IRC logs of #tryton for Saturday, 2009-01-24

chat.freenode.net #tryton log beginning Sat Jan 24 00:00:01 CET 2009
X0d_of_N0dcedk: hey, I've got two fields I want to equally streach across to fill a line00:09
X0d_of_N0dcedk: how would I do that?00:10
X0d_of_N0dI tried to create a group and use fill="1" and expand="1" and that didn't work00:10
X0d_of_N0dand I tried fill="1" on the field...00:10
carlosgood night!!!00:14
cedkX0d_of_N0d: it is not possible00:14
X0d_of_N0dhum00:14
X0d_of_N0ddamn00:14
X0d_of_N0dthat sucks00:14
cedkX0d_of_N0d: fill attribute is to set the last column to fill00:14
cedkX0d_of_N0d: but the client can remember the column size of view00:15
cedkX0d_of_N0d: if you enable the option in the form preferences00:15
X0d_of_N0dso I need to set a static column size00:15
cedkX0d_of_N0d: or you can set a width to the columns00:15
cedkX0d_of_N0d: but I think the best is to let the user decide00:16
X0d_of_N0duser decide?00:16
cedkX0d_of_N0d: yes adjusting the size00:16
X0d_of_N0dI'm trying to design a view00:16
X0d_of_N0dhow?00:16
cedkX0d_of_N0d: in Option>Form>Save Columns width00:17
cedkX0d_of_N0d: I need to go now00:17
cedkbye00:17
X0d_of_N0dcedk: later00:17
-!- vengfulsquirrel(n=ian@c-71-202-125-182.hsd1.ca.comcast.net) has joined #tryton00:22
vengfulsquirrelX0d_of_N0d: Hey dude, I think I quite possibly took the longest lunch anyone has ever taken and called "lunch".00:22
X0d_of_N0dlol00:23
X0d_of_N0dI'm fighting with the tryton ui xml stuff for the ldap module00:25
X0d_of_N0dman, coding is so easy... making things look right is such a pain00:25
X0d_of_N0dwhy can't everything just be commandline00:25
vengfulsquirrelha have you ever tried to format UI's on the command line00:26
vengfulsquirrelimpossible00:26
X0d_of_N0dever use xdialog?00:26
X0d_of_N0dhehe00:26
vengfulsquirrelNope00:26
X0d_of_N0dcommandline ui00:26
vengfulsquirrelI mainly do web development though00:26
vengfulsquirrelI'm still trying to figure out how everything even works in Tryton.00:27
X0d_of_N0dyeah, I really wish Tryton (or more accurately terp) had leveraged a more html style syntax00:27
X0d_of_N0dfor hr *use* hr00:27
X0d_of_N0dthen, if you need to, extend that00:28
X0d_of_N0dbut I guess that's the point of the web client, isn't it...00:28
X0d_of_N0dbah... give me curses00:28
ikks_ncurses rules big time!!!!00:30
X0d_of_N0dIf I'm in this project long enough I'd definitely want to build an ncurses interface to tryton00:31
X0d_of_N0dit would just make sense00:31
vengfulsquirrelYeah html/css kind of sucks though.00:31
ikks_The main problem in curses is utf-8900:31
ikks_utf-800:31
X0d_of_N0dyeah00:31
X0d_of_N0dI wonder what the roadmap is for that00:32
ikks_python has a nice curses integration.00:32
X0d_of_N0durxvt supports utf-800:32
ikks_I'm using rxvt-unicode00:32
ikks_but I have issues with tabs and accents00:32
X0d_of_N0dpython has an OK curses integration... I've never used curses in C so I don't know, but it seems like there are some problems with coding00:32
ikks_encoding in C too :(00:33
ikks_But vim and mutt manages it in a sweet way00:33
X0d_of_N0dI need to start using mutt00:33
ikks_If you plan to use mutt, also look at mairix to find mail.00:35
ikks_ACTION loves mutt00:35
X0d_of_N0done of my coworkers uses mutt, the other uses muttator on thunderbird00:36
vengfulsquirrelYeah I'm trying to figure out how to get the system to do the same thing that you can do from the gtk client but via python.00:37
vengfulsquirrelCompletely unrelated to UIs.00:37
vengfulsquirrelha00:37
X0d_of_N0dI also hate the ui stuff because I'm not really sure how it should look in the first place00:40
X0d_of_N0dikks_: so there is no gui designer for tryton, right?00:43
X0d_of_N0dikks_: you just have to edit the xml by hand00:43
ikks_no X0d_of_N0d there isn't00:43
ikks_yep, that's the way00:43
-!- juanfer(n=juanfer@190.157.120.122) has joined #tryton00:52
X0d_of_N0dis "toolbar" no longer in "tree"?01:44
X0d_of_N0dthe caps lock key needs to be striken from the keyboard01:59
vengfulsquirreldefinately02:01
X0d_of_N0dbrb02:27
