IRC logs of #tryton for Monday, 2009-02-02

chat.freenode.net #tryton log beginning Mon Feb 2 00:00:02 CET 2009
-!- FWiesing(n=Wiesinge@194.208.185.12) has left #tryton01:34
-!- vengfulsquirrel(n=ian@c-71-202-125-182.hsd1.ca.comcast.net) has joined #tryton05:11
-!- vengfulsquirrel(n=ian@c-71-202-125-182.hsd1.ca.comcast.net) has left #tryton05:11
-!- yangoon(n=mathiasb@p549F7CDE.dip.t-dialin.net) has joined #tryton05:20
-!- nicoe(n=nicoe@ip-80-236-193-136.dsl.scarlet.be) has joined #tryton07:25
-!- enlightx(n=enlightx@host-78-13-119-242.cust-adsl.tiscali.it) has joined #tryton08:00
-!- Gedd(n=ged@77.109.113.162) has joined #tryton08:06
-!- carlos(n=carlos@89.7.24.44) has joined #tryton08:19
-!- sharkcz(n=dan@plz1-v-4-17.static.adsl.vol.cz) has joined #tryton08:31
-!- cedk(n=ced@gentoo/developer/cedk) has joined #tryton08:49
-!- enlightx(n=enlightx@host198-42-static.85-94-b.business.telecomitalia.it) has joined #tryton09:47
-!- LordVan(n=lordvan@gentoo/developer/LordVan) has joined #tryton09:57
-!- LordVan(n=lordvan@gentoo/developer/LordVan) has joined #tryton10:01
-!- Gedd(n=ged@ip-80-236-193-136.dsl.scarlet.be) has joined #tryton10:04
-!- simahawk(n=simao@host-84-222-63-237.cust-adsl.tiscali.it) has joined #tryton10:08
-!- Timitos(n=Timitos@88.217.184.172) has joined #tryton11:49
-!- bechamel(n=user@85.201.86.139) has joined #tryton11:49
-!- enlightx(n=enlightx@host-78-13-119-242.cust-adsl.tiscali.it) has joined #tryton12:47
-!- jporcel(n=jporcel@62.57.72.13.static.user.ono.com) has joined #tryton13:00
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1508:dfa3b49d1209 trytond/trytond/web_service/db.py: Use template0 when restoring database13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1509:102e1348acb0 trytond/trytond/tools/convert.py: Remove unused import13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1510:b07795fac9a7 trytond/trytond/ (21 files in 9 dirs): Add modelstorage and modeldb13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 374:a0c1b92f0d47 account/ (account.py journal.py move.py tax.py): Fix for new modeldb13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 260:c4743cf2265f account_invoice/ (invoice.py payment_term.py): Fix for new modeldb13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 58:96116eede04a analytic_account/account.py: Fix for new modeldb13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 39:c3676e3463cd analytic_invoice/invoice.py: Fix for new modeldb13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 40:560c972316a2 analytic_purchase/purchase.py: Fix for new modeldb13:14
CIA-10tryton: ced roundup * #766/Exception: ('Argument must be string or unicode.', 'Traceback (most recent call last):\n File "/trytond/netsvc.py", line 282, i...: [resolved] Fix with changeset 53270480858f13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 32:0aa8ae67fa9a analytic_sale/sale.py: Fix for new modeldb13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 127:cbde71ecbb87 product/uom.py: Fix for new modelstorage13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 216:1bfc452d3eaf purchase/purchase.py: Fix for new modeldb13:14
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 113:d03be09667ba sale/sale.py: Fix for new modeldb13:15
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 474:53270480858f stock/ (location.py product.py): Fix for new view_header_get must return string for issue76613:15
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 475:970cf34ca7b1 stock/packing.py: Fix for new modeldb13:15
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 61:23b942619d75 timesheet/line.py: Fix for new view_header_get must return string13:15
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1173:9b9390ec9d0c tryton/tryton/gui/window/win_import.py: Improve memory used of import csv by removing StringIO13:21
-!- BleSS(n=neo@69.Red-83-39-51.dynamicIP.rima-tde.net) has joined #tryton13:22
BleSSI've seen a demo about how can be customized an openerp project13:28
BleSSI'm supposed that can be made the same with tryton, doesn't?13:28
cedkBleSS: globally, yes but there is some differences13:34
BleSSdoes OpenObject framework is also used by tryton?13:36
BleSSI was reading about differences between openerp and tryton, there are changes important and interesting overall related to conection/network13:38
carlosBleSS: as far as I know, OpenObject is an abstraction of the TinyERP/Tryton infrastructure so we may say that Tryton uses a modified version of OpenObject framework13:39
carloswell, is more correct to say that OpenObject is a modificed version of the original TinyERP/Tryton infrastructure13:40
BleSSI question myself if those changes could not be sent to openerp, and avoid a fragmentation. (I mean to have a community greater)13:44
cedkBleSS: we try it but it was not possible13:45
