johannesV (~johannes@M1567P030.adsl.highway.telekom.at) joined #gnuenterprise.
<johannesV> ohannes@Boston:~/prj/gnue/gnue-appserver$ svn update
<johannesV> johannes@svn.gnuenterprise.org's password: 
<johannesV> svn: Berkeley DB error while opening environment for filesystem /var/svn/gnue/db:
<johannesV> DB_RUNRECOVERY: Fatal error, run database recovery
<johannesV> svn update still running on ash ?
reinhard (~reinhard@M1250P006.adsl.highway.telekom.at) joined #gnuenterprise.
bddebian (~bddebian@ip68-4-154-50.oc.oc.cox.net) left irc: "Client exiting"
dimas (~ds@195.218.177.46) joined #gnuenterprise.
<reinhard> fsck
<reinhard> subversion down again!!
sjc (~sjc@cpc2-seve3-4-0-cust112.popl.cable.ntl.com) joined #gnuenterprise.
sjc (~sjc@cpc2-seve3-4-0-cust112.popl.cable.ntl.com) left #gnuenterprise ("Client exiting").
holycow (~rtaylor@h24-86-227-162.vf.shawcable.net) left irc: "Leaving"
snewpy (~markl@203.173.28.39) joined #gnuenterprise.
mdupont_ (~mdupont@p5080F90D.dip.t-dialin.net) joined #gnuenterprise.
mdupont__ (~mdupont@p5080FAE7.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out)
<johannesV> <johannesV> svn: Berkeley DB error while opening environment for filesystem /var/svn/gnue/db:
<johannesV> <johannesV> DB_RUNRECOVERY: Fatal error, run database recovery
SachaS (sacha@217-173-231-44.dialin.powersurf.li) joined #gnuenterprise.
<SachaS> hi everyone
<reinhard> hi SachaSh
<reinhard> oh
<reinhard> hi SachaS
<SachaS> hallo reinhard
<SachaS> wie gohts (hows going)?
<reinhard> guat :)
<SachaS> bisch fliissig (your working hard)?
snewpy (~markl@203.173.28.39) left irc: "Leaving"
<SachaS> was h?sch vor morn? (what are you up to tomorrow)?
snewpy (~markl@2001:db0:1019:1:99e2:930e:fec5:17bd) joined #gnuenterprise.
<SachaS> see yas later
SachaS (sacha@217-173-231-44.dialin.powersurf.li) left #gnuenterprise ("Leaving").
siesel (~jan@xdsl-213-196-212-101.netcologne.de) joined #gnuenterprise.
snewpy_ (~markl@203.217.45.40) joined #gnuenterprise.
Lainielea (Lainielea@h24-70-165-100.ok.shawcable.net) joined #gnuenterprise.
Lainielea (Lainielea@h24-70-165-100.ok.shawcable.net) left #gnuenterprise.
snewpy (~markl@2001:db0:1019:1:99e2:930e:fec5:17bd) left irc: Read error: 60 (Operation timed out)
<reinhard> siesel: you have a minute?
<siesel> yes
<reinhard> i've thought i'd look at making xmlrpc unicode safe
<siesel> nice
<siesel> IMHO pw_xmlrpc has some kind of unicode flag
<reinhard> i think we had some discussion about
<siesel> just py_xmlrpc doesn't support unicode at all
dimas (~ds@195.218.177.46) got netsplit.
<reinhard> putting type conversion into the xmlrpc layer
<siesel> remove conversion from appserver 
<siesel> yes, we discussed it :)
<reinhard> like that you can transparently transport not only strings and unicode
<reinhard> but also objects like a mx.datetime object
dimas (~ds@195.218.177.46) returned to #gnuenterprise.
<siesel> so we have to transfer type information too
<reinhard> is this still an aim?
<reinhard> yes
<reinhard> we would
<reinhard> i consider transporting every value in form of a (type, stringified_value) tuple
<siesel> that 
<siesel> oops
<siesel> IMHO we don't have to do this for all values
<reinhard> good point
<reinhard> only for such that can't be transported natively
<reinhard> so a string will be transported as a string
<siesel> like most functions have type definitions in the .grpc 
<reinhard> a unicode could be transported either as unicode (if the transport can do that) or as a utf-8 string
<reinhard> siesel: the problem is the variable types
<reinhard> like
<reinhard> fetch () returns a 2-dimensional list where every element can be of different type
<siesel> yes, IMHO variable types should be transported in the way you described (type, stringified_value)
<reinhard> ok i see
<reinhard> i think i understand
<reinhard> if type information is available in the grpc file, there's no need to transport it
<reinhard> if it isn't available, you have to
<reinhard> is that what you mean?
<siesel> a client can fetch it with directory methods
<siesel> yes. it is.
