IRC logs of #tryton for Tuesday, 2010-01-19 #tryton log beginning Tue Jan 19 00:00:02 CET 2010
-!- irclog( has joined #tryton00:47
-!- FWiesing( has left #tryton00:47
-!- tekknokrat( has left #tryton00:53
-!- vengfulsquirrel( has joined #tryton02:28
-!- ikks(n=ikks@ has joined #tryton04:25
-!- ikks(n=ikks@ has joined #tryton04:47
-!- yangoon( has joined #tryton05:18
-!- LucaSub( has joined #tryton08:24
-!- enlightx( has joined #tryton08:38
-!- vengfulsquirrel( has joined #tryton09:06
-!- paepke( has joined #tryton09:08
-!- cedk(n=ced@gentoo/developer/cedk) has joined #tryton09:10
-!- bechamel( has joined #tryton09:38
CIA-5reihle roundup * #1385/Problem with CalDAV and Lightning 1.0: As Timitos wrote in his post on Bugzilla, I connected with Evolution, and the calendar entries looked as they should. That makes it clear that the ...10:12
-!- LucaSub( has left #tryton10:13
CIA-5matb roundup * #1386/Module company doesn't install together with party_bank_de:11:12
CIA-5[new] Following two modules install without problems on 1.411:12
CIA-5 ...11:12
-!- paepke_( has joined #tryton11:23
-!- essich_( has joined #tryton12:11
CIA-5C?dric Krier <> default * 1490:9f1d4df6387c tryton/tryton/gui/window/view_form/model/
CIA-5Send signal record-changed to the wrong parent12:29
CIA-5It must be send to the ModelRecord of the field and not to the12:29
CIA-5Bertrand Chenal <> default * 2143:0334cc7b7b46 trytond/trytond/ (ir/ res/ res/ir.xml): Add rules on sequences12:40
CIA-5Bertrand Chenal <> default * 2144:8b79ad8ee8a3 trytond/trytond/model/ Merge12:40
CIA-5Bertrand Chenal <> default * 307:ec7bf061fcc0 party/party.xml: Add groups on sequence type12:40
CIA-5Bertrand Chenal <> default * 548:2b285d38426d account/ (journal.xml move.xml): Add groups on sequence types12:41
CIA-5Bertrand Chenal <> default * 400:c40e7ed7a84b account_invoice/invoice.xml: Add groups on sequence type12:41
CIA-5Bertrand Chenal <> default * 229:92d1ebcc025a sale/sale.xml: Add Sale Administrator group. Add groups on sequence type12:41
CIA-5Bertrand Chenal <> default * 358:1590eba6f1a1 purchase/purchase.xml: Add groups on sequence type12:41
CIA-5Bertrand Chenal <> default * 632:9912452bb68e stock/shipment.xml: Add groups on sequence types12:42
CIA-5Bertrand Chenal <> default * 633:d4786180e68d stock/ Merge12:42
CIA-5htgoebel roundup * #1387/"Create Database" does not reconnect to server:12:48
CIA-5[new] * Start Server12:48
CIA-5* Start Client, but do _not_ connect to a database12:48
CIA-5* stop server12:48
CIA-5* In the Client:12:48
CIA-5- select "Create Database"12:48
CIA-5- can n ...12:48
CIA-5Bertrand Chenal <> default * 2145:be03fd18a19b trytond/CHANGELOG: Updated CHANGELOG12:53
-!- ikks(n=ikks@ has joined #tryton12:57
CIA-5htgoebel roundup * #1357/ERROR:dispatcher:Exception calling method execute on wizard ir.module.module.config: [invalid] can't reproduce it :-(13:23
-!- yangoon( has joined #tryton13:23
-!- woakas(n=woakas@ has joined #tryton13:40
CIA-5C?dric Krier <> default * 2146:f7d3fd252538 trytond/trytond/model/
CIA-5Fix test on _search_domain_active for XML-RPC for issue138014:13
CIA-5Tuple in XML-RPC are list14:13
CIA-5C?dric Krier <> default * 2147:e68d1e920d6d trytond/CHANGELOG: merge14:13
CIA-5C?dric Krier <> default * 634:552208701792 stock/
CIA-5Fix test on to_location_warehouse for XML-RPC for issue138014:17
CIA-5Tuple in XML-RPC are list14:17
CIA-5ced roundup * #1380/xml-rpc search_read anormal behavior with [('active', '=', False)]: [resolved] Fix with changeset f7d3fd252538 and changeset 55220870179214:19
-!- juanfer(n=juanfer@ has joined #tryton15:25
-!- sharoon(n=sharoont@ has joined #tryton15:34
-!- rednul_( has joined #tryton16:28
CIA-5ced roundup * #1386/Module company doesn't install together with party_bank_de:16:41
CIA-5[invalid] It works here.16:41
CIA-5It seems that some custom module (even if not installed) add a domain on a field in the wrong way.16:41
CIA-5C?dric Krier <> default * 1491:565b82b46c33 tryton/tryton/gui/window/view_form/screen/
CIA-5Test range before use it17:37
CIA-5get_visible_range can resturn None17:37
-!- fil_( has joined #tryton17:41
-!- johbo_(n=joh@ has joined #tryton17:42
