2008-08-07T00:13:32 <dreimark> re
2008-08-07T00:24:48 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4041:10ac3ad07faa 1.8-wsgi-fkrupicka/MoinMoin/web/ (cgi_interface.py _fallback_cgi.py): Take the single-file cgi-adapter from flup/pep333 for fallback
2008-08-07T00:24:49 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4042:ab601d804f26 1.8-wsgi-fkrupicka/MoinMoin/web/ (cgi_interface.py frontend.py): Factor out common frontend code
2008-08-07T00:24:50 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4043:793bf51b6c76 1.8-wsgi-fkrupicka/ (MoinMoin/web/scgi_interface.py wiki/server/moin.scgi): Added SCGI adapter
2008-08-07T00:24:51 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4044:0b5f28e79da7 1.8-wsgi-fkrupicka/ (3 files in 3 dirs): Point moin.wsgi to the new real application
2008-08-07T00:24:52 <zenhase> now, some commits
2008-08-07T00:24:55 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4045:40dd44a3fcff 1.8-wsgi-fkrupicka/ (4 files in 3 dirs): Get rid of the mod_python adapter
2008-08-07T00:24:58 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4046:ba0daaf24d63 1.8-wsgi-fkrupicka/MoinMoin/ (script/server/standalone.py server/server_standalone.py): Fixed: MoinMoin.server.server_standalone is gone, don't import ConfigClass from here anymore
2008-08-07T00:25:02 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4047:50d37349a9aa 1.8-wsgi-fkrupicka/MoinMoin/ (4 files in 2 dirs): Remove unused request/server code
2008-08-07T00:25:05 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4048:95decb0aeadd 1.8-wsgi-fkrupicka/MoinMoin/auth/ (http.py sslclientcert.py): Remove special cases for removed request/server code
2008-08-07T00:25:08 <CIA-53> Florian Krupicka <florian.krupicka@googlemail.com> default * 4049:c9240417af81 1.8-wsgi-fkrupicka/MoinMoin/action/serveopenid.py: Fixed: import of MoinMoinFinish
2008-08-07T00:59:25 <dreimark> gn
2008-08-07T00:59:35 <ThomasWaldmann> gn dennda
2008-08-07T00:59:39 <ThomasWaldmann> gn dreimark
2008-08-07T00:59:44 <dennda> gn
2008-08-07T01:00:01 <ThomasWaldmann> tab-completion error :)
2008-08-07T01:00:14 <dennda> realized
2008-08-07T02:59:17 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4624:21addb82af77 1.8-storage-cdenter/MoinMoin/storage/ (backends/fs17.py _tests/test_backends_fs17.py): fs17 storage backend: made code basically work, added tests
2008-08-07T02:59:17 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4625:40b44d1b242f 1.8-storage-cdenter/MoinMoin/storage/backends/fs17.py: merged storage repo
2008-08-07T09:06:28 <dreimark> moin
2008-08-07T09:09:56 <ThomasWaldmann> moin
2008-08-07T09:15:54 <ThomasWaldmann> dennda: the docstrings in storage/init read a bit strange
2008-08-07T09:16:25 <ThomasWaldmann> "An Item can have Metadata or Revisions, or both.
2008-08-07T09:16:33 <ThomasWaldmann> " for example.
2008-08-07T09:20:57 <ThomasWaldmann> I would rather say "An Item has item-level metadata (non-revisioned) and stores metadata and data revisioned.
2008-08-07T09:22:43 <ThomasWaldmann> And "An Item has the same number of revisioned data and revisioned metadata revisions."
2008-08-07T09:27:51 <ThomasWaldmann> There are also quite some German grammar rules being applied to the English docstrings.
2008-08-07T09:29:27 <ThomasWaldmann> For example, it is usually NOT x-y in English, neither xy, but x y (x and y being words).
2008-08-07T09:33:18 <ThomasWaldmann> almost all docstring need to be more precise. A vague description a first might be good to get a quick grasp of what it does, but the ABC is used by implementors of real backends and thus you need to describe what goes in, what happens and what comes out in a very precise and complete way.
2008-08-07T09:57:10 <TheSheep> A vague description at fist MIGHT be good... ;)
2008-08-07T09:57:26 <TheSheep> does RFC define MIGHT?
2008-08-07T09:58:17 <TheSheep> I always get it wrong, because in Polish "must not" means "doesn't have to"
2008-08-07T10:08:30 <ThomasWaldmann> same for german
2008-08-07T10:44:26 <starshine> I think it's about you MAY do this and you SHALL do that... you may do (are permitted but not required to) and shall do (are required) - might do == may do, imperitive intent.
2008-08-07T10:45:54 <TheSheep> I like more humane phrases, like "if the implementer chooses so" ;)
2008-08-07T10:46:18 <starshine> in idiomatic american english, "I might do that" normally describes uncertainty if they would but positive attitude about doing it, while smiling strongly and sayign the same thing implies that one really does intend to and also thanks someone for the suggestion.
2008-08-07T10:46:56 <TheSheep> "sounds like a good idea"
2008-08-07T10:46:59 <starshine> as opposed "SHOULD NOT" - I shouldn't do it + wicked grin == I plan to do it and I plan to be pleased when it annoys some specific person
2008-08-07T10:47:52 <starshine> ...and often thanks the other person in the conversation for the suggestion ;P
2008-08-07T10:48:20 <TheSheep> "Such is Human Perversity."
2008-08-07T10:49:37 <TheSheep> starshine: but that's a deeper layer than just language
2008-08-07T10:49:56 <TheSheep> starshine: you usually don't venture there in technical writing
2008-08-07T10:50:28 <TheSheep> I mean, you could do the same in any language, even with only gestures
2008-08-07T10:51:23 <starshine> actually in technical writing you have to be aware of it in order to make sure to avoid it
2008-08-07T10:51:33 <starshine> such nuances make otherwise obvious text unclear.
2008-08-07T10:51:58 * TheSheep imagines a technical manual written with lots of sarcasm and parabolas
2008-08-07T10:52:16 <starshine> mind you such politics *are* part of the underpinning of the RFCs, so, that is still applicable
2008-08-07T10:52:20 <starshine> woot
2008-08-07T10:52:22 <starshine> that'd be fun
2008-08-07T10:52:34 <starshine> copyright note has to mention the easyhaven attack shark
2008-08-07T10:53:04 <starshine> "and we didnt want to have a copyright note, but our lawyers said if we didn't, they'd sic the attack shark on us. so here it is"
2008-08-07T10:53:06 <TheSheep> "should your goal be total confusion, you SHOULD use sarcasm"
2008-08-07T10:53:29 <TheSheep> that doesn't sound right
2008-08-07T10:53:33 <starshine> s/should your/if your/ s/, you/, then you/
2008-08-07T10:54:05 <TheSheep> nah, too straightforward, omitting words is fun
2008-08-07T10:54:34 <TheSheep> I also think that RFCs has too little alliteration
2008-08-07T10:54:45 <starshine> else you MAY overanalyze your opponent reaction.
