2008-03-07T01:05:08 <ThomasWaldmann> looks like noone uses wsgi AND twikidraw
2008-03-07T01:09:38 <johill> considering how many people use twikidraw I don't find that surprising
2008-03-07T01:10:21 <ThomasWaldmann> twikidraw uses querystring args AND POSTed data when saving
2008-03-07T01:10:31 <ThomasWaldmann> that was broken for wsgi
2008-03-07T01:11:09 <ThomasWaldmann> (and I am asking myself if evaluating both is the usual thing as cgifieldstorage does not do it)
2008-03-07T01:12:50 <TheSheep_> it doesn't?
2008-03-07T01:12:55 <TheSheep_> that's weird
2008-03-07T01:13:27 <TheSheep_> I guess you want to evaluate them separately
2008-03-07T01:14:12 <ThomasWaldmann> yes, that would avoid conflicts
2008-03-07T01:14:33 <ThomasWaldmann> but for simplicity, moin does just update the same form dict
2008-03-07T01:15:12 <ThomasWaldmann> also, the wsgi code calls fieldstorage with
2008-03-07T01:15:31 <ThomasWaldmann> form = cgi.FieldStorage(fp=self.stdin, environ=self.env, keep_blank_values=1)
2008-03-07T01:16:23 <TheSheep_> yes, cgi need the Content-length header, in particular
2008-03-07T01:16:38 <ThomasWaldmann> thfcgi also does it with keep_blank
2008-03-07T01:16:38 <TheSheep_> needs
2008-03-07T01:16:43 <ThomasWaldmann> all others dont
2008-03-07T01:16:56 <TheSheep_> ah
2008-03-07T01:17:16 <TheSheep_> hmm.. conistency might be desired
2008-03-07T01:17:26 <ThomasWaldmann> (or dont use fieldstorage, have to review that)
2008-03-07T01:18:22 <ThomasWaldmann> the question then is, what's better for moin?
2008-03-07T01:19:16 <TheSheep_> I'd guess that if you don't want a field to appear in a posted form, then you don't give it a name
2008-03-07T01:19:43 <TheSheep_> if it has a name it should appear, even if the value is blank
2008-03-07T01:21:23 <TheSheep_> (I actually spent several hours debugging a toy wiki script in which I forgot to give a name to the textarea)
2008-03-07T01:22:22 <ThomasWaldmann> ok, so fcgi and wsgi is right with keep=1
2008-03-07T01:23:13 <TheSheep_> you can always check for blank values explicitly
2008-03-07T01:23:31 <ThomasWaldmann> keep_blank_values: flag indicating whether blank values in
2008-03-07T01:23:31 <ThomasWaldmann> URL encoded forms should be treated as blank strings.
2008-03-07T01:23:31 <ThomasWaldmann> A true value indicates that blanks should be retained as
2008-03-07T01:23:31 <ThomasWaldmann> blank strings. The default false value indicates that
2008-03-07T01:23:31 <ThomasWaldmann> blank values are to be ignored and treated as if they were
2008-03-07T01:23:37 <ThomasWaldmann> not included.
2008-03-07T01:24:09 <ThomasWaldmann> so it is only for url encoded forms
2008-03-07T01:26:29 <ThomasWaldmann> hmm, no, the src tells it is also for multipart/* stuff
2008-03-07T01:28:45 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3208:4f6d7765e1f5 1.7/MoinMoin/request/request_wsgi.py: wsgi: fix TWikiDraw saving a drawing by also evaluating the query string args
2008-03-07T01:29:31 <ThomasWaldmann> now for the consistency review...