vengfulsquirrelX0d_of_N0d: Hey I have some more questions03:11
X0d_of_N0dok03:11
vengfulsquirrelOh you are back03:12
vengfulsquirrelSo ignoring configurable boms tell me if this sounds alright03:12
X0d_of_N0dok03:13
vengfulsquirrelPhantom boms have multiple inputs and have multiple outputs, during the explosion of a bom that references a phantom bom we must merge the inputs and ouputs of the phantom bom.03:13
vengfulsquirrel I propose we remove the phantom product from the outputs and push all the phantom bom's outputs into the current outputs.03:13
vengfulsquirrelAlso all inputs to the phantom boms will be appeneded to the current inputs list.03:13
vengfulsquirrelMy major question is how do you think we should handle bringing in phantom products from stock if for example a production order was unfinished?03:14
X0d_of_N0dwait...03:14
X0d_of_N0dbom input? you mean bom lines?03:14
vengfulsquirrelSorry so a given bom has a list of inputs and a list of outputs.03:14
X0d_of_N0da bom is a list of products used to produce another product (or, I guess, multiple products)03:15
vengfulsquirrelYes03:16
vengfulsquirrelSo if in that list of products there is a "phantom" subassembly/product then we need to merge it into the current bom at the start of a production so as to not require an additional production order.03:16
vengfulsquirrelSince it will be constructed on the production line, right ?03:16
vengfulsquirrelIgnore workcenters and routings for now.03:17
X0d_of_N0dok, yeah03:17
X0d_of_N0dyeah, that is kind of a trick....03:18
X0d_of_N0dhum...03:18
vengfulsquirrelSo would you say bringing in phantom subassemblies from stock is kind of like the concept of WIP ?03:18
X0d_of_N0dyeah03:18
vengfulsquirrelOr is WIP more of like a partially completed single product.03:19
vengfulsquirrelLike a half painted car or something.03:19
X0d_of_N0dboth03:19
vengfulsquirrelha damn, well i have no idea how to handle a half painted car03:19
vengfulsquirrelthat will have to be added later somehow03:20
X0d_of_N0da half painted car would be taken care of at the workcenter03:20
vengfulsquirrelright but i mean if you had to push it back into storage for some reason03:20
X0d_of_N0dI would see it as producing nothing but using half the resources03:20
X0d_of_N0da half painted car....03:20
X0d_of_N0dhum03:20
vengfulsquirreloh okay, you mean since the product would already be in production03:21
X0d_of_N0dyeah, if you need to stock it again then you need to do something external to the system03:21
X0d_of_N0dwe just can't deal with that right now03:21
vengfulsquirrelgreat because i don't want to03:21
vengfulsquirrelanyways back to phantom boms then03:21
X0d_of_N0dI'm sure with time we could, but it's not a priority03:21
vengfulsquirreldoes my problem make sense?03:21
X0d_of_N0dyeah, but you don't need to care about it until we have workcenters03:22
vengfulsquirrelOr its not a problem its just an unsolved use-case.03:22
X0d_of_N0dusers would need to manually put in how much stock they're returning. during this time all the phantoms could be listed.03:23
vengfulsquirrelYou mean when a production goes unfinished?03:24
X0d_of_N0dyeah03:24
vengfulsquirrelDuring assignation I might need to implode parts of the bom to check for phantom products in stock, so maybe I can put a checkbox to do that at the Waiting state.03:24
vengfulsquirrelDo you think that would be fare enough ?03:25
vengfulsquirrel*fair03:25
X0d_of_N0da checkbox at the waiting state?03:25
X0d_of_N0dhum03:25
vengfulsquirrelYeah I guess the waiting state is still controversial, but that is where i proposed having an automatic substitute checkbox as well.03:26
X0d_of_N0dI feel like states are things that should go into the database.....03:26
X0d_of_N0dif it's not going to be saved inbetween draft and assigned then it doesn't need a state there03:27
vengfulsquirrelYeah they are, but they also reflect the user interface possiblities, similar to how you move a sale through multiple states.03:27
X0d_of_N0dit's a process03:27
vengfulsquirrelIt is going to be saved.03:27