2008-08-07T10:56:04 <starshine> how about "foo is stored with revisions. bar and bang is stored without revisions."
2008-08-07T10:56:15 <starshine> s/bang is/bang are/
2008-08-07T10:57:14 <TheSheep> hmm... no rhythm :)
2008-08-07T10:57:38 <TheSheep> bar and bang's revisions stored are not :D
2008-08-07T10:58:23 <TheSheep> great master yoda's power is, but proper english sentence he not form can
2008-08-07T10:58:42 <starshine> 3 shall be the number of the counting, thou shalt count thou to 3. thou shalt not count to 4, neither shalt thou count to 2 (unless thou dost continue on to 3). 5 is right out.
2008-08-07T10:59:46 <starshine> so is this an object made of two parts, where the whole object is revision-stored, but the parts inside it one has also got internal revisions?
2008-08-07T11:00:10 <PawelPacana> moin
2008-08-07T11:00:16 <starshine> hi PawelPacana
2008-08-07T11:00:53 <TheSheep> PawelPacana: _o/
2008-08-07T11:07:07 * starshine .sleep()
2008-08-07T11:08:05 <TheSheep> good night starshine_away
2008-08-07T11:12:41 <ThomasWaldmann> moin TheSheep and PawelPacana
2008-08-07T11:13:05 <ThomasWaldmann> PawelPacana: my fs17 backend waits for some converter script to do a large scale practical test :)
2008-08-07T11:20:09 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4626:09575e289cfc 1.8-storage-cdenter/MoinMoin/storage/ (_tests/test_backends_fs17.py backends/fs17.py): fs17 storage backend: added attachment support, with tests
2008-08-07T11:20:21 <TheSheep> ThomasWaldmann: you are fast!
2008-08-07T11:20:51 <ThomasWaldmann> there's not much time left...
2008-08-07T11:21:55 <ThomasWaldmann> the good thing is that it is read-only and single-use, that makes things much easier :)
2008-08-07T11:28:13 <ThomasWaldmann> dennda: btw, can you explain why read() specialcase the chunksize<0 part?
2008-08-07T11:29:01 <ThomasWaldmann> for normal files and also for StringIO (file-like api) it is documented that it will return all content if it is negative.
2008-08-07T11:29:42 <ThomasWaldmann> so the backend does not need to solve that on its own, but can rely on it
2008-08-07T11:40:57 <zenhase> moin
2008-08-07T11:43:09 <ThomasWaldmann> hi zenhase
2008-08-07T11:57:56 <zenhase> hi thomas
2008-08-07T12:17:40 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4627:d19f3c416c24 1.8-storage-cdenter/MoinMoin/storage/ (_tests/test_backends_fs17.py backends/fs17.py): fs17 storage backend: fix pathes, make simple pages in a life wiki work
2008-08-07T12:21:48 <ThomasWaldmann> dennda: pages with Umlauts break
2008-08-07T12:35:51 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4628:9a08b475beba 1.8-storage-cdenter/MoinMoin/storage/ (_tests/test_backends_fs17.py backends/fs17.py): fs17 storage backend: improve names for Page/Attachment backend classes, fix tests
2008-08-07T12:43:15 <ThomasWaldmann> dennda: 110 failing tests in your repo!
2008-08-07T12:48:49 <ThomasWaldmann> some of the failures are cause by your usage of an empty MemoryBackend,
2008-08-07T12:49:17 <ThomasWaldmann> but quite many others are showing that your stuff is broken
2008-08-07T13:00:34 <ThomasWaldmann> PawelPacana: please also run the tests more often, including test_sourcecode
2008-08-07T13:01:10 <ThomasWaldmann> (and fix the trailing space)
2008-08-07T13:03:50 <ThomasWaldmann> PawelPacana: you also have that specialcasing of chunksize < 0 in your _read_revision_data. Does it make sense or did you just c&p it?
2008-08-07T13:11:45 <ThomasWaldmann> wtf is remove_item()?
2008-08-07T13:12:16 <ThomasWaldmann> (needed by the tests to clean up)
2008-08-07T13:13:52 <TheSheep> ThomasWaldmann: shouldn't be there, I think
2008-08-07T13:14:02 <TheSheep> ThomasWaldmann: in tests, I mean
2008-08-07T13:14:36 <TheSheep> ThomasWaldmann: where do you have it?
2008-08-07T13:15:41 <ThomasWaldmann> just run the tests
2008-08-07T13:15:56 <ThomasWaldmann> it's failing all over the place because of that
2008-08-07T13:16:40 <ThomasWaldmann> i intended to run test_security, because ACLs seem to be broken somehow, but I can't, because that also fails because of this
2008-08-07T13:18:30 <TheSheep> grep 'remove_item' *.py in storage/_test returns nothing...
2008-08-07T13:18:58 <ThomasWaldmann> yeah, but it should be there
2008-08-07T13:20:06 <ThomasWaldmann> (that doesn't mean that every backend must be able to implement this, but at least for testing we currently need it)
2008-08-07T13:20:23 <TheSheep> ./storage/backends/sql1.py: def remove_item(self, name):
2008-08-07T13:20:31 <TheSheep> only place where such a function exists
2008-08-07T13:20:58 <TheSheep> ThomasWaldmann: I don't recall anything like that from spec
2008-08-07T13:21:36 <ThomasWaldmann> what spec?
2008-08-07T13:21:52 <TheSheep> ThomasWaldmann: of the storage api
2008-08-07T13:22:07 <ThomasWaldmann> and why are I am discovering this and not the mentors?
