings had previously assumed the .be/bugs/XYZ/comments/ directory existed, which wasn't true for comment-less bugs. Now it checks, and creates the directory if necessary.
* becommands/severity and status now handle --complete appropriately.W. Trevor King2008-12-041-0/+2
| | | | | | | | | | I also disabled interspersed options and arguments in cmdutils.CmdOptionParser. See http://docs.python.org/library/optparse.html Now $ be severity xyz --complete returns available severities. It had previously returned --help --complete
* Per-tree status levels working.W. Trevor King2008-12-043-19/+57
|
* bug severity verification now works with per-tree severities.W. Trevor King2008-12-042-6/+8
| | | | | | | | | | I adjusted the YAML format following http://pyyaml.org/ticket/11 Unicode support To remove '!!python/unicode' escapes and allow unicode in the output. We can always have unicode in the output because the output is encoded (as per the BugFile.encoding setting) before being sent to the outside world.
* Per-tree settings now passed into bug module.W. Trevor King2008-12-042-7/+27
| | | | | | | | becommands/severity gets the configured settings appropriately. Todo: adjust setting-validation to compare against the current values. setup becommands/severity to --complete severities.
* Added per-tree configurable severities.W. Trevor King2008-12-043-27/+38
| | | | | | | They currently have no effect, but you can see them with $ be set There's a lot of information in this one 'settings' variable. I think set will have to be specialized to handle arrays smoothly...
* Added support for bug reporter field.W. Trevor King2008-12-021-0/+1
| | | | Closes e2f6514c-5f9f-4734-a537-daf3fbe7e9a0
* Fixed comparison bug in Bug._get_time.W. Trevor King2008-12-021-3/+8
|
* Converted Comment to the settings_object system.W. Trevor King2008-12-021-56/+28
|
* Bug moved to the settings_object system.W. Trevor King2008-12-021-56/+31
|
* Added decorator-style properties to bugdir. Created settings_object module.W. Trevor King2008-12-025-157/+467
| | | | | | | | settings_object.SavedSettingsObject encapsulates some of the common settings functionality in the BE BugDir, Bug, and Comment classes. It's a bit awkward due to the nature of scoping in python subclasses, but it's better than reproducing this code in each of the above classes. Now I need to move Bug and Comment over to *this* system ;).
* Added decorator-style properties to libbe/comment.py.W. Trevor King2008-12-013-77/+209
| | | | | Also some typo corrections and some reworking of bug/bugdir to better support the lazier loading.
* Added libbe/properties to make property management easier.W. Trevor King2008-11-282-105/+586
| | | | | libbe/bug has been moved over to the new system. comment and bugdir still to go.
* Added becommands/comment completion.W. Trevor King2008-11-273-4/+12
|
* Added bugid_args option to cmdutil.default_complete.W. Trevor King2008-11-271-3/+23
| | | | | | | | Now most of the bug-id arguments support Bash completion. Since there will hopefully be lots of bugs in the database, I decided to filter the list of available bugs. Currently, we just auto-complete active bugs for most commands, with the exceptions of open (obviously) and status (which needs to work on all types of bugs).
* Command completion simplified and working for list, dummies for other cmds.W. Trevor King2008-11-271-2/+22
| | | | | | | | All the other commands currently use default_complete(), which has no effect other than catching the --complete option and effectively aborting execution. This closes 8e1bbda4-35b6-4579-849d-117b1596ee99
* Merged --commands and --options into --complete. Simpler that way.W. Trevor King2008-11-271-12/+8
|
* Basic bash completion is now supported.W. Trevor King2008-11-271-3/+26
| | | | I'm still working on a clean implementation though...