<reinhard> ok
<reinhard> you ok with me trying that over the weekend?
<siesel> I'm ok.
<reinhard> excellent
<reinhard> are you working at something currently?
<reinhard> windows driver stuff?
<siesel> Just let me tell you two or three tips for it.
<siesel> (working on gtk2/wx uidriver)
<reinhard> sure )
<reinhard> sure :)
<reinhard> (good so we probably won't conflict each other)
<dcmwai> siesel, hello siesel :)
<dcmwai> long long long long time no see...
<siesel> 1. type information is passed in every xmlrpc message, so probably new types can be added or just unknown types encoded in a special way
<siesel> 2. the best point to start is the pw_xmlrpc library, as its becomming standart, and it should be able to support unicode automatically (some command line switch)
<siesel> 3. Some times ago I tried to implement type checks for rpc, and did it mostly via type(a)==UnicodeType, isinstance(a,mxDatetime) etc.
<siesel> quite similar to the stuff in native_to_rpc, just to mention it again
<reinhard> is that this stuff?:
<reinhard>       # care for special types
<reinhard>       if hasattr(args[i],'_type'):
<reinhard>         if args[i]._type=='base64':
<reinhard>           args[i]=xmlrpc.base64(args[i]._value)
<reinhard>         elif args[i]._type=='binary':
<reinhard>           args[i]=xmlrpc.base64(args[i]._value)
<reinhard>         elif args[i]._type=='boolean':
<reinhard>           args[i]=xmlrpc.boolean(args[i]._value)
<reinhard>         elif args[i]._type=='datetime':
<reinhard>           args[i]=xmlrpc.datetime(args[i]._value)
<siesel> 4. we have to care for standart compatibilty, i.e. if something is already defined in the xmlrpc definition we should do it in their way
<siesel> ___
<siesel> yes, seems to be
<reinhard> ok
<reinhard> thank you for the hings
<reinhard> thank you for the hints
<reinhard> and i agree with 4.
<reinhard> do you know a good doc about xmlprc
<siesel> more hings as hints, as most is normal gnue policy
<reinhard> (not the python library but the concept)
<reinhard> ?
<siesel> The xmlrpc spec can be found at: http://xmlrpc.org/spec
<reinhard> oh
<reinhard> sounds logical :)
<siesel> Its worth a read. Just 1-2 pages, very easy
<reinhard> thanks
<reinhard> now a last, quite gnu/linux specific question
<reinhard> is there a tool to monitor the xmlrpc traffic?
<reinhard> i.e., when i call some procedure
<reinhard> can i see the xml going over the line somehow?
<siesel> still on the library, the best docu is: /usr/lib/python2.3/xmlrpclib.py
<siesel>  ;)
<siesel> Yes, you can, I'll check how
<reinhard> lol
<siesel> btw. the library documentation for xmlrpclib is very good. Probably this should be your first read: http://python.org/doc/2.3.3/lib/module-xmlrpclib.html
<reinhard> ok
<siesel> there is a xmlrpc debugger at: http://ose.sourceforge.net/browse-example.php?example=xmlrpc-debugger.py
<siesel> reinhard: something else, can you change the group for -rw-rw-r--    1 reinhard reinhard   361748 Mar 11 17:59 log.0000000641 to gnue?
<siesel> its on ash:/var/svn/gnue/db and prevents everybody (except you) to do a svn up
<reinhard> i changed it
<reinhard> but not even i could do a svn update
hadsa (mircbot@80.179.2.79.forward.012.net.il) joined #gnuenterprise.
<reinhard> now it works
<reinhard> weird
hadsa (mircbot@80.179.2.79.forward.012.net.il) left #gnuenterprise.
Nick change: mdupont_ -> mdupont
jamest (~jamest@gw.math.ksu.edu) joined #gnuenterprise.
dsmith (puwtnshc5q@oh-strongsvillecadent1-1f-100.clvhoh.adelphia.net) joined #gnuenterprise.
_florin_ (florin@st8-nelbo.cj.pcnet.ro) joined #gnuenterprise.
lekma (~lekma@pauguste-5-82-67-200-175.fbx.proxad.net) joined #gnuenterprise.
<_florin_> hey!
<_florin_> anybody here for gnue-mysql-bugs?
<_florin_> (bugreports?)
lekma (~lekma@pauguste-5-82-67-200-175.fbx.proxad.net) left irc: Client Quit
<jamest> um
<jamest> bugs?
<_florin_> yep.
<jamest> in our code?
<_florin_> a form that used to work with 0.5.2 doesn't work with 0.5.3.
<jamest> gack
<_florin_> since my form is bugfree, your code is buggy :-)
<jamest> I'm not mysql user but I'll try and help
<_florin_> ok, so hang on. trying to reproduce...
<_florin_> ok.