2008-03-07T02:03:27 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3209:799875c9fe2d 1.7/MoinMoin/request/ (5 files): when setting up the request.form dict, make all request methods consistently use keep_blank_values=1 (see cgi.FieldStorage)
2008-03-07T02:39:22 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3210:b6edaabd06b7 1.7/ (3 files in 3 dirs): fix some comments (we still need a Config INSTANCE, but not because of logging)
2008-03-07T06:07:04 <dreimark> moin
2008-03-07T06:07:19 <dreimark> I get an AttributeError: Formatter instance has no attribute 'page'
2008-03-07T06:08:07 <dreimark> if I try to use wikiutil.renderText from an action
2008-03-07T06:08:37 <dreimark> text_moin_wiki.py", line 1400, in format
2008-03-07T06:08:43 <dreimark> self.hilite_re = self.formatter.page.hilite_re
2008-03-07T06:37:42 <dreimark> bbl
2008-03-07T09:13:56 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3211:8a1f1a9dd37f 1.7/ (19 files in 11 dirs): fix. misc problems found by pydev (unused/double imports, unused attributes, ...)
2008-03-07T09:21:07 <ThomasWaldmann> mvirkkil: moin :) btw, the docbook code has some comments about "crappy / even more crappy" code. maybe check if they still apply.
2008-03-07T09:23:31 <ThomasWaldmann> moin grzywacz :)
2008-03-07T09:23:44 <grzywacz> morning ThomasWaldmann!
2008-03-07T09:23:47 <ThomasWaldmann> see my last 1.7 changeset
2008-03-07T09:24:28 <ThomasWaldmann> you maybe want to revert parts of it (and maybe USE the unused stuff I have removed)
2008-03-07T09:24:58 <ThomasWaldmann> i also commented out 2 debug logging lines that used non-existing attributes
2008-03-07T09:26:16 <ThomasWaldmann> mvirkkil: when you are working on the admonitions, see contrib/stylesheets/ in the moin tree
2008-03-07T09:28:28 <grzywacz> ThomasWaldmann, just from looking at the diff it seems a little bit odd to me, but I won't have time until evening to give it a careful review
2008-03-07T09:29:49 <dreimark> hi ThomasWaldmann
2008-03-07T09:29:56 <dreimark> hi grzywacz
2008-03-07T09:33:15 <ThomasWaldmann> grzywacz: yes, I found it odd, too, but the stuff was not used :D
2008-03-07T09:33:38 <grzywacz> hey dreimark
2008-03-07T09:51:38 <dreimark> ThomasWaldmann: seems I did a stupid fix a year ago
2008-03-07T09:52:32 <dreimark> format(self, formatter, inhibit_p=False)
2008-03-07T09:52:43 <dreimark> in text_moin_wiki
2008-03-07T09:54:26 <dreimark> that , inhibit_p=False should be killed
2008-03-07T09:55:01 <dreimark> or someone later will kill me because if that is not removed every other format needs it too
2008-03-07T09:57:43 <dreimark> http://hg.moinmo.in/moin/1.6/diff/1c3d65c8ae8f/MoinMoin/parser/text_moin_wiki.py
2008-03-07T10:01:49 <ThomasWaldmann> johill: line 17 and 19 of serveopenid.py look suspect
2008-03-07T10:13:17 <ThomasWaldmann> dreimark: not sure I am understanding what it was about and what exactly you want to fix now
2008-03-07T10:52:51 <johill> ThomasWaldmann: yeah, randomString predates wikiutil.get_random_string
2008-03-07T10:53:07 <johill> but what's with line 19?
2008-03-07T11:59:52 <ThomasWaldmann> johill: the 2 lines ending with ... import server
2008-03-07T12:01:10 <johill> oh heh
2008-03-07T12:01:23 <johill> yeah the first one is bogus
2008-03-07T12:02:18 <ThomasWaldmann> ok, i remove it
2008-03-07T12:03:27 <johill> ok thanks
2008-03-07T12:12:21 <ThomasWaldmann> johill: sreq_req = sreg.SRegRequest.fromOpenIDRequest(openidreq.message)
2008-03-07T12:12:35 <ThomasWaldmann> (line 305 of serveopenid)
2008-03-07T12:12:53 <ThomasWaldmann> is that doing some side-effect or for what is it?