2008-08-07T13:22:16 <TheSheep> ThomasWaldmann: I even remember johill explixitly saying there won't be such a thing
2008-08-07T13:22:48 <TheSheep> ThomasWaldmann: oh, we asked about it, about two weeks ago, if you recall, it was connected to the item renaming thing
2008-08-07T13:23:10 <ThomasWaldmann> that means we just break 100 tests just because
2008-08-07T13:23:15 <PawelPacana> sql1.py? this is just old piece of code moved to backends recently
2008-08-07T13:23:30 <PawelPacana> it lived somewhere else before
2008-08-07T13:24:01 <TheSheep> ThomasWaldmann: let me find the logs
2008-08-07T13:24:04 <PawelPacana> i don't think it has been touched recently
2008-08-07T13:24:41 <ThomasWaldmann> TheSheep: i don't need to read some "don't change the history" blabla in the logs
2008-08-07T13:24:45 <ThomasWaldmann> i need working tests
2008-08-07T13:25:37 <PawelPacana> ThomasWaldmann: the trailing spaces came first after switching to pydev, i'll fix that
2008-08-07T13:25:40 <TheSheep> ThomasWaldmann: so add some def remove_item(*args, **kw): pass ?
2008-08-07T13:25:49 <TheSheep> ThomasWaldmann: I mean, should we add it?
2008-08-07T13:26:04 <ThomasWaldmann> it needs to remove, not pass X)
2008-08-07T13:26:57 <TheSheep> ThomasWaldmann: whole item with all the history?
2008-08-07T13:27:20 <ThomasWaldmann> it is used by MoinMoin._tests.nuke_item :)
2008-08-07T13:41:19 <ThomasWaldmann> where is dennda ?
2008-08-07T13:50:49 <ThomasWaldmann> dennda: test_security also falls over PageEditor:1000 telling that I did not change page contents (for a new page that never existed before with non-empty content). That code works in moin/1.8 repo, so some change in PageEditor seems to have broken it.
2008-08-07T14:08:37 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4629:466a3dc7f2b6 1.8-storage-cdenter/MoinMoin/storage/ (_tests/test_backends.py backends/fs.py backends/memory.py): storage: fs/memory-backends: Remove special-casing when reading a negative amount of chunks. Built-in read() operations already handle negative chunksizes for us. Adding Test for that
2008-08-07T14:08:53 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4630:8278a93cb4ab 1.8-storage-cdenter/MoinMoin/storage/ (backends/fs17.py _tests/test_backends_fs17.py): Merge with my own branch, 1.8-storage-cdenter
2008-08-07T14:09:17 <dennda> ThomasWaldmann: Dennda is still semi-asleep and working on the points you critizised :)
2008-08-07T14:10:15 <dennda> ThomasWaldmann: < ThomasWaldmann> And "An Item has the same number of revisioned data and revisioned metadata revisions." <-- This is both, hard to understand and wrong
2008-08-07T14:11:12 <dennda> As for the general docstring critizism: I am considering switching to javadoc or something similar in the last week
2008-08-07T14:15:46 <ThomasWaldmann> dennda: you won't change our documentation system
2008-08-07T14:15:52 <dennda> Umm???
2008-08-07T14:16:01 <dennda> You are already using something similar in quite a few places
2008-08-07T14:16:03 <ThomasWaldmann> all I request is that docstrings are complete
2008-08-07T14:16:16 <ThomasWaldmann> epydoc
2008-08-07T14:16:23 <dennda> yes, or epydoc for that matter
2008-08-07T14:17:03 <ThomasWaldmann> and in general i suggest you read storage/* code again line by line and fix all stuff you'll find then
2008-08-07T14:17:44 <dennda> First thing I'm gonna do is: Take a shower
2008-08-07T14:17:47 * ThomasWaldmann fixes meanwhile the security issues
2008-08-07T14:33:40 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4631:73707c0349f0 1.8-storage-cdenter/MoinMoin/Page.py: security fix: don't let wrong api usage cause ACL security problems (TODO: fix AssertionErrors)
2008-08-07T14:33:44 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4632:8d2b6e5a7ca9 1.8-storage-cdenter/MoinMoin/storage/ (_tests/test_backends.py backends/fs.py backends/memory.py): merged storage repo
2008-08-07T14:34:22 <ThomasWaldmann> dennda: now fix the AssertionErrors you'll get (after making test_security work)
2008-08-07T14:36:21 <ThomasWaldmann> btw, if self._item is not None: ... does not help anything if you can't do anything making sense for the "else" branch.
2008-08-07T14:36:58 <ThomasWaldmann> for such cases you rather want to use assert to make sure all api usage is correct
2008-08-07T14:58:44 <dennda> ThomasWaldmann: Many of the tests fail because they assume an underlay in storage. So I suggest I work on the migration script with PawelPacana next. That way we can a) test your backend, since underlay is in 1.7-storage-format, b) achieve the actual goal of having a migration script, c) fix the tests
2008-08-07T15:02:14 <ThomasWaldmann> dennda: as i said, some are failing because of that, but many others are failing due to stuff being just broken
2008-08-07T15:02:22 <ThomasWaldmann> and I expect you fix them
2008-08-07T15:02:50 <ThomasWaldmann> and PawelPacana can please start working on the converter meanwhile
2008-08-07T15:03:59 <ThomasWaldmann> I expect fixing the tests will also surface some other problems that you also need to fix
2008-08-07T15:04:46 <ThomasWaldmann> you first please fix that tests that fail not just because you have an empty wiki, but due to other reasons, like AttributeErrors etc.
2008-08-07T15:05:45 <ThomasWaldmann> to have a better overview, I also suggest you use py.test.skip("Test broken because we use an empty test wiki") where appropriate
2008-08-07T15:07:09 * ThomasWaldmann finds it a bit strange to neither be able to do automated tests nor to be able to do reasonable practical testing because wiki is empty
2008-08-07T15:11:24 <ThomasWaldmann> brb
2008-08-07T15:16:59 <CIA-53> Pawel Pacana <pawel.pacana@gmail.com> default * 4611:c2a4fccf8e2d 1.8-mercurialbackend-ppacana/MoinMoin/storage/_tests/test_backends_hg.py: (hg tests) Added skips aforementioned by Alex: wrong version of hg, test_concurrent.
2008-08-07T15:17:00 <CIA-53> Pawel Pacana <pawel.pacana@gmail.com> default * 4612:6f3f8d8227b0 1.8-mercurialbackend-ppacana/MoinMoin/storage/backends/hg.py: (hg) Prevent invoking external merge tool.
2008-08-07T15:17:00 <CIA-53> Pawel Pacana <pawel.pacana@gmail.com> default * 4613:e44297de7856 1.8-mercurialbackend-ppacana/MoinMoin/storage/backends/hg.py: (hg) Removed obsoleted check on read().
2008-08-07T15:17:02 <CIA-53> Pawel Pacana <pawel.pacana@gmail.com> default * 4614:34530ffa3147 1.8-mercurialbackend-ppacana/MoinMoin/storage/ (5 files in 2 dirs): PEP8 fixes (trailing spaces).