<_florin_> so i was starting gnue-forms form.gfd
<_florin_> and then trying to do a search (toolbar -> magnifying glas click).
<_florin_> and it says:
<_florin_> snip...
<_florin_> DB000:     self._dataConnection.rollback()
<_florin_> DB000:   File "/usr/lib/python2.3/site-packages/MySQLdb/connections.py", line 143, in rollback
<_florin_> DB000:     NotSupportedError, "Not supported by server")
<_florin_> DB000:   File "/usr/lib/python2.3/site-packages/MySQLdb/connections.py", line 33, in defaulterrorhandler                            
<_florin_> DB000:     raise errorclass, errorvalue
<_florin_> DB000: _mysql_exceptions.NotSupportedError: Not supported by server
<_florin_> and btw
<_florin_> same errors (rollback) for the clear-form button.
<jamest> does mysql support begin; commit; rollback; ?
<_florin_> mysql -u user
<_florin_> use borg_test;
<_florin_> begin; commit; rollback;
<_florin_> Query OK, 0 rows affected (0.00 sec)
[08:48] Last message repeated 2 time(s).
<_florin_> yep. seems like.
<jamest> can you give me a little more of the error msg
<_florin_> DB000: Traceback (most recent call last):
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/uidrivers/wx/ToolBar.py", line 64, in <lambda>
<_florin_> DB000:     lambda event, l=self.driver, e=event, f=self.form: l.dispatchEvent(e,_form=f))
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/common/events/EventController.py", line 107, in dispatchEvent
<_florin_> DB000:     handler(event)
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/GFInstance.py", line 877, in executeRollback
<_florin_> DB000:     event._form.rollback()
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/GFForm.py", line 560, in rollback
<_florin_> DB000:     block.processRollback(recover)
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/GFObjects/GFBlock.py", line 464, in processRollback
<_florin_> DB000:     self._dataSourceLink._dataObject.rollback()
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/common/datasources/drivers/Base/DataObject.py", line 155, in rollback
<_florin_> DB000:     self._dataConnection.rollback()
<_florin_> DB000:   File "/usr/lib/python2.3/site-packages/MySQLdb/connections.py", line 143, in rollback
<_florin_> DB000:     NotSupportedError, "Not supported by server")
<_florin_> DB000:   File "/usr/lib/python2.3/site-packages/MySQLdb/connections.py", line 33, in defaulterrorhandler
<_florin_> DB000:     raise errorclass, errorvalue
<_florin_> DB000: _mysql_exceptions.NotSupportedError: Not supported by server
<_florin_> that's the whole traceback.
<_florin_> (btw, i'm using python 2.3, mysql 4.0.16, on a debian sarge, linux-ppc)
<siesel> mysql dbsig driver doesn't support rollback and commit IMHO
<_florin_> best sollution: switch to postgresql?
<jamest> :)
<_florin_> ok. since you guys work on postgres it might be best for me to stick with you for now...
<_florin_> BUT: ready for another bug? :-)
<jamest> well
<jamest> we still need to fix mysql
<jamest> you can probably comment out a line here and get it working
<jamest> if its an emergency
<_florin_> well... i'm in an early stage of developing a program here.
<_florin_> so i could stick with mysql or switch to postgres...
<jamest> # NOTES:
<jamest> # Supports transactions if the MySQL server is compiled w/transaction support
<jamest> # (which it does NOT by default)
<_florin_> which file is that?
<siesel> jamest: I found the error, Base/DataObject.py calls direct the native procedure instead of calling the drivers connection objects 'commit'/'rollback'
<jamest> svn/gnue/gnue-common/src/datasources/drivers/mysql/mysql/DataObject.py
<jamest> siesel: ah, cool
<jamest> so replace _dataConnection w/ connection I imagine?
<jamest> but that looks like it's depreciated anyway
<siesel> yes, but its still in use somewhere :(
lekma (~lekma@pauguste-5-82-67-200-175.fbx.proxad.net) joined #gnuenterprise.
<siesel> _florin_: can you do an "svn up" or should I send you a diff which fixes the error?
<_florin_> in the installed version, is it somewhere in /usr/lib/gnue/python/gnue/ ?
<jamest> in forms
<_florin_> i don't have a svn version.
<_florin_> but i'll get one.
<_florin_> (do i need an account or something?
<_florin_> )
<siesel> Its in /usr/lib/gnue/python/gnue/common/datasource/drivers/Base/DataObject.py
<siesel> no you don't need an account
<jamest> svn co svn://www.gnuenterprise.org/var/svn/trunk
<jamest> IIRC
<jamest> er
<jamest> lol, it's in the topic 
<jamest> svn co svn://svn.gnuenterprise.org/var/svn/gnue/trunk gnue
<_florin_> ok, on my way...