2008-03-07T12:31:14 <johill> ThomasWaldmann: that code is never triggered
2008-03-07T12:31:19 <johill> data is always False
2008-03-07T12:31:33 <johill> it's some placeholder code for when simple registration is supported
2008-03-07T12:31:49 <johill> and that has a typo to
2008-03-07T12:31:56 <johill> it should read sreg_data not sreq_data
2008-03-07T12:32:06 <johill> oh n/m
2008-03-07T12:32:16 <johill> I think it's just incomplete
2008-03-07T12:32:34 <johill> feel free to remove the data parameter and the whole chunk of code
2008-03-07T12:34:00 <ThomasWaldmann> maybe better do that yourself. i have no idea about that stuff and not test environment.
2008-03-07T12:36:31 <CIA-38> Johannes Berg <johannes AT sipsolutions DOT net> default * 3212:5c5dbfafea66 1.7/MoinMoin/action/serveopenid.py: comment out simple registration code in openid provider
2008-03-07T13:03:18 <johill> darn I always get bitten by this
2008-03-07T13:03:23 <johill> property doesn't work on non-object classes
2008-03-07T13:06:38 <xorAxAx> then make it newstyle :)
2008-03-07T13:43:40 <johill> I did :)
2008-03-07T13:43:58 <johill> but I always think about it for a minute or two and try to find the error elsewhere
2008-03-07T14:10:53 <ThomasWaldmann> apropos newstyle, it complains about conftest.py
2008-03-07T14:14:23 <ThomasWaldmann> E1002:181:MoinClassCollector.setup: Use super on an old style class
2008-03-07T14:15:01 <ThomasWaldmann> (and some others)
2008-03-07T14:15:55 <xorAxAx> thats wrong, the class is a newstyle one
2008-03-07T14:27:28 <ThomasWaldmann> hmm, property does not work in oldstyle? but we have some property stuff in multiconfig.DefaultConfig class
2008-03-07T14:28:18 <johill> ThomasWaldmann: get works but not set
2008-03-07T14:28:30 <johill> ThomasWaldmann: set in oldstyle will simply remove the property instead of invoking the setter
2008-03-07T14:28:50 <ThomasWaldmann> hehe
2008-03-07T14:30:44 <ThomasWaldmann> def make_subscribable_events_prop():
2008-03-07T14:30:59 <ThomasWaldmann> then this maybe has a problem (multiconfig.py)
2008-03-07T14:31:23 <johill> ThomasWaldmann: indeed
2008-03-07T14:31:45 <johill> just make it a new-style class :)
2008-03-07T14:32:30 <ThomasWaldmann> could there be anything going wrong then?
2008-03-07T14:34:34 <johill> http://pastebin.ca/932032
2008-03-07T14:42:55 <ThomasWaldmann> ah, ok X)
2008-03-07T14:45:50 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3213:6162f334bd5a 1.7/MoinMoin/ (7 files in 5 dirs): fix some more minor problems find by pydev (unused/double imports, wrong indentation, ...)
2008-03-07T14:45:51 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3214:ff32c72b9b79 1.7/MoinMoin/action/serveopenid.py: merged main
2008-03-07T14:46:32 <ThomasWaldmann> hm, grammar++
2008-03-07T15:07:42 <dreimark> johill: ThomasWaldmann I do want to add a first version of the gallery arnica parser/action
2008-03-07T15:08:11 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3215:f12af017cfd1 1.7/MoinMoin/config/multiconfig.py: fix DefaultConfig to be a new style class (it is using properties!)
2008-03-07T15:09:25 <ThomasWaldmann> if it is still "unstable", it maybe could just live on ParserMarket
2008-03-07T15:09:42 <johill> maybe we can have a moin branch for it?
