*** yure has joined #gnuenterprise *** neilt has joined #gnuenterprise morning all bbl *** btami has quit IRC good morning neilt you're up early :-) actually about an hour later than normal what broke with the pictures in devguide? nothing broke it just for the different target formats, you need the pics in different formats for pdf, you need them as pdf for dvi, you need them as ps so I just missed the .ps version? yes and for the html output, the png's were ending up in the wrong dir sorry for that. no problem at all I liked your changes, btw when does the web site rebuild the docs? thanks the web site always shows the docs for the last release ok What I see in reading the docs is that they are pretty good to start with I think much better than when I was involved last time. good to hear that i am aware that the devguide is still missing a lot but it's like with many projects you need 90% of the time for the first half of the project and for the second half of the project, you need another 90% of the time ;-) maybe I can help with that? sure, actually you were already doing so yesterday :-) the docs will not build for me dvips devguide -o devguide.ps This is dvips(k) 5.95a Copyright 2005 Radical Eye Software (www.radicaleye.com) dvips: ! unexpected eof on DVI file make: *** [devguide.ps] Error 1 oh try "make pdf" to only generate the pdf version pdf version works find fine o ok make: *** No rule to make target `pdf'. Stop. err make devguide.pdf would that be IIRC but when I do 'make', the pdf builds first and works ok ah ok good make: `devguide.pdf' is up to date. make: `devguide.pdf' is up to date. make: `devguide.pdf' is up to date. for html you can do "make devguide.html" etc but I don't have a .dvi file (should I have)? no dvi is just another document format make: `devguide.html' is up to date. like pdf, postscript, txt ah you *don't* have a dvi? so no wonder the conversion from dvi to postscript fails :) if you do "make devguide.dvi", is there any appearant error message? Output written on devguide.pdf (23 pages, 165289 bytes). Transcript written on devguide.log. oh there is a bunch of other stuff in the middle so your tex version creates pdf and not dvi interesting seems to be the first line after make devguide.dvi LANG=C texi2dvi devguide.texi yep that should create a devguide.dvi not a devguide.pdf that seems to make sense as there is a separate command texi2pdf for pdf generation but it's not a big issue anyway as you can probably live well with only pdf and html working ok, but it would be nice to be able to check all of the docs when committing changes the dvi and ps output will be identical to the pdf, just a different file format so no real need to check them separately *** jamest has joined #gnuenterprise *** johannesV has quit IRC *** johannesV_ has joined #gnuenterprise neilt: the wiki is back up * jamest thinks you were asking about it jamest: excellent, thanks reinhard: can we make the following change in the make file? make devguide.dvi echo -n "@set UPDATED " > version.texi grep '$Date:' 01-intro.texi 02-concepts.texi 03-basicdb.texi 04-ui.texi 05-references.texi 06-procedures.texi | cut --delimiter=' ' --fields=3 | sort | tail -1 >> version.texi tail: `-1' option is obsolete; use `-n 1' Try `tail --help' for more information. make: *** [version.texi] Error 1 *** kilo has quit IRC neilt: tail -n 1 neilt: tail -1 was deprecated mnemoc: exactly, -n 1 works for me, but not sure about others as I have no way to test? -n works everywhere -1 was a non-posix exception deprecated by recient glibc excellent, I will make the change, thank you. np :) neilt: yes, please make that change sorry was on the phone -n 1 should work for everybody thanks, done *** btami has joined #gnuenterprise *** jcater has quit IRC hmmm, tail -# is depreciated, that's good to know neilt: fwiw, I just uploaded your changes in documentation it's just the "postweb" script in appserver directory I will gladly do so whenever you make improvements in the devguide while your problems with dvi persist reinhard: thanks *** jcater has joined #gnuenterprise jamest: thanks for fixing wiki ! *** tophe has joined #gnuenterprise np, sorry it took so long may i ask another one? you know 1 day, 1 ask sure :) :) per person? ;-) gack! * jamest hides can you setup dcl or any other you like for bugtracking, please ? *** tophe has quit IRC *** tophe has joined #gnuenterprise #rubyonrails *** tophe has left #gnuenterprise jamest ? (knock, knock, knock) * jamest is hiding i can and I can setup whatever we wish i know :) reinhard ?, jcater ? is DCL ok for you,or want something instead ? I would enter my bugs in whatever works it doesn't have to be DCL i know if I were to decide, I would use roundup as it's python and seems to work pretty well what is the current status of DCL? dunno haven't seen mdean in a while the original plan was IIRC that dcl would be ported to python and gnue forms still listed as beta on the web site, last update security fix March 2005 but AFAICT that was never even started, and it's been some time so I don't really know how "part of gnue" he actually feels Roundup has been deployed for: bug tracking and TODO list management (the classic installation) customer help desk support (with a wizard for the phone answerers, linking to networking, system and development issue trackers) issue management for IETF working groups sales lead tracking from the web page of roundup http://www.edgewall.com/trac/details.html also looks interesting, but not sure if the license is compatable seems roundup is a bit mature it is not that hard to make gnue forms for whatever bugtrack system but how could we use them? connection to remote db? dimas: that'd pull time from other stuff, time that is in very short supply :( trac is a wiki with svn viewer, with an issue tracker integrated trac is popular for python and non-python projects *** btami has quit IRC jamest: would make an interesting proof of concept for GNUE *** btami has joined #gnuenterprise trac license is BSD i vote for roundup, if it counts... is there a public instance of roundup on the web? if you don't need wiki, skip trac imo neilt: yes, it would be, but again who'd have time to do it. I know for a fact I wouldn't have time for ages jamest: i could, but would it work better then web-application? i would think web would be better seems so *** kmk has joined #gnuenterprise i really don't have a preference i'll setup what the masses want (within reason) so far we have: reinhard, btami for roundup * btami feels masses with reinhard :) lol * jamest decides it's best not to watch this btami reinhard mass feeling thing roundup looks nice an simple. That appeals to me I'm 100% indifferent so does that mean the masses have spoken? hey! no weight jokes :D lol hi ppl I think he was referring to your HUGE gonads ;) no, no, it was pretty much just a weight slam jcater has to have huge gonads to be putting up with the people here ;) * jcater worries that his gonads are being discussed in irc i don't think that's weight related though more just swelling due to his users always busting them lol *** SachaS has joined #gnuenterprise *** klasstek has joined #gnuenterprise from the roundup docs 3. Long running roundup-server instances ======================================== Note that long running Python processes may deplete your system memory since the current [1]C Python implementation never free()s allocated memory. [1] http://evanjones.ca/python-memory.html we'd run cgi roundup i'd imagine so this shouldn't impact us but I was wondering if this would impact appserver display mask for float value, would it work now? jamest: that's something johannes and lekma were fighting with quite some time dimas: i don't think so jamest: that's rather lame ok jamest: the memory management thing I know I will get killed for saying this but python is no longer my fav lang, I like ruby more :) * jamest adds chillywilly to the list of people that must be "re-educated" soon it's not a big problem anyway this memory thing as it only means python doesn't return memory to the os, but it keeps it for further reuse um, but you could still exhasut the system memory exhasut too cu all *** reinhard has quit IRC chillywilly: why ruby? from the ruby faq Ruby?s syntax and design philosophy are heavily influenced by Perl. yep, python it is then :) doesn't really seem like perl to me it's more OO than Python adn evern has encapsulation mechanisms even* for quick hacks, perl is actually pretty effective, But I would not use it for a project. I like the way it handles the multiple inheritance thing Ruby, like Smalltalk, only supports single inheritance it does have some perl-like shortcuts that can be used but it does have a very powerful mix-in concept: a class definition may include a module, which inserts that module?s methods, constants, etc. into the class. i used to use perl for small stuff until i realized it didn't gain me