<_florin_> do i need any extra-hackery to run the svn-version direcly (like installing or something)?
<reinhard> in the gnue-common directory
<reinhard> run setup-cvs.py
<_florin_> ok. so while i'm downloading, can you look over another debugging-output? related to dropdown-lists...
<_florin_> i changed the style of an input field to "dropdown" and now it says:
<_florin_> snip...
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/GFForm.py", line 674, in __refreshDisplay
<_florin_> DB000:     self.dispatchEvent('updateENTRY',object, _form=self)
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/common/events/EventController.py", line 107, in dispatchEvent
<_florin_> DB000:     handler(event)
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/uidrivers/_base/UIdriver.py", line 329, in updateEntry
<_florin_> DB000:     value = handler.getDisplayFiller(entry.getValue())
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/GFDisplayHandler.py", line 88, in getDisplayFiller
<_florin_> DB000:     return self._buildDisplayHelper(value, False)
<_florin_> DB000:   File "/usr/local/gnue/lib/python/gnue/forms/GFDisplayHandler.py", line 763, in _buildDisplayHelper
<_florin_> DB000:     if self.field._allowedValues.has_key(str(value)):
<_florin_> DB000: AttributeError: 'NoneType' object has no attribute 'has_key'
<_florin_> is it my mistake?
<reinhard> _florin_: dropdowns are broken :((
<reinhard> we know that
<_florin_> ok :-)
<_florin_> svn output:
<_florin_> svn: Expected version '2' of repository; found version '3'
<reinhard> which version of subversion do you use?
<reinhard> we updated to 1.0.0 yesterday
<reinhard> so you might have to update, too
<_florin_> svn --version
<reinhard> (sarge has 1.0.0
<reinhard> )
<_florin_> svn, version 1.0.0
<_florin_>    compiled Feb 23 2004, 20:25:06
<_florin_> Copyright (C) 2000-2004 CollabNet.
<_florin_> Subversion is open source software, see http://subversion.tigris.org/
<reinhard> oh
<reinhard> hmm
Action: reinhard goes back to his corner
<_florin_> so 1.0.0 is good?
<reinhard> ye
<reinhard> yes
wendall911 (~wendallc@wendallc.icehouse.net) joined #gnuenterprise.
<_florin_> so any chances for me to checkout?
<reinhard> i get the same error
<reinhard> jamest: you know what this error means?
<_florin_> no.
<_florin_> (oh, sorry -- thought you were asking me)
<reinhard> i'm asking jamest, our svn administrator :)
<jamest> users
<jamest> um
<jamest> lemme look
<jamest> grrr
<jamest> siesel: what version of svn you using?
bddebian (~bddebian@ip68-4-154-50.oc.oc.cox.net) joined #gnuenterprise.
<jamest> oooo
<jamest> getting close
<jamest> svnserve wasn't restarted after the upgrade
<jamest> so it was still at 0.33
<jamest> which is why devel accounts worked as they use ssh not svnserver
<_florin_> svn: Berkeley DB error while opening environment for filesystem /var/svn/gnue/db:
<_florin_> Permission denied
<_florin_> moving forward :-)
<_florin_> (does this /var/svn/gnue/db on my filesystem, or on the server?)
<jamest> server
<jamest> the .deb upgrade whiped out our wrappers
<jamest> wiped even
<jamest> good thing we don't back them up
<jamest> :)
<_florin_> backup? how do you spell? :-)
<jamest> actually we do backup the server
<_florin_> what does it take to get a ssh-svn account? this way i could download without snvserve
<_florin_> (i promise i'm not a psychopat with a strong desire to delete everything :-)
<jamest> it'll be just a few minutes
<jamest> n thing
<jamest> s/n/damn
<jamest> now it's launching a few 100 copies of svnserver
<jamest> due to my error of course :)
<jamest> should work now
<_florin_> yep. downloading...
<reinhard> what??
<reinhard> i get a commit mail after less than 2 hours??
<reinhard> there must be something wrong here
<jamest> don't commit!
Action: jamest grabs a trout
<_florin_> is this svn thing very different from cvs?
<_florin_> (i know it's the successor...)
<jamest> nope
<jamest> and yip
<jamest> commands are almost exactly the same
<jamest> for most commonly used things
<jamest> cvs up -Pd == svn up
<jamest> svn commit
<jamest> svn add
<_florin_> should be enough for me... i never used anything else besides commit/co/add/remove/update
<jamest> those should be the same
<jamest> svn move
<jamest> is new and nice
<jamest> you can move files, dirs about w/o issues
<_florin_> kewl :-)
<jamest> svn uses about 2x the disk space for a checkout
<jamest> as it keeps an unmodified copy on your system
<_florin_> why?