2008-03-07T15:09:53 <johill> that can just be based off 1.7 and not change anything but the parser :)
2008-03-07T15:09:57 <johill> or just put it into hg by itself
2008-03-07T15:11:57 <dreimark> If we can get a branch for it it would be nice
2008-03-07T15:12:20 <johill> you can stick it into git on git.sipsolutions.net ;)
2008-03-07T15:13:00 <ThomasWaldmann> i dont think a branch for a single extension makes much sense, but how about a moin/1.7-plugins branch for all 3rd party plugins?
2008-03-07T15:13:54 <johill> ThomasWaldmann: point is that I'd like to hack on it too so having it in hg/git makes sense, it doesn't even need to have the full moin source I guess
2008-03-07T15:14:11 <dreimark> ThomasWaldmann: no, thats sounds a bit it will be a 3rd party plugin forever
2008-03-07T15:14:59 <dreimark> I would prefer hg
2008-03-07T15:15:43 <ThomasWaldmann> it is no problem moving it to main repo when it has stabilized and is of popular use
2008-03-07T15:17:17 <johill> if you don't want to add a repo for it (why not? it should be dirt cheap?) I can host one too
2008-03-07T15:17:45 <ThomasWaldmann> i dont want to have N repos for N plugins
2008-03-07T15:17:59 <johill> why not?
2008-03-07T15:18:16 <ThomasWaldmann> because it makes the repo list too long :)
2008-03-07T15:18:21 <johill> heh
2008-03-07T15:18:31 <johill> see git.kernel.org ;)
2008-03-07T15:19:09 <johill> seriously though, that's fine, I can host it, though I'd prefer git since I'm already hosting a lot of git stuff :)
2008-03-07T15:20:11 <ThomasWaldmann> as long as you dont expect me to use git, I am fine with that :)
2008-03-07T15:21:22 <dreimark> for git i'll need a short list description too.
2008-03-07T15:21:28 <ThomasWaldmann> otherwise I think a repo for all extensions might make sense
2008-03-07T15:21:43 <ThomasWaldmann> (for all devs wanting to use rev ctrl)
2008-03-07T15:22:41 <johill> I guess that could work as long as the number of people using it is not too large
2008-03-07T15:22:43 <dreimark> yes, will 1.7-plugins be a clone of the current branch?
2008-03-07T15:22:49 <johill> (otherwise you always need to merge and stuff)
2008-03-07T15:22:58 <johill> it wouldn't need to be a clone at all
2008-03-07T15:23:07 <johill> it could just come with a plugins/* dir that you can put into your data/
2008-03-07T15:23:36 <johill> (I would actually like that... hg sucks when it comes to having multiple clones... all the history present a dozen times...)
2008-03-07T15:24:09 <dreimark> and htdocs/common with a place holder
2008-03-07T15:24:18 <ThomasWaldmann> i think just plugins/ is a good idea (much better history than with all those merges)
2008-03-07T15:24:21 <dreimark> currently i do have it there
2008-03-07T15:24:31 <johill> good point
2008-03-07T15:24:44 <dreimark> :)
2008-03-07T15:25:19 <ThomasWaldmann> dreimark: placeholder?
2008-03-07T15:26:40 <dreimark> I am not sure in the past there was a problem with empty dirs so we added always a file
2008-03-07T15:27:13 <ThomasWaldmann> we dont need an empty dir. the first dev wanting a htdocs dir can create it and put something into it
2008-03-07T15:27:31 <dreimark> cool
2008-03-07T15:27:54 * ThomasWaldmann creates a repo
2008-03-07T15:27:58 <dreimark> :) :)
2008-03-07T15:30:36 <johill> mind you, I'm going skiing next week, so it's not pressing for me anyway ;)
2008-03-07T15:32:12 <dreimark> I hope you do have some time to peak into the changed version
2008-03-07T15:32:50 <ThomasWaldmann> http://hg.moinmo.in/moin/1.7-extensions/
2008-03-07T15:32:54 <dreimark> I will add some unit tests next week.
2008-03-07T15:34:10 <dreimark> ThomasWaldmann: thanks !