anything over python and was much harder to read *** btami has quit IRC *** btami has joined #gnuenterprise jamest: perl does not have to be hard to read, that is up to the writer. well, my coder sucks python (based on my current knowledge) sucks for regular expressions so I, er he, writes hard to read perl that is evaluating text for match and change based on REs. compared to perl I believe you are right i hear ruby :) but I don't use REs inside apps all that much I would not use REs in apps either but quick data conversions etc perl is quick and easy this weekend is the ruby conference in San Diego unfortunately sold out. I may go anyway, maybe I can sneak in neilt: i tend to use bash, sed, awk and/or vi for most that now of course I use python as my desktop calculator in most cases now too :) * neilt now understands jamest's behavior *** btami has quit IRC *** btami has joined #gnuenterprise *** kmk has quit IRC *** kmk has joined #gnuenterprise *** kaimartin_ has joined #gnuenterprise *** johannesV_ has quit IRC *** kmk has quit IRC quick quesion wrt focus stops in forms right now there is code that builds a list of all objects that can get focus then if entries have focusorders set like focusorder="5" focusorder="1" focusorder="3" and there are actually 6 items that can get focus *** neilt has quit IRC then it attemptd to place the requested focus orders in place *** btami has quit IRC *** btami has joined #gnuenterprise and fill in the empty ones based upon order of entry in the gfd jamest, I think you're boring people... they keep leaving jcater: i get that a lot is this the behaviour we want fine w/me *** btami has quit IRC *** btami has joined #gnuenterprise why someone sets focusorder only 3 of 6 ? i think if one sets focusorder for an entry, it have to set it for all on that page this is the current attempted behavour so the 3/6 scenario maybe a bug it doesn't work at all at least mine doesn't if you have the above setup *** btami has quit IRC *** btami has joined #gnuenterprise and one of the items is focusorder="7" then it'll traceback i see i can see some advantage to this setup bbiab phone * jcater envisions jamest running to an old-style centrally-located black phone in that you'd only have to specify a few items in the list that needed moved about and it's autofill the rest but i could also see it putting all the specified ones in order at the front and filling in the rest after those try your first idea, and if one complains then reconsider... night all *** btami has quit IRC I'd see the former it's what I envisioned but could I suppose be unpredictable well, not unpredictable *** kaimartin_ has quit IRC *** kaimartin_ has joined #gnuenterprise *** jamest has left #gnuenterprise *** SachaS has quit IRC *** yure has quit IRC *** kaimartin_ has quit IRC *** kmk has joined #gnuenterprise *** jamest has joined #gnuenterprise *** kaimartin_ has joined #gnuenterprise *** kmk has quit IRC Hi. Is anybody listening with comprehensive gnue trigger experience? *** jcater has quit IRC *** SachaS has joined #gnuenterprise *** klasstek has quit IRC Anyone listening at all? ? yes :-) I think I need some basic trigger advice... in forms? yes. My ojective is to design a tool for my purchases. Obviously this involves suppliers, commodities and orders. Gnue should keep track of them and allow for easy input of data. *** sjc has quit IRC When it comes to the order form, I need to refer to the suppliers table in a way that is beyond a simple listbox. ok If an item is chosen from the list of available commodities, the form should auto-fill the entries for price and supplier. Every time the item is changed, a query to the database should be fired. ok Can you point me to working example of such a trigger? IANAP (I am not a python programmer) ok, just a sec (smiles in anticipation) now this trigger isn't perfect what it does is different, it queries the form to pull up all invoices in a certain state automatically so my accountant can select Pending from a dropdown and the form queries all invoices that haven't been approved yet ## value = unbound.fldfilterOn.get() # getParameter('queryType') if len(value): blkInvoice.initQuery() setParameter('queryType','%s' % value) blkInvoice.executeQuery() setParameter('queryType','A') (please let me know if all the xml enties are visible as kopete butchers them) in that trigger unbound is a block containing the dropdown named fldfilterOn that is how you'd get the current key value