X0d_of_N0dI don't think it should be saved between draft and assigned03:27
X0d_of_N0dif it can't be assigned it shouldn't go into some limbo state03:27
X0d_of_N0dit should go back to draft so you could do something else with it03:28
vengfulsquirrelOkay yeah its hard to explain in words03:28
vengfulsquirrelWe need an entire itnerface for manually subsituting things into the bom03:28
X0d_of_N0dif that's going to be a UI thing then that's cool, maybe make a circle instead of a square03:29
vengfulsquirrelSo I'm not sure where that would go03:29
X0d_of_N0dit would be a button on the bottom of the interface that would bring up a wizzard03:29
X0d_of_N0d"check for material availablily" or something03:29
X0d_of_N0dyeah, that'd be kind of a complex ui thing to code given the current architecture, but I really think it's sort of the right way to do things03:31
vengfulsquirrelYeah I just stole it from the packing paradigm of Draft->Waiting->Assigned->Done.03:34
X0d_of_N0dhum...03:34
X0d_of_N0dok03:34
vengfulsquirrelYeah do you think the wizard would be easier for the user?03:35
X0d_of_N0dvengfulsquirrel: much...03:35
X0d_of_N0dit would be harder for us to code I think, but easier for the end user03:35
X0d_of_N0dand more logical03:35
vengfulsquirrelbecause it would seem more maleable ?03:35
vengfulsquirrelOkay, so when I click pay on an invoice and a dialog box pops up that is using wizards right ?03:37
vengfulsquirrelWhereas I was talking about making a state in a workflow.03:37
-!- X0d_of_N0d(i=user@gateway/tor/x-45f67d87babcf217) has joined #tryton03:42
X0d_of_N0djoin #tryton03:43
vengfulsquirrelhey03:43
vengfulsquirrel(18:38:26) X0d_of_N0d: and more logical03:43
X0d_of_N0dhum03:43
X0d_of_N0dyes03:43
X0d_of_N0dhehe03:43
X0d_of_N0dok03:43
vengfulsquirrel(18:38:41) vengfulsquirrel: because it would seem more maleable ?03:43
vengfulsquirrel(18:40:07) vengfulsquirrel: Okay, so when I click pay on an invoice and a dialog box pops up that is using wizards right ?03:43
vengfulsquirrel(18:40:22) vengfulsquirrel: Whereas I was talking about making a state in a workflow.03:43
X0d_of_N0dok03:43
X0d_of_N0dright03:44
vengfulsquirrelYeah except I guess the waiting state in my workflow would become the user using the wizard.03:45
vengfulsquirrelI guess that might make sense03:46
X0d_of_N0dor maybe when you hit save it would check for the materials03:46
X0d_of_N0dand bring up a wizzard03:46
X0d_of_N0dexactly03:46
X0d_of_N0dcool03:47
X0d_of_N0dso, in order to avoid confusion, perhaps it would be wise to make the waiting box a circle or something to show it's different?03:47
vengfulsquirrelyeah okay well the wizard is going to need multiple states in itself03:48
vengfulsquirrel*screens03:48
vengfulsquirrelwhatever you want to call it03:48
X0d_of_N0dyou're thinking one for each subsitute?03:48
X0d_of_N0dwell... what would be the screens?03:49
vengfulsquirrelno, like explode the bom... that can be done at start i think, but then you need to choose if you want to auto allocate or auto pull phantoms03:49
vengfulsquirrelso you check whatever03:49
vengfulsquirreland then click03:49
vengfulsquirrelassign03:49
X0d_of_N0derm...wait03:50
X0d_of_N0dauto allocate?03:50
vengfulsquirrelha sorry so many words, assign03:50
vengfulsquirreldammit no i mean substitute03:50
X0d_of_N0dhehe, yeah03:51
vengfulsquirrelauto-substitute and auto-use-phantoms-from-storage03:51
X0d_of_N0dlets just drop subsitution for right now03:51
X0d_of_N0dI really think that should be for a later date03:51
X0d_of_N0dlater milestone03:52
vengfulsquirrelwell then we don't even need a wizard pretty much03:52
X0d_of_N0dit's easy to add on later, lets just not worry about it right now03:52
X0d_of_N0dcool03:52
vengfulsquirrelbut that seems to work better with my workflow plan03:53
X0d_of_N0dso then we really don't even have a waiting stage for milestone 103:53
X0d_of_N0d?03:53
vengfulsquirrelactually maybe not03:54
vengfulsquirreleither way03:54
X0d_of_N0dwe'll worry about that in a later milestone03:54