2008-03-07T15:39:06 <ThomasWaldmann> that's maybe also interesting for our theme devs
2008-03-07T15:42:29 <ThomasWaldmann> btw, I'll move some moin related sites from old server to new server soon
2008-03-07T15:51:10 <CIA-38> Reimar Bauer <rb.proj AT googlemail DOT com> default * 0:d164d12d06b8 1.7-extensions/ (16 files in 3 dirs): initial version of arnica (image gallery parser and slideshow action)
2008-03-07T15:52:08 <dreimark> just add #format arnica to a page and upload some images
2008-03-07T15:52:26 * ThomasWaldmann looks
2008-03-07T15:52:53 <dreimark> #format arnica help=1 shows some help
2008-03-07T15:54:53 <johill> dreimark: cloning now
2008-03-07T15:55:21 <ThomasWaldmann> dreimark: + url = url.replace('//', '/') <-- ??
2008-03-07T15:55:59 <dreimark> happens on localhost sometimes // instead of /
2008-03-07T15:57:27 <johill> but that shouldn't matter
2008-03-07T16:02:19 <ThomasWaldmann> dreimark: that should be fixed at some other place then :)
2008-03-07T16:02:29 <ThomasWaldmann> + # only users which are allowed to delete sould use this tools <--- typo
2008-03-07T16:03:50 <johill> there should be more CSS involved
2008-03-07T16:04:41 <dreimark> may be we should add a spellchecker for comments to the tests too ;)
2008-03-07T16:05:09 <ThomasWaldmann> +import os, re, Image <-- see other file
2008-03-07T16:06:16 <dreimark> the wrong url looks like GET //Example?action=AttachFile&do=get
2008-03-07T16:06:41 <dreimark> yeah i though I did it similiar in both files, seems I dreamed that
2008-03-07T16:08:04 <ThomasWaldmann> self.width = str(int(self.thumbnail_width) + int(self.text_width)) huh?
2008-03-07T16:08:50 <ThomasWaldmann> maybe it would be much easier to use an int internally :)
2008-03-07T16:10:04 <dreimark> I think this part gets completly changed we should try to get the argparser involved
2008-03-07T16:10:13 <johill> dreimark++
2008-03-07T16:10:19 <johill> but we need to think about that a lot
2008-03-07T16:10:21 <ThomasWaldmann> and please see the refactored AttachFile.py for some cleaner URL/Link generation
2008-03-07T16:10:37 <johill> since currently all the parser invocations are in the other parsers
2008-03-07T16:13:30 <ThomasWaldmann> + text = text.replace('\\n', '\n') ?
2008-03-07T16:14:58 * dreimark looks
2008-03-07T16:16:02 <ThomasWaldmann> i, ii, z?
2008-03-07T16:16:16 * ThomasWaldmann .oO(long way to go with that src!)
2008-03-07T16:18:49 <dreimark> he, I know they are to short and not self describing
2008-03-07T16:19:05 <dreimark> but I like to add first some tests before changing this part
2008-03-07T16:19:22 <dreimark> and I did already many changes
2008-03-07T16:22:49 <dreimark> ThomasWaldmann: those 3 vars controls images in n columns of a table
2008-03-07T16:24:51 * ThomasWaldmann guessed so
2008-03-07T16:31:30 <CIA-38> Reimar Bauer <rb.proj AT googlemail DOT com> default * 1:33d03349c532 1.7-extensions/data/plugin/ (action/arnica_slides.py parser/text_x_arnica.py): small fixes for comments, separate import for Image
2008-03-07T16:50:34 <dreimark> johill: yes, CSS is completly missing at the moment.
2008-03-07T16:51:09 <dreimark> the unit parser change is applied or ?