2008-08-07T15:17:05 <CIA-53> Pawel Pacana <pawel.pacana@gmail.com> default * 4615:94e8648eaca4 1.8-mercurialbackend-ppacana/MoinMoin/storage/backends/research/ (_example_setparent.py test_memory_commit.py): (hg) Removed old hg examples.
2008-08-07T15:17:08 <CIA-53> Pawel Pacana <pawel.pacana@gmail.com> default * 4616:2408a3186e65 1.8-mercurialbackend-ppacana/MoinMoin/storage/ (2 files in 2 dirs): (storage) Further PEP8 fixes.
2008-08-07T15:41:59 <ThomasWaldmann> dennda: merging this ^^ will save you a few sourcecode checker failures :)
2008-08-07T16:04:40 <ThomasWaldmann> waldi: any link news?
2008-08-07T16:23:14 <CIA-53> Bastian Blank <bblank@thinkmo.de> default * 4279:6a1a1509be07 1.8-dom-bblank/MoinMoin/converter2/moinwiki_in.py: Moin Wiki input converter - Comment known problem in freelink interwiki support
2008-08-07T16:31:22 <zenhase> hmm, is there a way to exempt directories from test_sourcecode tests?
2008-08-07T16:32:04 <zenhase> i have about 20 failing tests with flup in contrib :o
2008-08-07T16:34:19 <ThomasWaldmann> zenhase: see the test source
2008-08-07T16:50:45 <dreimark> zenhase: which directories ?
2008-08-07T16:52:17 <dreimark> you can call it from a dir also
2008-08-07T16:52:42 <dreimark> sorry "got it" flup has that
2008-08-07T17:05:43 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4641:94481f511ce2 1.8-storage-cdenter/MoinMoin/storage/ (7 files in 3 dirs): Merge 1.8-mercurialbackend-ppacana
2008-08-07T17:18:14 <dreimark> byeongweon: the nesting problem isn't fixed
2008-08-07T17:18:31 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4642:22b86faa6aa0 1.8-storage-cdenter/MoinMoin/storage/backends/hg.py: storage: mercurial-backend: PEP8 Fix. Remove trailing whitespace
2008-08-07T17:18:36 <dennda> PawelPacana: What editor do you use?
2008-08-07T17:19:17 <dennda> test_sourcecode passes now ;)
2008-08-07T17:45:05 <ThomasWaldmann> dennda: btw, I fixed the acls not working problem
2008-08-07T17:45:31 <dennda> Wasn't that 4631:73707c0349f0?
2008-08-07T17:50:22 <ThomasWaldmann> no, more brokenness
2008-08-07T17:50:50 <ThomasWaldmann> that changeset was just that it does not just ignore ACLs if someone uses API wrong
2008-08-07T17:51:10 <ThomasWaldmann> (will raise AssertionError for that case and not keep quiet as it did before)
2008-08-07T17:51:33 <dennda> Well then commit the patch :)
2008-08-07T17:52:57 <ThomasWaldmann> I am just doing a review what else was broken :)
2008-08-07T17:58:06 <ThomasWaldmann> btw, 2nd error in 1 line of code X)
2008-08-07T18:13:49 <waldi> ThomasWaldmann: yes. i make an uri class to handle this whole problems
2008-08-07T18:16:38 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4643:16a2efa9fb66 1.8-storage-cdenter/MoinMoin/ (Page.py security/__init__.py): fix ACL processing, add assertion to avoid this kind of bugs
2008-08-07T18:16:38 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4644:2abe00edbf1e 1.8-storage-cdenter/MoinMoin/storage/backends/hg.py: merge storage repo
2008-08-07T18:20:27 <CIA-53> Bastian Blank <bblank@thinkmo.de> default * 4280:bc943db328ae 1.8-dom-bblank/MoinMoin/util/ (_tests/test_uri.py uri.py): Add generic URI class
2008-08-07T18:20:28 <CIA-53> Bastian Blank <bblank@thinkmo.de> default * 4281:0c3c7b9b25a5 1.8-dom-bblank/MoinMoin/converter2/moinwiki_in.py: Moin Wiki input converter - Split interwiki links directly, use correct escaping for all URIs
2008-08-07T18:20:29 <CIA-53> Bastian Blank <bblank@thinkmo.de> default * 4282:ca7e493f3c76 1.8-dom-bblank/MoinMoin/converter2/ (_tests/test_link.py link.py): Link converter - Use URI class for correct escaping
2008-08-07T18:23:28 <ThomasWaldmann> dennda: ^^
2008-08-07T18:23:43 <dennda> seen
2008-08-07T18:25:39 <ThomasWaldmann> now even the standard underlay pages work ;)
2008-08-07T18:26:13 <ThomasWaldmann> excep umlauts
2008-08-07T18:26:47 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4645:53817b625e42 1.8-storage-cdenter/MoinMoin/storage/backends/memory.py: storage: MemoryBackend: Adding basic implementation for news method
2008-08-07T18:27:13 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4646:d51d421df8f6 1.8-storage-cdenter/MoinMoin/ (Page.py security/__init__.py): Mergy my own repo, 1.8-storage-cdenter
2008-08-07T18:29:18 <CIA-53> Bastian Blank <bblank@thinkmo.de> default * 4283:29553876dcb4 1.8-dom-bblank/MoinMoin/converter2/ (_tests/test_moinwiki_in.py moinwiki_in.py): Moin Wiki input converter - Check for invalid interwiki links
2008-08-07T18:29:48 <ThomasWaldmann> dennda: i fix the caching encoding problem
2008-08-07T18:30:36 <dennda> ThomasWaldmann: How do you test standard underlay pages work? When browsing localhost:8080 the first thing I get is that AssertionError I need to fix
2008-08-07T18:31:12 <ThomasWaldmann> works for me :) with fs17
2008-08-07T18:32:38 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4647:e523f7278bd2 1.8-storage-cdenter/MoinMoin/caching.py: fix caching for non-ascii item names
2008-08-07T18:32:49 <dennda> How is that dependant on the backend??
2008-08-07T18:32:59 <dennda> That was meant to be only one question mark, btw.
2008-08-07T18:33:20 <ThomasWaldmann> i don't know. i use fs17 for data and it works.
2008-08-07T18:34:16 * dennda tests
2008-08-07T18:37:55 <dennda> ThomasWaldmann: Try a clean clone. As expected, I get the exactly same error with data_backend = fs17.FSPageBackend("/home/dennda/Desktop/data_einssieben")
2008-08-07T18:38:08 <dennda> By the way, I guess I need to point that to wiki/underlay, correct?