carlosBleSS: well, Tryton is a fork of TinyERP with a lot of rewrites not small changes13:45
carlosand then, TinyERP guys evolved their system to OpenObject13:46
BleSSwell, I need create/customize a little application (erm) abount accounts, reporting, etc13:48
BleSSI want to know if is posstible also intryton13:48
cedkBleSS: yes, it is possible to customize it13:48
-!- nicoe(n=nicoe@ip-80-236-193-136.dsl.scarlet.be) has joined #tryton13:48
BleSSwell, then I'm going to startin by the developer manual, thanks13:49
-!- ikks(n=igor@190.144.69.234) has joined #tryton14:00
-!- BleSS(n=neo@69.Red-83-39-51.dynamicIP.rima-tde.net) has left #tryton14:10
carlosTimitos: Hi, seems like the chart of accounts is working now. I moved all root types to a common parent as you suggested and seems like it works now (not sure whether that was the required fix or just a mix of different changes I did...)14:11
carlosTimitos: but anyway, it works now :-P14:11
Timitoscarlos: great14:12
carlosI'm now with the taxes part and it should be functional14:12
carlosTimitos: however, I have a doubt about the way I wrote the chart of accounts14:12
carlosTimitos: I have two chart of accounts, the regular one and a simplified version14:13
Timitoscarlos: Did you put them into one module? Is this needed? Perhaps it would be better to create two modules?14:14
carlosthe accounts are the same, but the simplified version have reports simplified, and thus, some account types are different from the regular one14:14
carlosTimitos: well, you should use one or the other depending on your past year incomes and number of employees14:15
carlosso I was thinking on adding a wizard to switch from one to the other14:16
carlosand in Tryton you get different chart of accounts templates, so you choose the one you want to use, instead of having to look for all modules before you decide which one to use14:16
Timitoscarlos: hm. this is indeed a special feature for spain. in germany you should always keep the same chart if it is possible14:18
carloswell, you cannot change it in the same year14:21
carlosbut I think it may be useful if you do a mistake and start with the wrong one14:22
-!- LordVan(n=lordvan@gentoo/developer/LordVan) has joined #tryton14:22
carlosgiven that the account numbers and usage is the same, and the only difference are the reports14:22
Timitoscarlos: yes. as only the reports are different there should be not so much to do.14:23
carlosbut I need both chart of accounts, right? there is no other easy way to do it in Tryton, isn't it?14:24
carlosTimitos: also, I was thinking on adding another chart of accounts template for individuals that are their own company in the same module. Why do you think is better to split it?14:25
Timitoscarlos: i haven´t thought about that yet.14:25
carlosok14:25
carlosI have it done anyway, so I could migrate it later if we find a good way to do it14:25
Timitoscarlos: for me i would prefer to split it into more modules. so i only have those accounts and account.types i really need.14:26
carlosaren't those just templates until you create a chart of accounts for a company?14:27
Timitoscarlos: but are you able to select these charts by your needs? i haven´t tried that yet. perhaps your way will work too.14:28
carlosTimitos: what I get right now is the same as having two different modules that provide a chart of accounts14:29
carlosI get the basic chart of accounts that comes with Tryton by default14:29
carlos+ the German one (I activated it)14:30
carlos+ both chart of accounts that my module provides14:30
carlosthe only problem I had having both chart of accounts in the same module is that I had to use different IDs for the xml entries14:30
carlosACTION is using two different .xml files, one for each chart of accounts14:31
TimitosACTION is testing carlos account chart and has now a freezing client.....14:39
carlosTimitos: I tested it with 1.0 series and it takes a while, but it never froze my client14:40
Timitoscarlos: i will try again14:41
carlosACTION -> out14:53
carlossee you later14:53
Timitoscarlos: cu14:57
carlosback15:21
carlosTimitos: were you able to load my module or is it still causing problems?15:51
Timitoscarlos: it worked now.15:51
carlosok15:51
Timitoscarlos: it takes really much time.15:51
carlosI guess that's because it's huge15:52
carlostwo 'big' chart of accounts...15:52
Timitosit seems to be really huge. and i thought my one is already huge :-)15:52
carlosIt may be an argument to split it15:53
carlosor maybe to improve the parser/importer performance :-P15:53
carlosyeah, yours is also big15:53
carlosin fact, yours is bigger than the ones from Spain15:54
Timitoscarlos: but mine does not take so much time. but i think this is because of your two charts in one module15:55
carlos449K vs 349K + 360K15:56