<jamest> so diffs, reverts, etc don't require net access
<jamest> and are fast
<_florin_> oh. that's kewl... does it do patche automagically, too?
<_florin_> i mean patches
<jamest> dunno
<jamest> not sure what you mean
<_florin_> like... you change something and want to build a patch to reflect your changes.
<_florin_> since it keeps a clean copy, it could diff your changes agains the copy for you...
<reinhard> svn diff
<reinhard> does that
Action: dsmith is liking svn more and more
<_florin_> how big is a gnue checkout? (trying to estimate by download time here...)
<dsmith> By keeping a local copy of the unmodified file, svn only sends diffs to the repo, optimizing network usage.
<reinhard> _florin_: 103 M for me
<dsmith> With cvs, you send the whole file, and the diff is done at the server (iirc)
<_florin_> i heard something about atomic commits...
<_florin_> like... grouping some files together to form an entity which is to be treated as atomic.
<dsmith> Yeah, the repo is versioned, not individual files.
<_florin_> how do you guys use that? on a per directory basis?
<_florin_> (like forms, designer, common...)
<reinhard> i try to use it on a logical basis
jcater (~jason@w202.z065105010.mem-tn.dsl.cnc.net) joined #gnuenterprise.
<reinhard> like i try to commit things in a single commit
<reinhard> that are related changes
<reinhard> for example
<reinhard> when i would change a function in common
<jamest> i try to use it on an infrequent basis
<reinhard> and also changed some calls of that function in gnue-forms and gnue-appserver
<jamest> as it means i had to do work
<reinhard> i would commit all of them together
<reinhard> jamest: lol
<_florin_> so atomic commits are not statical (as in "dicated by the server/repository")?
<jcater> morning all
<_florin_> so thanks to everybody -- download complete :-)
<jbailey> Heya Mr. ca[r]ter.
<_florin_> gotta go now, running out of money (i'm in an internet cafe somewhere in transilvania :-)
<_florin_> bye and thanks again!
<dsmith> !!
_florin_ (florin@st8-nelbo.cj.pcnet.ro) left #gnuenterprise.
thierry__ (~thierry@AStrasbourg-251-1-1-7.w82-127.abo.wanadoo.fr) joined #gnuenterprise.
thierry_ (~thierry@AStrasbourg-251-1-1-7.w82-127.abo.wanadoo.fr) left irc: Read error: 110 (Connection timed out)
Nick change: thierry__ -> thierry
<jamest> gack!
<jamest> an gnue svn checkout from an internet cafe!?!?!?
<jamest> that dude is even sicker than at least a few of the people in here
<dsmith> The mind boggles
lekma (~lekma@pauguste-5-82-67-200-175.fbx.proxad.net) left irc: "Leaving"
dimas (~ds@195.218.177.46) left irc: "÷ÙÛÅÌ ÉÚ XChat"
<wendall911> I work from the Internet Cafe! What's weird about that?
<wendall911> hehe
<jamest> well, that proves my point
<jamest> "than at least a few"
<jamest> :)
Action: jamest doesn't see how one could afford to do that
<wendall911> jamest: that is one of the funniest things I've heard in a while I must admit, lol
<wendall911> jamest: I can just see reinhard --> "any of you guys know how to check in from the Internet Cafe?"
<jamest> :)
<jbailey> jamest: Many Internet Cafés offer monthly rates.
<jbailey> jamest: The one in new york that I used was $50 for unlimited for a month.
<jamest> wow, i didn't know that
<jamest> i knew some profs here that use them overseas and last I knew it was outrageous cost
<jamest> ah, that's not too bad really
<jbailey> Overseas it tends to be.
<jbailey> This place in Manhatten had like 800 computers.
<wendall911> jbailey: that's pretty cool little fact
<siesel> jamest: if you can go zo intercafe's to play multiplayer games several hours long,
<siesel> why not better play with the goats ?
<wendall911> that internet cafe would make a killer compile farm
<wendall911> might as well put those unused cpu cycles to work ;)
<jbailey> *lol*
<jbailey> It's a very cool café, too.  They reformat the machines after every customer.
<siesel> bbl
siesel (~jan@xdsl-213-196-212-101.netcologne.de) left irc: "Client exiting"
lekma (~lekma@pauguste-5-82-67-200-175.fbx.proxad.net) joined #gnuenterprise.
<jamest> here's youre knoppix CD and your meal
<jamest> your
<jbailey> jamest: I think i'm going to put together a Debian LiveCD just so that people stop using Knoppix. =)
<jamest> but 
<jamest> but
[11:28] Last message repeated 1 time(s).
<jamest> knoppix is debian
<jbailey> Sure, but shouldn't it be debian/knoppix gnu/linux?
Action: jbailey hides
<jbailey> =)
<jamest> lol
<jamest> i figured it was some crazy thing against their use KDE
<jamest> they probably also used screwedUpCaps in their var names too!