2008-08-07T18:38:56 <dennda> hm
2008-08-07T18:38:57 <dennda> ok
2008-08-07T18:39:04 <dennda> with "wiki/underlay" it works
2008-08-07T18:39:08 <dennda> nice
2008-08-07T18:39:40 <dennda> except it is a bit buggered now ;)
2008-08-07T18:42:32 <CIA-53> Bastian Blank <bblank@thinkmo.de> default * 4284:0e3eed5620f4 1.8-dom-bblank/MoinMoin/converter2/link.py: Link converter - Fix pagelinks
2008-08-07T18:54:30 <ThomasWaldmann> dennda: as a simple implementation for the item search does not depend on the backend, couldn't that be done in the base class instead of NotImplementedError?
2008-08-07T18:54:50 <ThomasWaldmann> and if one wants to optimize, one still can overwrite it
2008-08-07T18:55:24 <ThomasWaldmann> i could also copy the stuff from hg backend, but why duplicate X)
2008-08-07T19:00:29 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4648:3735825943aa 1.8-storage-cdenter/MoinMoin/storage/ (__init__.py backends/fs.py backends/hg.py backends/memory.py): storage: Backends: Small refactoring. Extracting generic backend.search() method and putting it into ABC.
2008-08-07T19:00:41 <dennda> Your wish is my command.
2008-08-07T19:01:57 <dennda> I need a plan on what to eat for late lunch. Criteria: Not too fat (had enough pizzas already), not too expensive, self-made, cooked fastly, without creating a mess. Ideas? :)
2008-08-07T19:02:48 <TheSheep> dennda: Schnitzel
2008-08-07T19:03:22 * TheSheep just made some
2008-08-07T19:03:57 <TheSheep> I guess spagetti would qualify as ft+messy
2008-08-07T19:03:59 <TheSheep> fat
2008-08-07T19:04:20 <dennda> I had noodles yesterday :)
2008-08-07T19:04:23 <grzywacz> depends on spaghetti
2008-08-07T19:04:30 <dennda> And Schnitzel... Seriously. Am I German? Oo
2008-08-07T19:04:47 <grzywacz> You can make a tasty spaghetti with a chicken and some good sauce that won't be very fat.
2008-08-07T19:04:49 <TheSheep> dennda: it's a good old polish meal, called Schabowy here
2008-08-07T19:05:38 <dennda> http://stompbox.typepad.com/blog/2008/08/arroz-con-pollo.html <-- Looks good, but too much work
2008-08-07T19:05:52 <TheSheep> dennda: I can't make anythng else except for cooked chicken in white sauce and scrambled eggs
2008-08-07T19:06:04 <dennda> heh
2008-08-07T19:06:10 <TheSheep> dennda: fruit salad!
2008-08-07T19:06:29 <dennda> I want a meal, not dessert
2008-08-07T19:07:45 <TheSheep> obion soup
2008-08-07T19:07:47 <TheSheep> onion
2008-08-07T19:07:59 <TheSheep> with cheese
2008-08-07T19:08:02 <dennda> I want a meal, not an appetizer
2008-08-07T19:08:12 <dennda> And besides, my roommate cooked that the other day
2008-08-07T19:08:16 <dennda> And it was terrible :D
2008-08-07T19:09:02 <TheSheep> hash browns it is then
2008-08-07T19:09:10 <TheSheep> with schnitzel ;)
2008-08-07T19:09:33 <dennda> hm probably
2008-08-07T19:09:36 <zenhase> hash browns? :o
2008-08-07T19:09:37 <dennda> I'll just go and get something
2008-08-07T19:09:44 <dennda> kartoffelpuffer
2008-08-07T19:10:14 <TheSheep> greasy++
2008-08-07T19:10:22 <dennda> great and easy?
2008-08-07T19:11:56 <TheSheep> something like that
2008-08-07T19:16:14 <ThomasWaldmann> zenhase: you are checking the wsgi state of affairs for twisted?
2008-08-07T19:20:44 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4649:a97af1f4e9f7 1.8-storage-cdenter/MoinMoin/action/AttachFile.py: kill useless link rels for attachments
2008-08-07T19:20:45 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4650:64c3a50234b7 1.8-storage-cdenter/MoinMoin/storage/backends/fs17.py: fs17 storage backend: fix for listing attachment items if there is no attachments/ dir
2008-08-07T19:20:56 <dennda> ok here's the plan
2008-08-07T19:21:19 <dennda> gnocchi + sauce al arrabiata + cheese == epic win (hopefully) and for only 3,54EUR
2008-08-07T19:21:23 * dennda is gone cooking
2008-08-07T19:25:18 <zenhase> ThomasWaldmann: yes
2008-08-07T19:25:53 <zenhase> i have started a twisted_interface which shall wrap up everything in a twisted.web2 resource
2008-08-07T19:26:29 <zenhase> deployment script will be something like a .tap or whattheyarecalled then
2008-08-07T19:27:27 <zenhase> dennda: possible, if spices are already available :)
2008-08-07T19:28:13 <dennda> zenhase: sure
2008-08-07T19:28:26 <dennda> but the sauce is already quite spicy
2008-08-07T19:39:31 <ThomasWaldmann> zenhase: doesn't twisted also use wsgi meanwhile?
2008-08-07T19:41:12 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4651:8a334900d0f7 1.8-storage-cdenter/MoinMoin/storage/backends/fs17.py: fs17 storage backend: remove dummy search_item, base class has a generic one now
2008-08-07T19:45:10 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4652:53ae547e88ce 1.8-storage-cdenter/MoinMoin/ (Page.py storage/__init__.py): fix RootPage.getPageList, clarify docstring to avoid such errors
2008-08-07T19:45:26 <ThomasWaldmann> bad names byte ...
2008-08-07T19:46:31 <ThomasWaldmann> brb
2008-08-07T19:47:01 <dennda> ThomasWaldmann: you did see that I already did that?
2008-08-07T19:47:01 <dennda> 19:00
2008-08-07T19:48:11 <dennda> arg, never mind
2008-08-07T20:04:58 <ThomasWaldmann> too much gnocchi error :P
2008-08-07T20:05:24 <ThomasWaldmann> heh
2008-08-07T20:38:54 <waldi> too much garlic error?
2008-08-07T20:50:57 <zenhase> dennda: how can the sauce be already spicey? you need the spices to make the sauce? :O
2008-08-07T20:51:27 <zenhase> dennda: or is it one of those 'pre-configured pasta sauces'?