carlosACTION is talking about file size15:56
-!- udono(n=udono@dynamic-unidsl-85-197-24-127.westend.de) has joined #tryton16:20
-!- juanfer(n=juanfer@190.144.69.234) has joined #tryton17:09
cedkcarlos: I don't think it is the xml parser fault17:15
cedkI think it is the insert into the database that is slow17:15
carloscedk: I was just joking17:15
carlosI'm sure it's the database insert, it's a lot of information17:16
udonocarlos: how many records you try to insert?17:16
carlosudono: it's a chart of accounts17:17
carlosactually, two of them17:17
cedkthe main issue, I think it is that we create on INSERT per record17:17
bechamelupdating indexes  and one by one insert are costly but i don't see a way to improve that17:18
cedkin psycopg, there is the possiblity to make multiquery in one request17:18
bechamelcedk: the problem is when there are reference for id that are in the same file17:18
carlosthe total number of <record> tags are 2084 in total (both chart of accounts)17:19
cedkbechamel: to have multiqueries, we must create a new xml syntax17:19
carlosand that's without the tax information17:19
carloscedk: I'm wondering, why are we using 'parent' references instead of nested tags?17:20
carlosI'm not sure whether that's going to improve the performance, but at least nested tags seems more correct when using XML, don't you think so?17:22
cedkcarlos: no, because you can have many trees in one model17:22
bechamelcarlos, cedk and above all the xml reflect the db, parent is just a foreign key17:24
carlosyou said it, trees, without loops, right?, so that shouldn't be a problem...17:25
carlosbechamel: so the XML is more an exported SQL table using XML instead of SQL17:25
bechamelcarlos: xml is a ready-to-be-imported-in-the-db file :)17:26
carlosI guess the performance of the python code that process such xml file is better as it's right now than if it needs to map the nested tags to foreign keys17:27
cedkcarlos: we can have trees with loop17:27
carloscedk: really? in that case there is no way to use nested tags. I didn't get any loop in my chart of accounts neither I was able to image its utility, so I assumed there aren't such loops17:29
udonoWe are using XML only for inserting data out of modules and via XMLrpc. For me it is not a problem, that it takes some time...17:29
carlosudono: well, as I said, I was joking about improving its performance, given that the speed problem that Timitos got is easily fixable using different modules for different chart of accounts17:31
cedkcarlos: loop in account charts is forbidden but for other models it can be allowed17:33
carlosok17:33
cedkcarlos: by the way, we use a sax parser which has better performance on big file17:37
carlosdo you think the speed may change if I merge both xml files in a single one?17:39
cedkcarlos: no, I don't think it will be visible17:40
carlosok17:40
cedkcarlos: but I think like timitos, that it will be better to have two modules17:40
carlosyeah, that's what I'm going to do17:41
-!- simahawk(n=simao@host-84-222-63-237.cust-adsl.tiscali.it) has joined #tryton18:05
cedkcarlos: and by the way, the installation of a module is all in one transaction, so that can also slowdown the import process if there is a lot of data18:31
carlosindeed18:32
Timitosthis is an important notice. yes18:32
-!- vengfulsquirrel(n=ian@c-71-202-125-182.hsd1.ca.comcast.net) has joined #tryton18:51
-!- FWiesing(n=Wiesinge@194-208-185-012.tele.net) has joined #tryton19:19
-!- panthera(n=daniel@unable-to-package.org) has joined #tryton19:43
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1328:3b386601061b trytond/trytond/ir/model.py: Restart cache on fields_view_get after model access change19:47
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1329:201f5660bc9a trytond/trytond/osv/orm.py: Add missing fuzzy test in _where_calc and _order_calc for issue69719:47
CIA-10tryton: Bertrand Chenal <bch@b2ck.com> default * 1330:4a3b8eac3a13 trytond/trytond/tools/convert.py:19:47
CIA-10tryton: xml import: Fetch the correct default value when a field belongs to19:47
CIA-10tryton: the parent model of the record.19:47
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1331:3f717d2af4a8 trytond/trytond/tools/convert.py: Use default value from base model before of inherit model for xml import19:47
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1332:54e82a5ef3e0 trytond/trytond/res/user.py: Reset cache for check and fields_view_get when writing res.user19:47
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1333:3c307125cf73 trytond/trytond/osv/orm.py: Disable by default sort on One2Many fields19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1334:72e1824a61bf trytond/trytond/osv/orm.py:19:48
CIA-10tryton: Fix Modified Preorder Tree Traversal19:48
CIA-10tryton: - _update_tree works if only one record has changed19:48