<bddebian> Knoppix != Debian
<jbailey> jamest: It's a love the sinner, hate the sin sort of thing.
<jamest> bddebian: you're right, knoppix supports more hardware
Action: jamest runs
<bddebian> Doh.. :-)
<reinhard> jamest: you ever heard of Gnoppix?
<jamest> yeah
<wendall911> reinhard: it's way cool, even better, I like Morphix
dustym (~dustym@66.182.192.34) left irc: Read error: 110 (Connection timed out)
dustym (~dustym@66.182.192.34) joined #gnuenterprise.
dustym (~dustym@66.182.192.34) left irc: "Leaving"
Nick change: dsmith -> dsmith-lonch
johannesV (~johannes@M1567P030.adsl.highway.telekom.at) left irc: "Client Exiting"
lekma (~lekma@pauguste-5-82-67-200-175.fbx.proxad.net) left irc: "Leaving"
Nick change: dsmith-lonch -> dsmith
Nick change: bddebian -> bddeb_away
<reinhard> does anybody use common.rpc outside GNUe?
<jcater> not that I know of
<reinhard> thanks
jamest (~jamest@gw.math.ksu.edu) left irc: Read error: 104 (Connection reset by peer)
jamest (~jamest@gw.math.ksu.edu) joined #gnuenterprise.
<jcater> reinhard: would it be possible to only do the setup.py ChangeLog and other stuff that calls svn, during an sdist ?
<jcater> i.e., I logged in as root to do a ./setup.py install
<jcater> and couldn't
<jcater> kind of surprised me
<reinhard> the two things normally done when run from svn are
<reinhard> get ChangeLog and build translations
<reinhard> what about building the translation?
<jamest> why are we doing them each time?
<jcater> reinhard: my thinking is people can't just grab a snapshot of our svn
<jcater> and install on their machine
<jcater> they'd have to get a prebuilt tarball somewhere
<jcater> unless they have a full install
<jcater> hmm
<jamest> do the tarballs ship minus translations
<reinhard> the tarballs contain all that stuff
<reinhard> the problem is when you get svn snapshot "tarballs"
<reinhard> which weren't made with setup.py sdist
<reinhard> but with tar :)
<reinhard> if the tarballs were done with setup.py sdist
<reinhard> they would be installable with no dependency
<reinhard> i mean with no extra dependency :)
<jamest> http://story.news.yahoo.com/news?tmpl=story&u=/040312/482/nyet14403121827
<reinhard> would it be possible to build the snapshots with ./setup.py sdist?
<jcater> um
<jamest> that has got to be the most horrible picture of britney spears in existance
<jamest> someone posted it in another channel
<jcater> possibly
<reinhard> that would solve all that problems
<jcater> it introduces new problems
<jcater> but would solve that problem
<reinhard> and would make the snapshot fully functional tarballs
<reinhard> which problems would it introduce?
<jcater> well, they were originally introduced (the snapshots) for people who wanted to participate in developement but had trouble w/cvs(now svn) could just grab a tarball
<jcater> but now the repository has to be fully functional for a snapshot to be created
<jcater> that may be an ok change
<reinhard> i think the point is
<reinhard> there will always be files that are generated but have a ugly dependency
<reinhard> that starts with the ChangeLog
jamest (~jamest@gw.math.ksu.edu) left irc: Read error: 104 (Connection reset by peer)
<reinhard> and is also the case for other things
<reinhard> like generated translations
reinhard (~reinhard@M1250P006.adsl.highway.telekom.at) left irc: Read error: 54 (Connection reset by peer)
<jcater> everyone's bailing on me
reinhard (~reinhard@M1250P006.adsl.highway.telekom.at) joined #gnuenterprise.
<reinhard> grrr
<reinhard> what was the last line you got?
<jcater> we can try doing sdist snapshots and see how it goes
<jcater> <reinhard> and is also the case for other things
<jcater> <reinhard> like generated translations
<reinhard> maybe documentation
<reinhard> (tex or docbook)
<reinhard> these files will always be in tarballs
<reinhard> but they will never be in svn
<reinhard> so the question is
<reinhard> do we want the snapshots to be "daily tarballs"
<reinhard> or are they just an svn co without svn?
<jcater> we can try sdist tarballs
<jcater> and see if anyone complains
<jcater> may be best in the long run
<jcater> I just miss being able to ./setup.py install :)
<jcater> not that I do it often
<reinhard> jcater: from a snapshot?
jamest (~jamest@gw.math.ksu.edu) joined #gnuenterprise.