vengfulsquirrelif you don't install the substitute module then ... wait no wizard is better03:54
vengfulsquirrelYeah okay so if you do install the substitute module we will make the assign=> button start a wizard03:55
vengfulsquirrelwhereas now it will just go straight to assigned or fail03:55
X0d_of_N0dwe won't do a substitute module, we'll just add it to the next rev of the mrp module03:55
vengfulsquirrelso we can just ignore it for now because that's how we will integrate it later, which is unobtusive03:55
vengfulsquirrelyeah exactly except i want to make sure it will cleanly fit in later03:56
vengfulsquirrelwell sure enough03:56
vengfulsquirreland that sounds fine ot me03:56
X0d_of_N0dright03:56
vengfulsquirrel*to03:56
X0d_of_N0dcool03:56
vengfulsquirrel wait except the phantom thing, what do you think about that03:57
X0d_of_N0dthe phantom thing can't happen until we have WIP, so that's not something we can worry about now03:57
X0d_of_N0djust explode the bom and don't worry about it03:57
vengfulsquirrelokay04:00
vengfulsquirreldone and done04:00
X0d_of_N0dok, one more thing04:00
vengfulsquirrelyeah i have a bunch more things actually04:00
X0d_of_N0dok, also "done", in your workflow, doesn't need to be there. Once itmes are done the people who look at workorders don't care if things get moved to storage...04:00
vengfulsquirreli have to go soon though but go ahead04:00
X0d_of_N0dhummm04:01
X0d_of_N0dthis is quick04:01
X0d_of_N0donce you draft the moves the workorder is fulfilled04:01
X0d_of_N0dthat's it04:01
vengfulsquirrelHmm04:02
vengfulsquirrelYeah someone has to re-stock it right?04:02
vengfulsquirrelHaving draft moves attached to a finalized thing seems kind of awkward.04:02
X0d_of_N0dright, but the guys who look at workorders don't have to restock it04:02
vengfulsquirrelYeah but the workorders will all be marked done04:02
vengfulsquirrelthe parent production order just wouldn't be done04:03
X0d_of_N0dhum.....04:03
vengfulsquirrelhmm yeah okay i'll look at that04:03
vengfulsquirrelSomeone asked for that04:03
vengfulsquirreland it made sense to me04:03
X0d_of_N0dcool04:04
X0d_of_N0dalso, I think unfinished should be marked as "canceled" in the db... just imho04:04
vengfulsquirrelyeah i think that is right, except maybe unfinished goes to canceled04:05
vengfulsquirrelthat's the same problem as above though04:05
vengfulsquirrelgotta work out that draft orders thing04:05
X0d_of_N0dright04:06
vengfulsquirrelbut something that was unfinished will maybe take different outputs04:06
X0d_of_N0dok04:06
X0d_of_N0dso you said you had some stuff you need?04:06
vengfulsquirrelyeah crap where do i write all this down though04:06
vengfulsquirreli wish the mrp document had like an associated scratch pad04:06
vengfulsquirreli guess it'll be in the log04:06
vengfulsquirreli'll review it tomorrow, yeah my problem is04:07
vengfulsquirrelfor the configurable bom04:07
vengfulsquirreli was thinking you could associate a list of by-products with each configurable bom input and those would be thrown into the outputs when those bom lines were selected04:08
vengfulsquirrelActually here is a simpler solution04:09
X0d_of_N0dhum04:09
X0d_of_N0dyeah, ok04:09
vengfulsquirrelA configurable bom has a list of by-product outputs(independent of any configuration) and once the bom is configured all the outputs can be added or removed.04:09
X0d_of_N0dACTION doesn't think configurable boms should have multiple outputs04:10
vengfulsquirrelThe system will still work just for weird configurable boms that might make differing by-products you'll have a hard time configuring tons of them.04:10
X0d_of_N0da configurable bom should be a list of components that are purchased or manufactured04:10
vengfulsquirrelYeah, but what about the outputs?  There will probably be by-products, at least, right ?04:11
X0d_of_N0dby products should only be produced during the manufacturing of components from raw materials04:11
vengfulsquirreloh04:11
vengfulsquirreli see04:11
vengfulsquirrelokay so single output then you think ?04:12