2008-08-07T20:55:41 * ThomasWaldmann creates a 1.7 user backend
2008-08-07T21:00:16 <zenhase> ThomasWaldmann: regarding twisted wsgi: yes, they have a wsgi-resource in twisted.web2
2008-08-07T21:11:56 <ThomasWaldmann> zenhase: can you find out whether it is usable? if yes, would there be a reason to have special support?
2008-08-07T21:25:19 <ThomasWaldmann> dennda: on what are you working right now?
2008-08-07T21:25:24 <zenhase> ThomasWaldmann: well, it is usable i think, problem with twisted is that it's quite complicated to setup itself
2008-08-07T21:25:36 <zenhase> i mean twisted deployment in general
2008-08-07T21:25:48 <xorAxAx> it is?
2008-08-07T21:25:59 <ThomasWaldmann> it's twisted at least :))
2008-08-07T21:26:33 <zenhase> xorAxAx: well, with all those different deployment methods? tac-files, tap-files, twistd with it's myriad of options, tons of different servers :)
2008-08-07T21:26:44 <xorAxAx> yeah, tac and tap :)
2008-08-07T21:27:27 <CIA-53> Byeongweon [tasyblue@gmail.com] default * 3980:238eceaf862d 1.8-guieditor-mbyeongweon/MoinMoin/converter/text_html_text_moin_wiki.py: add comment about prossible error
2008-08-07T21:27:29 <CIA-53> Byeongweon [tasyblue@gmail.com] default * 3981:63a209dee4b3 1.8-guieditor-mbyeongweon/MoinMoin/converter/text_html_text_moin_wiki.py: fix delimiter process routine to handle close delimiter also. see MoinMoinBugs/GuiEditorCantNest
2008-08-07T21:28:10 <zenhase> i wonder if it's sufficient to point twisted-users to where the WSGI-application lives and just let them setup the rest for themselves
2008-08-07T21:28:35 <zenhase> i mean, twisted users have to be somehow professionals, no normal user would deploy twisted i suppose
2008-08-07T21:29:36 * ThomasWaldmann .oO(no normal user would deploy on IIS, must be professionals)
2008-08-07T21:30:51 <dennda> ThomasWaldmann: Those gnocchis were way to heavy and kicked me off my feet for a moment. Back in business now and reinforced by some coffee
2008-08-07T21:31:10 <dennda> Working on the AssertionErrors
2008-08-07T21:34:19 <ThomasWaldmann> byeongweon: you have been told at least 2 times that the delimiter stuff does not work that way
2008-08-07T21:34:56 <ThomasWaldmann> it is completely useless to find the "longest inner delimiter" and add 1
2008-08-07T21:36:06 <ThomasWaldmann> the only delimiter interesting for you is the outer one and you have to roundtrip it somehow
2008-08-07T21:36:59 <byeongweon> hm..
2008-08-07T21:37:46 <byeongweon> the way you told me is save original delimiter notation to pre tag somehow.
2008-08-07T21:37:53 <byeongweon> isn't it?
2008-08-07T21:38:02 <ThomasWaldmann> i can use {{{foo foo}}} to delimiter anything that does not contain foo}}}
2008-08-07T21:38:38 <ThomasWaldmann> yes
2008-08-07T21:39:24 <ThomasWaldmann> iirc some gui editor formatter calls use title attr to do that kind of stuff
2008-08-07T21:39:58 <ThomasWaldmann> e.g. look at interwiki links
2008-08-07T21:41:11 <byeongweon> but if somebody insert pre tag by gui editor and add delimiter notaion into pre tag.
2008-08-07T21:41:34 <byeongweon> to handle that case. I should some unique string to outer delimiter?
2008-08-07T21:42:02 <ThomasWaldmann> can you insert a pre section with gui editor?
2008-08-07T21:42:16 <byeongweon> yes.
2008-08-07T21:42:24 <ThomasWaldmann> how?
2008-08-07T21:42:49 <byeongweon> format -> formatted
2008-08-07T21:45:36 <ThomasWaldmann> did you try it with multiple lines, does it work?
2008-08-07T21:46:27 <byeongweon> oh. I understand. that only works for single line.
2008-08-07T21:46:27 <ThomasWaldmann> (for new pre sections generated with that, you method is ok. just use your method if you don't find delimiter info in title attr)
2008-08-07T21:46:41 <ThomasWaldmann> yeah, it's broken :|
2008-08-07T21:47:11 <ThomasWaldmann> add it to your list in case you don't have that yet
2008-08-07T21:48:28 <byeongweon> ok. I will make 2 way of process delimiter
2008-08-07T21:49:28 <byeongweon> and put formatted out and modify work for multi line.
2008-08-07T21:54:09 <xorAxAx> ThomasWaldmann: why does it need to be roundtripped?
2008-08-07T22:00:53 <dennda> http://server.mpathy.de/winfraud.gif
2008-08-07T22:00:54 <ThomasWaldmann> so that it roundtrips :D if you use {{{foo, you don't want gui editor to change it to {{{{{
2008-08-07T22:04:53 <xorAxAx> why?
2008-08-07T22:05:14 <xorAxAx> you could also ask for whitespace roundtripping
2008-08-07T22:05:22 <xorAxAx> probably similarly useful
2008-08-07T22:05:38 <xorAxAx> (used to structure the page, doesnt have any visible effect)
2008-08-07T22:08:16 <ThomasWaldmann> the gui editor should at least try to not make bigger diffs than necessary
2008-08-07T22:09:07 <xorAxAx> yeah, but these issue only contributes minimally to the resulting diff
2008-08-07T22:09:17 <xorAxAx> i would suggest to prioritise it very lowly
2008-08-07T22:10:35 <ThomasWaldmann> ok
2008-08-07T22:10:49 <ThomasWaldmann> fixing creation of pre via gui is more important
2008-08-07T22:21:46 * ThomasWaldmann .oO(1.7 r/o user backend works)
2008-08-07T22:22:06 <ThomasWaldmann> where's the converter? :)
2008-08-07T22:39:30 <CIA-53> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4653:aeb2ed63f6e5 1.8-storage-cdenter/MoinMoin/ (storage/backends/fs17.py user.py): fs17 storage backend: added user backend, fix wrong API usage
2008-08-07T22:43:43 <dennda> You didn't want me to write it :)
2008-08-07T22:45:01 <ThomasWaldmann> yeah, still waiting for fixes :)
2008-08-07T22:45:07 <dennda> coming
2008-08-07T22:45:18 <ThomasWaldmann> PawelPacana: are you working on the converter?