CIA-5C?dric Krier <> default * 1492:927d887a8ba4 tryton/tryton/gui/window/view_form/view/ Scroll before setting the cursor17:55
CIA-5C?dric Krier <> default * 1493:9ea37f6d2ddf tryton/tryton/gui/window/ Set the cursor after reload17:55
cedkessich_: will you work on 181154?18:23
cedkessich_: or do you let me integret the patch?18:25
-!- pepeu(n=manuel@ has joined #tryton19:27
-!- vengfulsquirrel( has joined #tryton19:32
essich_cedk: did we agree on the coding still yet?19:36
essich_cedk: esp. usage of __all__?19:38
cedkessich_: no19:46
essich_cedk: so my work was for nothing. going to delete the reivew-set.19:48
essich_cedk: if you change your mind, let me know19:55
pantheraas of know, we have all of tryton1.4 in squeeze, yay :)19:56
cedkessich_: it is sad that you don't want to contribute because we avoid the usage of "import *"19:59
cedkessich_: and we are not the only one to do so:
cedkpanthera: great19:59
cedkpanthera: do you want to make a news on the website?20:00
essich_cedk: Tis styleguid is not relevant for Tryton, see
essich_cedk: you must not set up new rule every few weeks just because you had a new idea20:11
cedkessich_: I don't say that !20:12
essich_cedk: if "from M import *" is the only matter, we can agree.20:13
essich_cedk: but you alos dislike __all__20:14
cedkessich_: it is unuseful if you don't import *20:15
-!- LucaSub1( has joined #tryton20:15
-!- LucaSub1( has left #tryton20:15
cedkessich_: and it is revelant only for library not inside software20:15
essich_cedk: __all__ documnts the interface of the module. and documentation is always good.20:16
essich_cedk: __all__ is only a plus, no negative effects.20:17
essich_cedk: __all__ is recomenced by PEP8 (which says the underscore prefix is the old way)20:17
cedkessich_: we don't care20:18
essich_cedk: *you* do not care. I do20:18
cedkessich_: we don't use it in Tryton because it is not a library20:18
essich_cedk: this is q question of code quality20:18
cedkessich_: no20:18
essich_cedk: why no?20:18
cedkther is no use20:18
essich_cedk: it is documenting the interface, this is a use20:19
cedkessich_: there is no interface as it is not a library20:19
essich_cedk: every module is an interface, per definitionem20:19
cedkessich_: no20:20
cedkessich_: we don't use it like that20:20
essich_cedk: it does not matter how *you* are using it. it's still true20:21
vengfulsquirrelHa wow, this is an unfruitful argument.20:21
cedkessich_: we will not include different codeing style in Tryton20:22
essich_cedk: so I can not contribute any more? :-((20:22
vengfulsquirrelI think tryton does need more documentation but I don't think __all__ is going to help the parts that are the problem.20:22
cedk__all__ doesn't give any information20:23
cedkit is just to follow version change20:23
vengfulsquirrelWell it gives some, but it seems like if the module/interface is really important it would be better documented with a .rst file.20:23
cedkit is useful for base module but not for specific software like Tryton20:23
cedkvengfulsquirrel: __all__ just gives what will be imported by import *, but as we don't use it, it gives any information20:24
essich_cedk: I'm not going to wast my time discussion teh usefuillness of __all__20:25
essich_cedk: it seams as if you are still thinking of a small, irrelevante framework.20:26
cedkessich_: absolutly not20:26
essich_cedk: but if tryton gains momentum, there will be a lot of developers which need the information20:26
vengfulsquirrelcedk: I just meant it COULD be used to document which parts of a module are meant for public usage but I agree/think that in tryton's case rarely is that kind of documentation needed.20:27
essich_cedk: so you arguments about "not an interface" are irrelevant20:27
cedkessich_: we use the "old way"20:27
vengfulsquirrelTryton needs overview documentation and usage documentation, there are not that many interfaces that need to be documented for the developer.20:27
cedkvengfulsquirrel: +120:27
essich_cedk: you are using the "old way" but I should chance to "while true"? You are so labile20:28
cedkand once again __all__ gives just a list, it doesn't document more20:28
essich_cedk: cedk: __all__ documents the public interface20:29