of the dropdown but if you aren't wanting to query the entire block then you'd need just pull the desired values and populate the fields of the block i'll find a sample of that *** kmk has joined #gnuenterprise ok in this trigger from my shipping form (actually, this shouldn't even be in this form...sigh) blkItem.prevRecord() skuCode = str(blkItem.fldSKU.get()) if len(skuCode): results = dtsItemDetails.simpleQuery({"sku":skuCode}) if len(results): result = results[0] blkItem.fldDescription.set("%s" % result['description_short']) blkItem.fldLocation.set("%s" % result['location']) blkItem.fldSKU.set("%s" % result['sku']) blkItem.nextRecord() else: for a in range(3): beep() time.sleep(.25) form.showMessage('SKU Code not in Item table!\nUnable to accept!', 'Alert!!!') blkItem.fldSKU.set("") setFocus(blkItem.fldSKU) return this is cruft from an earlier interation of my shipping form but I think the code is still valid ignore the prevRecord() call as it's not needed when a sku is entered the form queries against a datasource for a matching record (that is the only function of that datasource and it is not tied to any blocks) if the query returns a result then the 1st result (result = results[0]) is used to populate the fields in the block if nothing is returned by the query then the form beeps 3 times and displays a warning message when the shipper hits the OK button the sku is wiped clean and focus is forced back to the sku field for another attempt so you should be able to take the post-change trigger as in the first sample and couple it with the datasource simple query in the second to populate your fields did that help? bbias *** kaimartin_ has quit IRC *** menomc has joined #gnuenterprise *** mnemoc has quit IRC *** kmk has quit IRC *** jcater has joined #gnuenterprise *** kmk has joined #gnuenterprise *** menomc is now known as mnemoc hmmm was it that bad? :) Ouups. I misssed the hole of your hints. But I could recover it from the log :-) Now I need some time to chew on it... kmk: you are kaimartin_? Thank you for responding so fast and comprehensive! yes. I didn't get the significance of the "alternative nick" when I configured the irc reader. when your nick is in use, you get logged as your "alternative nick" Ok. My alt nick is already reconfigured as kmk_ mine is menomc mnemoc: i'd never notice that :) :) when you over read it, you can't notice the diference :p jamest: The xml braces are visible (at least they found there way into the log) ok, kopete won't show anything inside so I don't see if my code is even close to right :) is everything visible. (using KSirc v1.3.11) \part(gnight) \part *** kmk has quit IRC *** SachaS is now known as SachaZzzz *** jcater has quit IRC *** jcater has joined #gnuenterprise *** jamest has quit IRC *** jcater has quit IRC *** ninja has joined #gnuenterprise what's the state of play with the 'financials' module? http://www.gnuenterprise.org/packages/financials.php says 'Status <$ status ( "financials" ) >' hi ninja only some base classes to use in your own implementation hi is there a generally agreed database schema for things like 'chart of accounts' and 'transactions' etc yet? you can look for gnue-luca which is the beginning of general-ledger application does any of this stuff need svn-based tools, or can i get by with standard 'apt-get'-able stuff? * ninja figures he's got about two hours to see what he can get set up :) i've been manually keeping my accounts records in about three different places, it's about time i consolidated them into one decent, flexible, customisable system i think gnue-luca only available from svn, too early to be released/packaged ok, i'll have to take a look when i've got more bandwidth to download it ninja: talk with reinhard, he'll wake up soon :) i'll try to add some notes to the AccountingInfo wiki page too, as I have some experience writing financials/ledger software at an old company many moons ago ninja: are you able to grab gnue-luca from svn? dimas: working on it - i'll download a copy to my homedir on my home server - i want that to serve accounting data/interfaces to my computer and my gf's so she can start handling some of this stuff too i'll check out the whole svn trunk and try to figure out how it fits together i looked at it all about a year ago, but didn't get anything useful working (just lots of dependency errors) and didn't have time to pursue it (then) i have few extra forms to gnue-luca classes as i played with it some too