* Added -a/n/m/r to becommands/diff.py.W. Trevor King2008-11-251-2/+2
|
* Added libbe/encoding.py to wrap input/output/file access appropriately.W. Trevor King2008-11-2514-159/+363
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | I borrowed most of the code for this. get_encoding() is from Trac http://trac.edgewall.org/browser/trunk/trac/util/datefmt.py format_datetime() Trac has a BSD license http://trac.edgewall.org/wiki/TracLicense I don't know if such a small snippet requires us to "reproduce the above copyright" or where we need to reproduce it if it is needed. The stdout/stdin replacement code follows http://wiki.python.org/moin/ShellRedirectionFails Because of the stdout replacement, the doctests executes now need an optional 'test' argument to turn off replacement during the doctests, otherwise doctest flips out (since it had set up stdout to catch output, and then we clobbered it's setup). References: http://wiki.python.org/moin/Unicode http://www.amk.ca/python/howto/unicode http://www.python.org/dev/peps/pep-0100/ I also split libbe/editor.py off from libbe.utility.py and started explaining the motivation for the BugDir init flags in it's docstring.
* Oops, should be passing libbe.tree.Tree.sort() args to children. Fixed.W. Trevor King2008-11-251-1/+4
|
* Added becommands/merge to join duplicate bugs.W. Trevor King2008-11-242-17/+39
|
* Adjusted becommands/set to handle user_id properly.W. Trevor King2008-11-241-7/+10
|
* Fixed broken doctest and inconsitent indentation from not wrapping comments.W. Trevor King2008-11-241-2/+4
| | | | | Also emptied becommands/__init__.py. I didn't understand the plugin interface when I wrote it.
* Removed auto-wrapping from comment.Comment.string().W. Trevor King2008-11-241-2/+4
| | | | | | | | | | It makes tracebacks almost illegible. I doubt markup/markdown systax or auto-formatting is really useful, since bugs-reports are ususally a short comment and a traceback. I also closed a4d38ba7-ec28-4096-a4f3-eb8c9790ffb2 and 7bfc591e-584a-476e-8e11-b548f1afcaa6, which have probably been fixed for a long time...
* Added Bug.comments(), BugDir.has_bug() & cleaned up diff.diff().W. Trevor King2008-11-244-23/+33
| | | | + some other minor fixes and cleanups.
* Replaced direct filesystem read from bugdir.py with RCS mediated read.W. Trevor King2008-11-243-60/+13
| | | | | | | | | Also replaced utility.FileString with StringIO() in cmdutil.py, which allowed the removal of utility.FileString and utility.get_file. The only remaining file().read() outside the RCS framework is the read in utility.editor_string(), but should probably not go through the RCS.
* Added 'allow_no_rcs' flag to RCS file system access methods.W. Trevor King2008-11-245-125/+204
| | | | | Now mapfile access has fewer special cases, and there is less redundant rcs.add/update code.
* Created bugdir.MultipleBugMatches so bugdir no longer imports cmdutil.W. Trevor King2008-11-232-4/+9
|
* Removed outdated beuuid import from libbe/bugdir.pyW. Trevor King2008-11-231-1/+0
|
* libbe/diff.diff() was missing newly created bugs. Fixed.W. Trevor King2008-11-231-2/+5
| | | | Also added blank lines to separate the new/modified/removed groups.
* Go back to lazy bug loading to get execution speed back up.W. Trevor King2008-11-234-38/+62
| | | | Fixes bug b3c6da51-3a30-42c9-8c75-587c7a1705c5
* Improved user-id saving/loading/caching & save user-id into duplicate bugdirs.W. Trevor King2008-11-231-16/+19
| | | | | | Fixes the duplicate bugs a403de79-8f39-41f2-b9ec-15053b175ee2 c894f10f-197d-4b22-9c5b-19f394df40d4
* Added bugdir user-id caching and save/load from settings file.W. Trevor King2008-11-232-1/+36
|
* Fixed "'NoneType' object is not callable" error in libbe/utility/Dir.__del__W. Trevor King2008-11-231-2/+5
|
* Added archive/project init code for `./test_usage.sh arch`.W. Trevor King2008-11-233-9/+1
| | | | Also some minor cleanups.
* Explicit rcs.cleanup() in bugdir test.W. Trevor King2008-11-224-19/+15
| | | | | | | | | | | | | Don't use del(rcs), because if there was an error, there is still a reference to rcs in the traceback, so it is never cleaned up. This can leave the external archive cluttering up your Arch install if you're using the Arch backend. See the __del__ documentation http://python.active-venture.com/ref/customization.html#l2h-175 for details. Also fixed some out-of-date method names in libbe.diff
* libbe.utility.Dir was complaining of a missing shutil in __del__().W. Trevor King2008-11-221-1/+2
| | | | | Presumably the cleanup process removed shutil before cleaning up the Dir. Now Dir keeps a local reference.