X0d_of_N0das far as I'm concerned if you have a configurable bom with by-products you'd put that in a normal bom, then put the normal bom in a configurable bom04:12
X0d_of_N0dso yeah04:12
vengfulsquirrelOkay yeah sorry if I'm off sometimes, I'm kind of fighting thin air04:13
vengfulsquirrelSo that sounds great because it super simplifies everything04:13
vengfulsquirrelyou can't make more than one output and that output is chosen when configuring the configurable bom.04:14
X0d_of_N0dcool04:14
vengfulsquirreli think phantom boms could probably still work correctly in that situation04:15
X0d_of_N0dright... except potential multiple outputs made by children of the configurable bom04:15
vengfulsquirrelyeah but that's fine we don't deal with that until production line time04:15
vengfulsquirrelhopefully that makes sense04:16
X0d_of_N0dyeah04:17
vengfulsquirrelokay great so i think i'm ready soon to start breaking ground04:18
X0d_of_N0dsweet04:18
vengfulsquirrelor at least make way more detailed designs04:18
vengfulsquirrelmaybe start talking about the db and stuff04:18
vengfulsquirrelgotta run for now though04:18
vengfulsquirrelthanks for the help, i'll talk to you next week04:19
X0d_of_N0dok, talk to you monday probably04:19
vengfulsquirrelyeah sounds good04:19
vengfulsquirrellater04:19
X0d_of_N0dnp, later04:19
-!- yangoon(n=mathiasb@p549F5C6D.dip.t-dialin.net) has joined #tryton05:20
-!- ikks(n=igor@190.12.153.202) has joined #tryton05:36
-!- Frank159_(n=chatzill@pD9E63563.dip.t-dialin.net) has left #tryton07:35
-!- carlos(n=carlos@89.7.24.44) has joined #tryton08:48
-!- sharkcz(n=dan@plz1-v-4-17.static.adsl.vol.cz) has joined #tryton09:05
-!- Gedd(n=ged@77.109.114.26.adsl.dyn.edpnet.net) has joined #tryton09:09
-!- enlightx(n=enlightx@host-84-221-85-222.cust-adsl.tiscali.it) has joined #tryton09:21
-!- cristi_an(i=5978d3ce@gateway/web/ajax/mibbit.com/x-768acd54bec2b78c) has joined #tryton09:41
-!- carlos(n=carlos@89.7.24.44) has joined #tryton12:22
-!- carlos(n=carlos@89.7.24.44) has joined #tryton12:38
-!- cedk(n=ced@gentoo/developer/cedk) has joined #tryton12:46
-!- sharkcz(n=dan@plz1-v-4-17.static.adsl.vol.cz) has joined #tryton13:38
-!- cristi_an(i=5978d3ce@gateway/web/ajax/mibbit.com/x-26b20177a2f7d72b) has joined #tryton16:49
-!- cristi_an(i=5978d3ce@gateway/web/ajax/mibbit.com/x-b41c644d4aa8cf97) has joined #tryton16:50
CIA-8tryton: matb roundup * #697/Sort order of almost all tables under Administration doesn't respect translation: [chatting] The issue is partiallly solved. There are 2 remaining issues: 1) As you can see on search_inv.png, untranslated items (name Invoice) ...17:35
yangooncedk: are you around?17:37
cedkyangoon: yes17:42
yangooncedk: I need a little help getting report overrides working17:43
yangooncedk: I don't understand exactly, what I have to do for https://bugs.tryton.org/roundup/issue65717:44
cedkyangoon: you must make like a new report17:44
yangooncedk: but is it enough to define in xml?17:45
cedkyangoon: yes17:45
yangooncedk: then it just doesn't work for me in respect to https://bugs.tryton.org/roundup/issue65717:46
yangooncedk: because src strings of report, that overrides, are still shown belonging to module, that is overridden17:47
cedkyangoon: you must not override17:47
-!- bechamel(n=user@85.201.86.139) has joined #tryton17:48
yangooncedk: just tried again with invoice.xml like http://paste.pocoo.org/show/101145/17:55
yangoondoesn't work17:56
-!- sharkcz(n=dan@plz1-v-4-17.static.adsl.vol.cz) has joined #tryton17:58
cedkyangoon: this is because you override and not create a new one18:00
-!- bechamel(n=user@85.201.86.139) has left #tryton18:11
-!- bechamel(n=user@85.201.86.139) has joined #tryton18:11
CIA-8tryton: C?dric Krier <ced@b2ck.com> default * 1482:83a97e60fcc8 trytond/trytond/osv/orm.py: Add missing fuzzy test in _where_calc and _order_calc for issue69719:48
CIA-8tryton: ced roundup * #697/Sort order of almost all tables under Administration doesn't respect translation: [resolved] Fix with changeset 83a97e60fcc819:48
yangooncedk I think I got it now, but there is one thing20:09