vengfulsquirrelPEP8 does not say you should use __all__ it just says IF you use import * then you should use __all__ over the old way of prefixing.  Where "old way" means the old way of supporting import *.  I don't think tryton uses import * right?20:29
cedkessich_: without _ is public with _ is not20:29
cedkvengfulsquirrel: right20:30
vengfulsquirrelha we could have documented something already20:30
essich_cedk: this is old style, PEP 8 recomends __all__20:30
bechamel2all: the python doc says : "Package authors may also decide not to support it, if they don’t see a use for importing * from their package", and everywhere we do import like "from tryton.model import modelsql" so for me __all __ is not needed20:30
vengfulsquirrelessich_: it only recomends __all__ IF you use import *20:30
cedkessich_: why is it old way ? I don't follow rule without explaination20:30
essich_vengfulsquirrel: this is the result of cedk's nit-picking :-(20:30
cedkessich_: import * was always forbiden20:31
essich_cedk: you forbate in on January, 13th. only 6 days ago.20:31
cedkessich_: It was missing,20:32
essich_bechamel: if we are not using "forom M import *", there is no need for prefix underscores, too. But tryton uses pefix underscore widespread.20:33
cedkessich_: look at the code there was no import *20:33
cedkessich_: it is useful even without import *20:33
cedkessich_: it is the "documentation"20:33
bechamelessich_: it just help people to know what is important and what is not20:34
cedkessich_: STOP20:34
essich_cedk: as you said: there is no from M import *20:35
bechamelactually I never saw myself using import *20:35
cedkessich_: except in __init__.py20:35
essich_cedk: don't be afraid, there are all ;-)20:35
essich_bechamel: shouldn't we help developers? or should we make their live harder as it is necessary?20:36
essich_cedk: report/ is designed for "from M import *"20:37
cedkessich_: why do you say that?20:38
essich_cedk: report/ has "from report import *", thus report/ has to be designe to be used this way.20:39
cedkessich_: it is design to be use like this: from import Report20:40
essich_cedk: you do not use it this way. See report/__init__.py20:40
essich_cedk: *you* are using "from M import *", but I must not design a moduel to be used liek this?20:41
essich_cedk: you are contraticting yoursefl20:41
cedkessich_: this is the last time I said: "Not for"20:41
cedkI stop there, I don'T want to spend more time on this20:42
essich_cedk: we are not taking about We are talking abour report.py20:42
essich_cedk: is the one which gets imported be 'from report import *'20:42
essich_cedk: is the one thar should define __all__20:42
vengfulsquirrelessich_: I see what you mean but its still not worth changing that since developers are not ever going to or be expected to use import *.20:44
vengfulsquirrelLet's talk about how to do .search() on a many2many which I think is more productive and I still have not resolved.20:44
essich_vengfulsquirrel: depend on where developers are coming from20:45
-!- essich_( has left #tryton20:45
-!- essich_( has joined #tryton20:45
bechamelfor everybody information: essich removed all his issues on codereview, so he only succeed to waist our (and his) time :(20:46
vengfulsquirrelessich_: The only case where import * would be used would be in files.  Any other case should use absolute imports.20:46
essich_vengfulsquirrel: experienced python programmers are quite used to this20:46
essich_bechamel: Thei are only "private"20:47
vengfulsquirrel essich_: Quite used to what?20:47
essich_vengfulsquirrel: maybe. But this is no argument not *not* using __all__20:47
essich_vengfulsquirrel: maybe. quite used to look at the interface (as defined in __all__) or to use "from M import *"20:48
essich_vengfulsquirrel: why sould * be allowed for absolut imports?20:49
vengfulsquirrelessich_: It just isn't worth the effort to change it, I tried using __all__ in the past and I found it hard to maintain.  Especially since I never use import *.  Experienced python developers do not use import *, as far as I know it is always frowned upon.20:49
essich_bechamel: this fruitless discussion about nit-picking issues and removing stuff which helps is wasting my time :-(20:50