<reinhard> you should be able to do that from a release as welll as be able to do it from svn co
<reinhard> only snapshot should make problems
<reinhard> i guess i could do another thing
<reinhard> i have this set of files not in svn but in release
<reinhard> i generate all these files on setup.py build/install as well as on setup.py sdist
<reinhard> even though some of them are not necessary to install
<reinhard> (ChangeLog being prominent example)
<reinhard> i could on setup.py build/install only build the files that are needed for install
<reinhard> for example
<reinhard> po/*.gmo would get built
<reinhard> always
<reinhard> but ChangeLog would only be built on setup.py sdist, not on build/install
<reinhard> does that make sense?
<jcater> I think that would be better
<jcater> I can understand the translations needing to be built
<jcater> but ChangeLog is really only needed on sdist anyway
<jcater> (and docs too, imho)
<reinhard> jcater: that's the question
<reinhard> jcater: currently, setup.py install installs some docs into $prefix/share/doc
<reinhard> i think this is quite unusual
<jcater> that setup.py install installs docs?
<jcater> or that it generates docs?
<reinhard> well
<reinhard> it installs docs
<reinhard> and for appserver
<reinhard> it also generates them (runs them through texinfo)
<reinhard> so that we get plain text versions
<reinhard> generation of ChangeLog should be fixed now
<reinhard> this is an incompatible change for all setup.py programs
<reinhard> but that's no problem at all because the i18n stuff is also an incompatible change :)
<reinhard> so we will have to release all tools together next time anyway
<reinhard> svn: Commit failed (details follow):
<reinhard> svn: Berkeley DB error while opening environment for filesystem /var/svn/gnue/db:
<reinhard> Permission denied
Action: jcater mutters something bad under his breath
Action: jcater may have to submit his first patch to subversion soon
<reinhard> http://www.jacksonh.net/jackson/images/godkills_bugs.jpg
<jcater> lol
<jcater> they know of the problem
<jcater> the solution is to mv svn svn.wrapper
<jcater> and create an svn script file that sets environ variables and calls svn.wrapper
<jcater> asinine
<reinhard> jcater: if you could fix svn then i would commit the fix for setup.py Changelog generation
<jamest> wtf
<jamest> i wrapped that this am
<jamest> reinhard: try now
<jamest> wtf
<jamest> those were wrapped at 9am
<jamest> and jason commited something at 3pm
<jamest> and it screwed up the perms
<jamest> ah , wait!
Action: jbailey waits.
<jamest> sigh
<jamest> it lost the g+s
<jbailey> Is your umask set wrong then?
<jamest> no, that's the wrapper
Action: jamest kicks svn
Action: jbailey still only uses CVS.
<jamest> ash:/var/svn/gnue/db# more /usr/bin/svn
<jamest> umask 002
<jamest> /usr/bin/svn.real "$@"
btami (~btami@napalm.napnet.hu) joined #gnuenterprise.
Vee2d2 (~vin@66.169.136.41) left irc: Read error: 110 (Connection timed out)
<jamest> i like svn
<jamest> but their multi user setup is a bit more touchy than i recall cvs being
<reinhard> ok committed
reed (reed@pcp04683586pcs.tupelo01.ms.comcast.net) joined #gnuenterprise.
reed (reed@pcp04683586pcs.tupelo01.ms.comcast.net) left #gnuenterprise.
<Vee> we're fixing to migrate to svn maybe.. 
<Vee> we're kinda teetering now though, because svn doesnt appear to letcha see who's doing what (as in cvs if everyone's using edit, you can do a cvs editors to see what people are working on, and get notification if someone starts editing something you're working on, etc))
snewpy_ (~markl@203.217.45.40) left irc: "Leaving"
Nick change: mdupont -> M7U90nt
snewpy (~markl@2001:db0:1019:1:99e2:930e:fec5:17bd) joined #gnuenterprise.
<jbailey> Vee: Can't you just do that with an on-commit hack of some sort?
<Vee> *shrug* the notification I'm talking about isnt related to commits though
<Vee> s/isnt/arent/
<Vee> modules are checked out readonly.. if I wanna work on file x, I cvs edit x. That changes the perms to writable on my local copy, and if anyone does a cvs editors, they'll see I'm editing file x and see a timestamp for when I issued the edit, too.. 
<jbailey> Ah, weird.  i've never done that.
<Vee> if you came along and did a cvs edit x on your copy, I'd get an email from the server telling me so.. 