yangooncedk: the print wizard for invoice has to find the internal name 'account.invoice'20:09
yangooncedk: so this field has to be the same for the new report20:10
yangooncedk: but there is a unique key on this field preventing the import of a second definition with this internal name20:10
yangooncedk: removing the unique key makes it work for me20:11
cedkyangoon: ok, I will remove it20:14
yangooncedk: ok, thx, it is ir_action_report_report_name_uniq UNIQUE(report_name)20:19
-!- vengfulsquirrel(n=ian@c-71-202-125-182.hsd1.ca.comcast.net) has joined #tryton20:38
-!- carlos(n=carlos@248.Red-83-39-84.dynamicIP.rima-tde.net) has joined #tryton20:53
CIA-8tryton: C?dric Krier <ced@b2ck.com> default * 1483:2c02e7aa0f5b trytond/trytond/sql_db.py: Add drop_constraint21:01
CIA-8tryton: C?dric Krier <ced@b2ck.com> default * 1484:61c4f969bc09 trytond/trytond/ir/action.py: Remove report_name_uniq constraint on ir.action.report21:01
cedkyangoon: done21:24
yangoonyes, great21:24
CIA-8tryton: C?dric Krier <ced@b2ck.com> default * 172:5439ae892af4 gentoo-overlay/dev-python/relatorio/ (Manifest relatorio-0.5.0.ebuild): Version bump21:28
-!- gremly(n=oscar@190.156.157.155) has joined #tryton21:31
yangooncedk: if I want to set the old report to inactive I get always:  WARNING:init:Field active of 293@ir.action.report not updated (id: report_invoice), because it has changed since the last update21:52
yangooncedk: which way do I have to go, if I want to disable the old report?21:52
bechamelyangoon: this mean that the record in the db is not the same that was imported the first time, i.e. it was changed by a user21:53
bechamelyangoon: and the code that import xml try to not overwrite user modifications21:53
bechamelyangoon: the easiest think to do is just to remove the record (has long as there is not other records poiting to it), and update all21:54
yangoonbechamel: I tried to remove all write_date and write_uid manually from the records, nevertheless the same message21:54
yangoonbechamel: from what does tryton decide, if a record has changed?21:55
bechamelyangoon: it's not enouhg the imported values are stored in a table (ir_model_data)21:55
yangoonok, thx!21:55
bechamelyangoon: if the actual record is not the same that the corresponding values in this table the record is not updated21:55
yangoonbechamel: so what do I have to do, it always possible that a user has changed the record21:56
bechamelyangoon: I never tried to remove records directly in ir_model_data but it should work21:56
bechamelyangoon: ?21:56
yangoonbechamel: if I want to set the old report to inactive21:57
bechamelyangoon: I think you cannot set the record to inactive because it was modified in the meanwhile with your tests21:58
yangoonbechamel: yes, that's it21:58
bechamelyangoon: but it should work21:59
yangoonbechamel: but this can happen always21:59
yangoonbechamel: provided the case, a user gets a module later and he has already worked on those records21:59
bechamelyangoon: yes, the best is not to change data "by hand" on thedb21:59
yangoonbechamel: how can I22:00
bechamelyangoon: in this case you only want to put active on an existing record22:00
yangoonsorry, provide on a first time import of the new module22:00
yangoonyes, ecactly22:00
yangoonsorry, relly slow on the keyboard today...22:01
bechamelyangoon: so just add a record with active false on the xml of your custom module, I think there is an example like that somewhere, let me chack22:01
bechamelyangoon: look this: http://hg.tryton.org/hgwebdir.cgi/modules/stock/file/19568e8d5cb8/stock.xml#l2122:03
yangoonbechamel: many thx22:03
bechamelyangoon: and: http://pastebin.com/m765be16d22:03
bechamelyangoon: it's actualy the opposite of what you want, the original record is set with active="0"22:04
bechamelyangoon: and the xml snippet "activate" the record22:04
yangoonbechamel: yes I see, trying at once22:05
bechamelyangoon: take care of the id of the record it's important to give the id of the record you want to overwrite prefixed with the name of the module it comes from.22:09