2008-08-07T22:51:31 <CIA-53> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 4654:04d26a5f83bc 1.8-storage-cdenter/MoinMoin/ (14 files in 5 dirs): storage: Partially applying patch from JohannesBerg. Killing obsolete nuke_page usage.
2008-08-07T22:54:05 <dreimark> re
2008-08-07T22:54:17 <ThomasWaldmann> obsolete?
2008-08-07T22:54:19 <dreimark> xorAxAx: do you review all closed gui bugs ?
2008-08-07T22:55:51 <dreimark> I can look at some but won't do it duplicated
2008-08-07T22:56:09 <dreimark> looks like I have time my student seems to be a frog
2008-08-07T22:56:41 <ThomasWaldmann> hehe
2008-08-07T22:57:07 <ThomasWaldmann> did she get green and eats flies?
2008-08-07T22:59:51 <dennda> ThomasWaldmann: Yeah, obsolete
2008-08-07T23:00:14 <dennda> The tests are run on a MemoryBackend, so that is not needed
2008-08-07T23:00:29 <ThomasWaldmann> dreimark: "Sei kein Frosch!" == "Don't chicken out!" :)
2008-08-07T23:00:35 <dennda> And it was causing tests to fail
2008-08-07T23:01:10 <xorAxAx> dreimark: no, i didnt review them
2008-08-07T23:01:10 <ThomasWaldmann> dennda: so if i create a page A in test1, i can create another page A in test2?
2008-08-07T23:01:33 <ThomasWaldmann> i rather expected the memorybackend keeps data while the engine runs.
2008-08-07T23:01:57 <xorAxAx> dreimark: i am doing code review, spot checks and integration tests, though
2008-08-07T23:02:55 <dennda> Well, if you create *another* MemoryBackend you can obviously have the same name there, too. But the backends are set up in test/wikiconfig.py (As of my pairprogramming-session with xorAxAx)
2008-08-07T23:02:56 <xorAxAx> btw, i think we should meet on next wednesday as well
2008-08-07T23:03:24 <xorAxAx> dennda: well, as we discussed, we need to fill the memory backend with a testset
2008-08-07T23:03:25 <ThomasWaldmann> dennda: so it doesn't help at all that it is a memory backend
2008-08-07T23:03:40 <xorAxAx> whats the problem, ThomasWaldmann?
2008-08-07T23:03:56 <ThomasWaldmann> the problem is removing stuff before creating an alternative
2008-08-07T23:04:43 <xorAxAx> removing what? :)
2008-08-07T23:04:58 <dreimark> xorAxAx: ok, I will sign the fixed ones and reopen the others
2008-08-07T23:05:00 <ThomasWaldmann> nuke_page, remove_item
2008-08-07T23:05:20 <xorAxAx> dreimark: you mean if you find bugs that were closed by him even though they are still there?
2008-08-07T23:05:24 <xorAxAx> ThomasWaldmann: please elaborate
2008-08-07T23:06:17 <ThomasWaldmann> if test1 created a page A and nuked it at the end, test2 could do just the same without any problem
2008-08-07T23:06:41 <ThomasWaldmann> now all nuke stuff was removed
2008-08-07T23:06:55 <dreimark> xorAxAx: I am not sure if there are more than the nested one, but that one I've already reopened twice
2008-08-07T23:07:12 <xorAxAx> dreimark: ok, please also reflect that on his page
2008-08-07T23:07:18 <ThomasWaldmann> that means we might have new failures just because of that
2008-08-07T23:07:35 <xorAxAx> ThomasWaldmann: you mean in the storage unrelated code?
2008-08-07T23:08:09 <xorAxAx> yes, it would be nice if the normal test classes get a new backend every time as well
2008-08-07T23:08:23 <ThomasWaldmann> everywhere where the removed code was used
2008-08-07T23:08:39 <xorAxAx> as soon as we have a test set that can be copied into a memory backend, we can start with a clean datadir
2008-08-07T23:08:45 <xorAxAx> ThomasWaldmann: which code was removed, i dont understand
2008-08-07T23:08:56 <ThomasWaldmann> nuke_page, remove_item
2008-08-07T23:09:10 <ThomasWaldmann> or just read the storage repo cs
2008-08-07T23:10:24 <ThomasWaldmann> I also don't get why remove_item is not implemented. there are occasions a storage client might want to do that
2008-08-07T23:10:55 <ThomasWaldmann> (that doesn't mean we expose that to Joe Doe on the web, but we could use it within moin and tools)
2008-08-07T23:11:20 <xorAxAx> ThomasWaldmann: we already talked about that
2008-08-07T23:11:29 <xorAxAx> and implementing that is easy, its just done on a different level:
2008-08-07T23:11:56 <xorAxAx> we get a few advantages by assuming that the set of items is only getting new items at normal runtime
2008-08-07T23:12:10 <zenhase> hmm, i just realized, that the twisted.web2 wsgi adapter isn't available in standard-twisted anyway :o
2008-08-07T23:12:13 <xorAxAx> if items should be removed, we can easily just run a migration script that removes the particular items
2008-08-07T23:12:22 <zenhase> web2 is only in the svn-trunk of twisted
2008-08-07T23:12:23 <xorAxAx> zenhase: oh, web2 ... the badly maintained one :)
2008-08-07T23:12:43 <xorAxAx> note that this could happen from a superuser action as well
2008-08-07T23:13:12 <ThomasWaldmann> so I have to do low level stuff because api lacks?
2008-08-07T23:13:20 <xorAxAx> no, not low level at all
2008-08-07T23:13:30 <ThomasWaldmann> but?
2008-08-07T23:13:31 <xorAxAx> its rather high level
2008-08-07T23:13:40 <xorAxAx> in fact, even higher level than storage api
2008-08-07T23:13:48 <zenhase> xorAxAx: well, since it isn't even in the twisted distribution of twisted, i wouldn't call it badly maintained ... more like 'never finished' ;)
2008-08-07T23:13:56 <xorAxAx> but its O(n) because the storage is rebuilt
2008-08-07T23:14:09 <xorAxAx> the assumption is that items are seldomly removed or removed in bulk quantities
2008-08-07T23:14:19 <xorAxAx> an action could e.g. set a metadata flag
2008-08-07T23:14:22 <xorAxAx> that hides the item
2008-08-07T23:14:37 <xorAxAx> which is then used to remove all bulk pages
2008-08-07T23:14:50 <xorAxAx> ThomasWaldmann: a nested loop (migration function) is higher level than the storage api
2008-08-07T23:15:38 <ThomasWaldmann> ok, hiding is clear, but it means that i have to completely rebuild the storage to just remove some items completely
2008-08-07T23:15:55 <ThomasWaldmann> sorry, but that doesn't sound very sane
2008-08-07T23:15:59 <xorAxAx> yes, but when do you remove items completly? as i said, the rebuild can be deferred
2008-08-07T23:16:08 <xorAxAx> why?