* Created and fixed bug 496edad5-1484-413a-bc68-4b01274a65eb.W. Trevor King2008-11-224-34/+126
| | | | | | | | I figured out why Arch was complaining. For non-Arch users, file system access has been tweaked a bit see the BugDir doc string for details. Also, you should now set BugDir.rcs instead of .rcs_name. .rcs_name automatically tracks changes in .rcs (the reverse of the previous situation), so read from whichever you like.
* Oops, these new submods are used by the new, classified Bug & BugDir.W. Trevor King2008-11-223-0/+602
| | | | I'd forgotten tell bzr...
* Another major rewrite. Now BugDir, Bug, and Comment are more distinct.W. Trevor King2008-11-2110-512/+477
| | | | | | | | | | | | | | | I pushed a lot of the little helper functions into the main classes, which makes it easier for me to keep track of what's going on. I'm now at the point where I can run through `python test.py` with each of the backends (by changing the search order in rcs.py _get_matching_rcs) without any unexpected errors for each backend (except Arch). I can also run `test_usage.sh` without non-Arch errors either. However, don't consider this a stable commit yet. The bzr backend is *really*slow*, and the other's aren't blazingly fast either. I think I'm rewriting the entire database every time I save it :p. Still, it passes the checks. and I don't like it when zounds of changes build up.
* names.creator() replaced by rcs.get_user_id().W. Trevor King2008-11-191-6/+0
|
* Merged bug.new_comment into bug.Bug.new_comment.W. Trevor King2008-11-191-8/+4
|
* Moved bug.new_bug code into bugdir.BugDir.new_bug.W. Trevor King2008-11-192-8/+4
| | | | | Also removed explicit comparisons from beweb/controllers.py, since they are now built into the Bug.__cmp__ method.
* Removed libbe/tests.pyW. Trevor King2008-11-191-31/+0
| | | | | | Not needed with stronger test.py. It's only use would be testing an installed libbe in place... Maybe that is useful enough? If so, we can bring it back.
* Major rewrite of RCS backends. RCS now represented as a class.W. Trevor King2008-11-1817-692/+1123
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Lots of changes and just one commit. This started with bug dac91856-cb6a-4f69-8c03-38ff0b29aab2, when I noticed that new bugs were not being added appropriately with the Git backend. I'd been working with Git trouble before with bug 0cad2ac6-76ef-4a88-abdf-b2e02de76f5c, and decided things would be better off if I just scrapped the current RCS architecture and went to a more object oriented setup. So I did. It's not clear how to add support for an RCS backend: * Create a new module that - defines an inheritor of rsc.RCS, overriding the _rcs_*() methods - provide a new() function for instantizating the new class - defines an inheritor of rcs.RCStestCase, overiding the Class attribute - defines 'suite' a unittest.TestSuite testing the module * Add your new module to the rest in rcs._get_matching_rcs() * Add your new module to the rest in libbe/tests.py Although I'm not sure libbe/tests.py is still usefull. The new framework clears out a bunch of hackery that used to be involved with supporting becommands/diff.py. There's still room for progress though. While implementing the new verision, I moved the testing framework over from doctest to a doctest/unittest combination. Longer tests that don't demonstrate a function's usage should be moved to unittests at the end of the module, since unittest has better support for setup/teardown, etc. The new framework also revealed some underimplented backends, most notably arch. These backends have now been fixed. I also tweaked the test_usage.sh script to run through all the backends if it is called with no arguments. The fix for the dac bug turned out to be an unflushed file write :p.
* Fixed another bug in git.strip_git() (bug 0cad).W. Trevor King2008-11-161-1/+1
| | | | | Also added git mode to test_usage.sh. I'll go through and add modes for the other RCSs...
* Merged in my git.py changed fixing Hubert Chathi's git set-root bug (0cad).W. Trevor King2008-11-161-2/+9
|\
| * Oops... *Now* I've fixed 0cadW. Trevor King2008-11-131-1/+5
| |
generated by cgit (git 2.34.1) at 2024-09-21 23:00:36 +0000