yangoonbechamel: yes, just ran into the trap22:09
yangoonbechamel: ;)22:09
yangoonbechamel: I now have http://pastebin.com/m64fc31022:12
yangoonbechamel: but get always the same warning22:13
yangoondo I work on the wrong table?22:13
bechamelyangoon: the warning is "not updated ... because it has changed since the last update" ?22:14
yangoonbechamel: yes22:14
bechamelyangoon: I will test here22:15
yangoonbechamel: ok, thx22:15
bechamelyangoon: i have the same error :) i don't remember how the code i gave you work correctly22:18
yangoonbechamel: puuh:)22:18
-!- oversize(n=manuel@94.219.105.90) has joined #tryton22:28
yangoonbechamel: I think, there is also a bug in updating timestamps in ir_model_data22:32
yangoonbechamel: when I am updating account_invoice, ir_model_data.write_date and ir_model_data.date_update are set to the actual date22:32
yangoonbechamel: when I am updating my custom module with the new report, the dates are not updated22:33
bechamelyangoon: and what is the bug ? you mean that the timestamps shouldn't be changed ?22:33
yangoonbechamel: don't know, either it should be changed in both cases or in none22:34
yangoonbechamel: and always ir_model_data.write_date < ir_model_data.date_update22:34
yangoonbechamel: so I don't know, how tryton evaluates "...changes since the last update..."22:35
bechamelyangoon: date_update is not used actualy ...22:36
yangoonbechamel: perhaps just evaluating write_date not null?22:36
yangoonbechamel: ok22:36
bechamelyangoon: i don't see how the xml I gave you works, it's crazy how am i able to forget this kind of stuff22:41
yangoonbechamel: :D22:41
yangoonbechamel: and I get still the warning after deleting the records from ir_model_data22:41
bechamelyangoon: yesn and it's the normal behaviour22:42
yangoonbechamel: seems to me, that it probably should work, but that some bug was introduced meanwhile22:42
bechamelyangoon: No it's working, but i don't how22:46
yangoonbechamel: thx anyway, have to eat sth. now, cu later22:53
-!- ikks(i=igor@190.102.215.179) has joined #tryton22:57
vengfulsquirrelDoes anyone think someone would want to use BOMs only without any sort of production order process ?  Maybe for  some sort of accounting purpose?23:28
bechamelvengfulsquirrel: it may be useful if the 'production' is very light or made in front of customers , like in a restaurant or a gift shop (to wrap object in paper), ...23:33
bechamelvengfulsquirrel: or maybe for promotion, "free mouse if you buy the laptop" (the package is a product by itself)23:34
bechamelvengfulsquirrel: but i'm not shure if it must be handled with bom23:35
vengfulsquirrelyeah that's a different idea, although that one is useful23:35
vengfulsquirrelthat's more of a like a product kit23:36
bechamelvengfulsquirrel: yes a kit. I think it's managed with bom in openerp23:37
vengfulsquirrelYeah I think that's a little weird because its like a marketing idea.23:38
bechamelvengfulsquirrel: it you only manage sale and purchase, a bom is not necessary. But inventory stuff need to which are real products in the kit23:40
bechamelsorry: ... need to _know_ which ...23:41
vengfulsquirrelyeah seems like that would be a modification between the sale and the packing generation23:44
-!- igor__(i=igor@190.102.205.198) has joined #tryton23:45
vengfulsquirrelI was thinking about putting simple BOMs, production orders and simple production order processing into the same module.23:47
bechamelvengfulsquirrel: you plan to have distincts object production orders (simple and not simple) ?23:49
vengfulsquirrelWhat do you mean ?23:49
bechamelvengfulsquirrel: ok  I misread, i did not see the "processing" word23:50
vengfulsquirrelOh right by that I just mean no shop floor control, it just helps you handle your stock.23:51
bechamelvengfulsquirrel: yes and this module should be easily extendable to handle more complex  cases23:53
vengfulsquirrelyeah hopefully, i'm kind of worried about some things not extending very well but we have to start somewhere23:55

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!