2008-03-07T16:58:39 <johill> dreimark: yeah I pushed that out
2008-03-07T17:05:25 <dreimark> fine
2008-03-07T17:06:44 <johill> you realise that 'fine' said like that means 'ok but who cares?' :P
2008-03-07T17:07:01 <dreimark> oh no
2008-03-07T17:07:11 <dreimark> I should besser say ok
2008-03-07T17:07:40 <johill> I know what you meant :)
2008-03-07T17:07:48 <johill> phh
2008-03-07T17:08:01 <johill> "I very much like your latex parser/macro for MoinMoin. However, it doesn't work if you use mod_wsgi (which is kind of very popular because it really doesn't make sense to generate pages everytime - if you're dealing with relatively static contents)."
2008-03-07T17:08:22 <johill> "... (namely, mod_wsgi and i suppose other caching systems as well have problems with `fork`!) ..."
2008-03-07T17:08:47 <johill> that's unpossible.
2008-03-07T17:08:54 <johill> he fixes it by using popen
2008-03-07T17:09:01 <johill> but... how is popen implemented if not using fork()?
2008-03-07T17:09:01 <xorAxAx> instead of?
2008-03-07T17:09:15 <xorAxAx> using the popen function, johill
2008-03-07T17:09:27 <xorAxAx> the difference is file descriptor inheritance probably
2008-03-07T17:09:52 <johill> you realise, of course, that popen() is a libc function that uses fork?
2008-03-07T17:10:04 <xorAxAx> so? :)
2008-03-07T17:10:28 <johill> so what's the problem?
2008-03-07T17:11:06 <xorAxAx> maybe stdout or stderr end up being connected to mod_wsgi?
2008-03-07T17:11:23 <xorAxAx> well, i would just ask for an error description
2008-03-07T17:11:45 <johill> I don't trust popen
2008-03-07T17:12:37 <johill> it uses the shell
2008-03-07T17:12:38 <johill> that's crap
2008-03-07T17:12:52 <johill> you almost never want that
2008-03-07T17:14:04 <xorAxAx> if it uses the shell, it uses a different search path probably
2008-03-07T17:14:39 <johill> nah, the path ought to be configured in the script
2008-03-07T17:15:32 <xorAxAx> well, maybe it isnt --> different behaviour :)
2008-03-07T17:15:51 <johill> we'll see.. I just asked what the error was
2008-03-07T17:27:03 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2433:76d2f1408506 1.6/README: just testing
2008-03-07T17:27:28 <johill> no proper error description...
2008-03-07T17:27:35 <johill> "it fails when using fork but doesn't when using popen"
2008-03-07T17:50:46 * ThomasWaldmann prepares to move hg.moinmo.in
2008-03-07T18:15:00 * ThomasWaldmann switches repo to read-only
2008-03-07T18:18:51 <starshine> it's going to move?
2008-03-07T18:20:03 <ThomasWaldmann> new server, new IP, still hg.moinmo.in
2008-03-07T18:20:08 <starshine> excellent
2008-03-07T18:20:25 <starshine> let me know when you think it's settled in, and I'll try to do a pull.
2008-03-07T18:20:41 <ThomasWaldmann> starshine: what IP does it have for you right now?
2008-03-07T18:21:23 <starshine> 78.47.126.99
2008-03-07T18:21:42 <ThomasWaldmann> ok, so you have a sane DNS :D
2008-03-07T18:21:49 <starshine> about 200ms turnaround time on ping
2008-03-07T18:22:43 <starshine> ..and the packets came back in order :)
2008-03-07T18:25:14 <ThomasWaldmann> total size is 871992230 speedup is 290.05
2008-03-07T18:25:23 * ThomasWaldmann loves rsync :)
2008-03-07T18:25:27 <dreimark> bbl
2008-03-07T18:31:27 <ThomasWaldmann> starshine: btw, there is a new moin/1.7-extensions repo for devs that want to use hg for their extensions (code, themes, ...)