<Vee> it's kinda handy
<Vee> when I asked in #svn earlier sussman hinted that they may be looking into it soon
<Vee> or something lik eit
Vee2d2 (~vin@66.169.136.41.ts46v-07.otnc1.ftwrth.tx.charter.com) joined #gnuenterprise.
jamest (~jamest@gw.math.ksu.edu) left #gnuenterprise ("Kopete 0.8.0 RC1 : http://kopete.kde.org").
dsmith (puwtnshc5q@oh-strongsvillecadent1-1f-100.clvhoh.adelphia.net) left irc: "Good Night"
M7U90nt (~mdupont@p5080F90D.dip.t-dialin.net) left irc: "Leaving"
M7U90nt (~mdupont@p5080F90D.dip.t-dialin.net) joined #gnuenterprise.
jcater (~jason@w202.z065105010.mem-tn.dsl.cnc.net) left irc: "home"
Nick change: M7U90nt -> md-party
Nick change: md-party -> _mdupont
Nick change: _mdupont -> _m7u90nt
holycow (~rtaylor@h24-86-227-162.vf.shawcable.net) joined #gnuenterprise.
btami (~btami@napalm.napnet.hu) left irc: Read error: 60 (Operation timed out)
jcater (~jcater@cpe-066-061-071-147.midsouth.rr.com) joined #gnuenterprise.
snewpy (~markl@2001:db0:1019:1:99e2:930e:fec5:17bd) left irc: Read error: 54 (Connection reset by peer)
snewpy (~markl@2001:db0:1019:1:99e2:930e:fec5:17bd) joined #gnuenterprise.
jamest (foobar@adsl-66-142-221-144.dsl.tpkaks.swbell.net) joined #gnuenterprise.
ogger (~ogger@57.Red-213-96-148.pooles.rima-tde.net) left irc: "Leaving"
snewpy (~markl@2001:db0:1019:1:99e2:930e:fec5:17bd) left irc: Read error: 54 (Connection reset by peer)
snewpy (~markl@2001:db0:1019:1:99e2:930e:fec5:17bd) joined #gnuenterprise.
dneighbo (~dneighbo@ip68-109-180-32.ph.ph.cox.net) joined #gnuenterprise.
reinhard (~reinhard@M1250P006.adsl.highway.telekom.at) left irc: "When all else fails, read the instructions."
derek (~dneighbo@ip68-109-180-32.ph.ph.cox.net) left irc: Read error: 110 (Connection timed out)
jbailey (~jbailey@dragonfly.fundserv.com) left irc: "Client exiting"
Nick change: bddeb_away -> bddebian
<chillywilly> hi
<jcater> word
<jcater> anyone using apache2?
Action: jcater is trying to figure out what to install to get mod-php under apache2
<wendall911> jcater: you just need mod_php
<wendall911> jcater: need -DPHP4 as a startup parameter
<ajmitch> jcater: there's no debian package, the php4 & apache2 people don't get along, so php4 sucks with apache2 due to threadsafety issues :)
<jcater> omfg
<jcater> thanks
<ajmitch> you can use it with the apache2-mpm-prefork, which is the same model as apache 1.x
<ajmitch> others are not guaranteed to work nicely, iirc
<wendall911> mpm address that issue
<jcater> ok, I chose preform
<jcater> prefork
<jcater> so that's a good start :)
<wendall911> ajmitch: it isn't so much that they don't get along, it's that the php4 devs think it is an apache issue...I tend to agree, as it is pushing the security model off onto modules, instead of apache handling it.
<ajmitch> wendall911: they blame each other :)
<ajmitch> I heard some comments from each camp at LCA
<wendall911> ajmitch: I think if it runs as a mod, it's an apache issue, if it runs as cgi, it's a php issue
<wendall911> ajmitch: apache guys are just pissed because mpm sucks
Action: ajmitch shoots neighbours
<wendall911> nothing really runs threadsafe under apache2, they need to quit blaming people and just fix it
<ajmitch> as an end user, I don't care too much as to who's to blame, so long as someone fixes it
<wendall911> agreed
wendall911 (~wendallc@wendallc.icehouse.net) left irc: "Download Gaim: http://gaim.sourceforge.net/"
holycow (~rtaylor@h24-86-227-162.vf.shawcable.net) left irc: Read error: 60 (Operation timed out)
holycow (~rtaylor@h24-86-227-162.vf.shawcable.net) joined #gnuenterprise.
holycow (~rtaylor@h24-86-227-162.vf.shawcable.net) left irc: "Leaving"
holycow (~rtaylor@h24-86-227-162.vf.shawcable.net) joined #gnuenterprise.
Nick change: bddebian -> bddeb_afk
Maniac (~Maniac@h24-77-228-236.wp.shawcable.net) joined #gnuenterprise.
Maniac (~Maniac@h24-77-228-236.wp.shawcable.net) left #gnuenterprise ("Leaving").
Nick change: bddeb_afk -> bddebian
jamest (foobar@adsl-66-142-221-144.dsl.tpkaks.swbell.net) left irc: "Kopete 0.8.0 RC1 : http://kopete.kde.org"
jcater (~jcater@cpe-066-061-071-147.midsouth.rr.com) left irc: "sleep"
--- Sat Mar 13 2004