CIA-10tryton: - can not use child_number in update but instead old_right - old_left19:48
CIA-10tryton: - set old_left if the same than old_right19:48
CIA-10tryton: - pre-select ids before update to prevent collision19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1335:d27fd05245e9 trytond/trytond/osv/orm.py: Don't loop on id in delete for _update_tree because _rebuild_tree will be called many times.19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1336:8c24e50b35ab trytond/trytond/ (ir/translation.py osv/osv.py): Fix Cacheable to prevent attributes collision19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1337:fc66b065fe66 trytond/trytond/web_service/db.py: Use template0 when restoring database19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 206:c890bd635857 account_invoice/invoice.py: If amount_to_pay_today is zero used amount_to_pay in pay invoice wizard19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 207:04fd17846df0 account_invoice/CHANGELOG: Add changelog19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 396:fd1c6a67a106 stock/packing.py: Missing context in assign_force of internal packing19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 397:e0aebda37086 stock/move.py: Set effective_date also for assigned moves19:48
CIA-10tryton: Bertrand Chenal <bch@b2ck.com> default * 398:96d4068b9809 stock/packing.py: Fix warehouse default value: use list instead of string for domain.19:48
CIA-10tryton: Bertrand Chenal <bch@b2ck.com> default * 399:f847ec3157f8 stock/packing.py: Fix on_change: clean location and address when custome/supplier is unset.19:48
CIA-10tryton: Bertrand Chenal <bch@b2ck.com> default * 400:881e02314ff6 stock/packing.py: Fix: use sequence on inventory packing creation.19:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1075:29471f54e330 tryton/tryton/gui/window/ (win_export.py win_import.py): Use None instead of False when calling fields_get for import/export for issue75919:48
CIA-10tryton: C?dric Krier <ced@b2ck.com> default * 1076:de86ed7470a2 tryton/tryton/gui/main.py: Add logout before backup database for issue76319:49
-!- panthera_(n=daniel@unable-to-package.org) has joined #tryton19:49
-!- paola(n=paola@host-84-223-76-210.cust-adsl.tiscali.it) has joined #tryton20:11
-!- cedk(n=ced@gentoo/developer/cedk) has joined #tryton20:17
-!- cristi_an(i=5978d3ce@gateway/web/ajax/mibbit.com/x-b0defc9016238164) has joined #tryton20:19
-!- Timitos(n=Timitos@88.217.184.172) has joined #tryton20:41
-!- bechamel`(n=user@85.201.86.139) has joined #tryton20:53
-!- enlightx_(n=enlightx@host-78-13-120-215.cust-adsl.tiscali.it) has joined #tryton21:07
cedkvengfulsquirrel: ping21:15
vengfulsquirrelcedk: Hey21:15
cedkvengfulsquirrel: I checked your patches21:16
cedkvengfulsquirrel: for check_type_for_moves21:16
cedkin the search call, you set double [[]]21:16
cedkthis is not needed, one [] is enough21:17
cedkand no need to set context=None in the call21:17
vengfulsquirrelHmm yeah but double [[]] is still okay right its just not the best?21:18
cedkvengfulsquirrel: it will be slower21:19
cedkand generate uggly SQL21:19
vengfulsquirrelOkay noted21:19
cedkvengfulsquirrel: if you fix those two remarks, you can generate the patch files for inclusion21:20
vengfulsquirrelFor the context=None, I read the coding guidelines and it said "Calling method with default values must explicitly use keywords arguments."21:20
vengfulsquirrelHa I think maybe I read too much into that, what is the context for ?21:21
cedkvengfulsquirrel: when you have values but here you don't have any21:21
cedkvengfulsquirrel: context is for contextual value, this can be used for many things like language, timestamp check, etc.21:22
cedkvengfulsquirrel: so what we want to say by explicit call, it is that when you have a value for an args that is default, you must use "="21:23
cedkvengfulsquirrel: because like that we can change the argument order without any problems21:23
cedkvengfulsquirrel: when you have the two patches, you can assign to me the issue21:23
cedkvengfulsquirrel: and ping me21:24
vengfulsquirrelcedk: Okay thanks, I think you meant "Never pass keyword arguments as positional arguments."21:28
cedkvengfulsquirrel: yes21:33
cedkvengfulsquirrel: you can update the wiki21:33
cedkvengfulsquirrel: if you have better english sentence21:33
-!- sharkcz(n=dan@plz1-v-4-17.static.adsl.vol.cz) has joined #tryton21:54
-!- paola_(n=paola@host-84-223-76-210.cust-adsl.tiscali.it) has joined #tryton22:47
-!- paola_(n=paola@host-84-223-76-210.cust-adsl.tiscali.it) has left #tryton22:47
-!- tekknokrat(n=gthieleb@dslb-088-074-150-146.pools.arcor-ip.net) has joined #tryton23:09
CIA-10tryton: vengfulsquirrel * r432 /wiki/CodingGuidelines.wiki: Cleared up arguments verbiage.23:24

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