2008-03-07T19:01:06 <niemeyer> Hello masters
2008-03-07T19:01:36 <ThomasWaldmann> hi gustavo :)
2008-03-07T19:03:15 <ThomasWaldmann> (we diskussed shorty about parameter parsing and quoting, maybe interesting for johill also)
2008-03-07T19:06:27 <niemeyer> Yeah, I was mostly wondering if it still enforced quotes
2008-03-07T19:06:50 <niemeyer> and pointing out that it'd be nice if they continued to be optional on non-ambiguous cases
2008-03-07T19:09:13 <grzywacz> re
2008-03-07T19:11:40 <ThomasWaldmann> hi grzywacz
2008-03-07T19:11:48 <grzywacz> hey ThomasWaldmann
2008-03-07T19:12:08 * ThomasWaldmann will do some final tests now and then switch the repos back to rw
2008-03-07T19:15:27 <niemeyer> ThomasWaldmann: Is that 1.7 coming?
2008-03-07T19:17:12 <CIA-38> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 3216:5baf4f41dc4c 1.7/wiki/config/more_samples/ldap_smb_farmconfig.py: added update todo to ldap/smb sample config
2008-03-07T19:18:05 <ThomasWaldmann> niemeyer: http://moinmo.in/MoinMoinTodo/Release%201.7.0 see there for the planned timing
2008-03-07T19:20:55 <ThomasWaldmann> ok, repo is rw again. please only use hg.moinmo.in (hg.thinkmo.de won't work any longer, so fix your repo/.hg/hgrc in case of problems).
2008-03-07T19:21:42 <niemeyer> * removed attachments direct serving (cfg.attachments)
2008-03-07T19:21:45 <niemeyer> What's that about?
2008-03-07T19:27:35 <ThomasWaldmann> it's a old feature, deprecated since long
2008-03-07T19:28:04 <ThomasWaldmann> direct serving of attachments by the web server (not via moin)
2008-03-07T19:28:49 <ThomasWaldmann> (has a big red warning in the docs because of the security problems :)
2008-03-07T19:28:59 <niemeyer> ThomasWaldmann: Hmm.. I'm using direct serving via the web server
2008-03-07T19:29:47 <ThomasWaldmann> really?
2008-03-07T19:29:47 <niemeyer> ThomasWaldmann: Shouldn't be hard to make it continue to work nevertheless.. I already have some patches in place to make the download URL more friendly
2008-03-07T19:29:52 <niemeyer> Yeah
2008-03-07T19:29:52 * ThomasWaldmann wonders :)
2008-03-07T19:30:04 <niemeyer> http://labix.org/editmoin
2008-03-07T19:30:10 <niemeyer> Check the download link
2008-03-07T19:31:34 <ThomasWaldmann> yeah, looks like
2008-03-07T19:31:51 <ThomasWaldmann> btw, i am trying to change the attachment urls
2008-03-07T19:32:04 <ThomasWaldmann> so the filename is in path_info, like a sub-item
2008-03-07T19:32:19 <niemeyer> That'd be nice
2008-03-07T19:34:01 <ThomasWaldmann> who may edit your labix.org wiki, only you yourself?
2008-03-07T19:43:58 <niemeyer> ThomasWaldmann: Me and selected contributors
2008-03-07T19:44:36 <niemeyer> ThomasWaldmann: Moin is being used more as a tool for easy content maintenance than as a traditional wiki there.
2008-03-07T19:44:48 <ThomasWaldmann> ok, so noone attaching evil php files :)
2008-03-07T19:45:56 <niemeyer> ThomasWaldmann: Well.. it's even more than that.. this is really a project oriented page. There are lots of places who offer high quality wiki services.. I'd rather see people using these if they want a more traditional wiki.
2008-03-07T19:47:20 <niemeyer> I have given edit access to pretty much everyone who asked for it, even then
2008-03-07T21:32:46 * ThomasWaldmann prepares moving the wikis MoinMoin: MoinMoinChat/Logs/moin-dev/2008-03-07 (last edited 2008-03-07 00:15:25 by IrcLogImporter)