2008-08-07T23:16:26 <xorAxAx> please present a use case
2008-08-07T23:16:55 <ThomasWaldmann> e.g. items with illegal content
2008-08-07T23:17:11 <xorAxAx> they are quickly flagged by the aforementioned action
2008-08-07T23:17:14 <ThomasWaldmann> and btw, note: we don't have that hide functionality
2008-08-07T23:17:16 <xorAxAx> by the superuser of the wiki
2008-08-07T23:17:34 <xorAxAx> well, its easy to add in the place that checks whether an item is deleted
2008-08-07T23:17:43 <ThomasWaldmann> so this is just another case of removing before solving otherwise
2008-08-07T23:17:57 <xorAxAx> an item is either existing, deleted or to be purged
2008-08-07T23:18:16 <xorAxAx> and not many places in the code (maybe 1 or 2 functions) are checking whether an item is deleted, so it should be easy
2008-08-07T23:18:22 <ThomasWaldmann> is this somewhere in the documentation?
2008-08-07T23:18:26 <xorAxAx> what?
2008-08-07T23:18:46 <xorAxAx> the purging is just an idea, as i just said, its trivial to implement
2008-08-07T23:18:48 <ThomasWaldmann> the reasoning for not having remove_item and how to solve
2008-08-07T23:18:55 <xorAxAx> ah, no
2008-08-07T23:19:00 <xorAxAx> its in the spec
2008-08-07T23:19:11 <xorAxAx> which hasnt been put on a wiki page for the last months
2008-08-07T23:19:15 <ThomasWaldmann> decide :)
2008-08-07T23:19:21 <xorAxAx> ?
2008-08-07T23:19:32 <ThomasWaldmann> either it is documented somewhere or not
2008-08-07T23:19:39 <xorAxAx> i never said it is documented
2008-08-07T23:19:51 <xorAxAx> in the spec != documented
2008-08-07T23:20:06 <ThomasWaldmann> a spec is documentation
2008-08-07T23:20:11 <xorAxAx> no
2008-08-07T23:20:16 <ThomasWaldmann> sure it is
2008-08-07T23:20:24 <xorAxAx> "
2008-08-07T23:20:29 <xorAxAx> A specification is an explicit set of requirements to be satisfied by a material, product, or service"
2008-08-07T23:20:56 <xorAxAx> but this set of requirements hasnt been seen in any kind of article in the last months
2008-08-07T23:21:13 <xorAxAx> you can only infer it from reading the code, the old design documents written by johill and his words
2008-08-07T23:21:30 <xorAxAx> even though i have been trying to motivate him to write it down
2008-08-07T23:21:32 <ThomasWaldmann> dennda: btw, docs/specs/whateveryoucallit - there is some stuff in the toplevel dir of your repo
2008-08-07T23:21:52 <dennda> Yes
2008-08-07T23:21:56 <ThomasWaldmann> I expect that you clean up there.
2008-08-07T23:22:11 <xorAxAx> and i strongly encourage to do so because it is very important to document design decisions (like this one)
2008-08-07T23:22:18 <dennda> There's much to be cleaned up. I planned to do that in the end
2008-08-07T23:22:18 <ThomasWaldmann> e.g. in the BUGS document, the bugs need to be reviewed
2008-08-07T23:23:20 <ThomasWaldmann> "the end is near" :)
2008-08-07T23:26:50 * ThomasWaldmann continues bug hunting
2008-08-07T23:28:05 * dennda too
2008-08-07T23:32:50 <xorAxAx> byeongweon: ping
2008-08-07T23:41:17 <xorAxAx> byeongweon: please report current status on your wikipage
2008-08-07T23:42:56 <dennda> ThomasWaldmann: You said you don't get the ACL-related AssertionError with your backend
2008-08-07T23:43:46 <dennda> ThomasWaldmann: Did you try opening the Page "asfjhAFGHWEHGbajghajg"?
2008-08-07T23:43:48 <dennda> :)
2008-08-07T23:44:26 <dennda> if you only tried that with pages from underlay, it is clear why it works
2008-08-07T23:44:55 <dennda> but I think the assertion you added to Page.getACL is wrong
2008-08-07T23:51:02 <byeongweon> xorAxAx: ok
2008-08-07T23:51:25 <byeongweon> xorAxAx: what kind of status?
2008-08-07T23:51:56 <xorAxAx> byeongweon: well, what you are currently doing, how you plan to fix the image regressions, that you will handle the regressions found by reimar etc.
2008-08-07T23:52:48 <ThomasWaldmann> dennda: yes, i get it for non-existing pages
2008-08-07T23:52:55 <dennda> ThomasWaldmann: Which is obvious
2008-08-07T23:53:10 <dennda> ThomasWaldmann: because the assertion is wrong
2008-08-07T23:53:24 <ThomasWaldmann> the code before was more wrong :)
2008-08-07T23:53:36 <byeongweon> xorAxAx: should I just write them to my wiki page?
2008-08-07T23:54:09 <dennda> why? it worked
2008-08-07T23:54:11 <xorAxAx> byeongweon: you should amend the existing page. did you see reimars bug reports?
2008-08-07T23:54:17 <xorAxAx> the ones he reopened
2008-08-07T23:54:22 <dennda> and I don't see anything wrong with it at first glance
2008-08-07T23:55:32 <byeongweon> xorAxAx: yes I saw his bug reports.
2008-08-07T23:56:16 <xorAxAx> byeongweon: which issue are you working on currently?
2008-08-07T23:56:48 <byeongweon> I'm currently doing the bug that thomas told me.
2008-08-07T23:57:02 <byeongweon> delimiter one.
2008-08-07T23:58:01 <xorAxAx> well, as i said its not very important
2008-08-07T23:58:17 <xorAxAx> if you mean the one where the {{{foo is changed to {{{{
2008-08-07T23:58:32 <xorAxAx> but there are other worse bugs related to {{{ stuff
2008-08-07T23:58:53 <ThomasWaldmann> dennda: it didn't give correct acls if self._item was not initialized correctly
2008-08-07T23:59:51 <dennda> ThomasWaldmann: If the page doesn't exist, there is no self._item MoinMoin: MoinMoinChat/Logs/moin-dev/2008-08-07 (last edited 2008-08-06 22:15:01 by IrcLogImporter)