essich_vengfulsquirrel: I use it in all my modules (except for quick'n'dirty)20:51
essich_vengfulsquirrel: And I seldom use "from M import *", but I design for others to use it that way if they like.20:52
vengfulsquirrelessich_: Real documentation is better and more easily maintained than __all__, I guess that is the point I tried to make earlier.  The maintenance required to keep __all__ relevant makes it not worth it.20:53
vengfulsquirrelessich_: Experienced python programmers would expect free text documentation that explains how to use the interface and the important components(that would probably span many python modules).  They would never open the file and check whats in __all__.20:54
essich_vengfulsquirrel: Documentation on a seperate place will not stay current, even in the short run. This is why Python has doc-strings20:55
-!- essich_( has left #tryton20:55
-!- essich_( has joined #tryton20:55
essich_vengfulsquirrell: I expcet usinf pydoc <module>20:56
vengfulsquirrelessich_: __all__ is just hard to maintain because no one uses it, and you change a function/class way in a file and forgot to update __all__ , documentation in a separate place will at least be seen by more people that can tell that it has become outdated and they will try to update it20:57
vengfulsquirrelessich_: And tryton's real documentation problem now is that more usage documentation is needed that __all__ would not help with.20:58
vengfulsquirrelI almost always know what function/field/model/class I need I just don't know how I should use it to get what I want done.20:58
vengfulsquirrelessich_: I think instead of spending time running around change _function to function and putting function in __all__ we should focus that energy on more usage documentation because that is what developers will expect to find.20:59
essich_vengfulsquirrell: I'm not changing module, just to all __all__. I've written now modules whre I've put in _all__21:00
essich_vengfulsquirrell: I'm not changing modules, just to add __all__ everywhere. I've written now modules whre I've put in _all__21:00
essich_vengfulsquirrell: but cedk does not like it. I should remove it.21:01
vengfulsquirrelWell to maintain consistency with the rest of the project __all__ must be removed from your new modules or added to all the tryton modules.21:01
vengfulsquirrelSo I don't think its worth it to add __all__ to the entire project, when it doesn't solve the/a problem that tryton is having it only adds more maintenance.21:01
essich_vengfulsquirrell: not necessary. The rule could be to add it whenever a module interfac changes.21:01
vengfulsquirrelNo I think it must be all or nothing in order to maintain consistency.21:02
essich_vengfulsquirrell: leaving it away does not help, too. And discussion it only wastes time21:02
essich_vengfulsquirrell: regarding documentation: a basic problem is that cedk is not even adding docstrings to new code21:06
vengfulsquirrelWe only need docstrings in certain places, I think the bigger problem is usage documentation.21:08
CIA-5htgoebel roundup * #1388/search_rec_name fails for ir.lang:21:14
CIA-5[new] Cédric Krier schrieb:21:14
CIA-5> > This will work as search_rec_name search first for the code:21:14
CIA-5> >21:14
CIA-5> > "Name","lang"21:14
CIA-5> > "Gustav Gast","en_U ...21:14
essich_vengfulsquirrell: writing a complete documentation is much, much work. Docstrings are much easier to write and halp other to understand the code.21:15
essich_vengfulsquirrell: Nevertheless, if you are taking about documentation of concepts, I fully agree :-)21:15
bechamelessich_: to be clear: cedk is nit-picking since day one, and yes it's painful, and I'm/was the first to suffer from it, but at the end the codebase is really homogeneous and it's a big plus IMO21:17
bechamelessich_: and for me it's such an advantage that I'm ready to put my ego in my pocket. I know from experience that if I was right and cedk didn't agree, it's a matter of time for the issue to get back in our discussion and get changed my way. I only need to be patient21:19
essich_bechamel: a homogeneous code and it's a big plus. shure.21:34
essich_bechamel: but cedk's nit-picking is *de-motivating*, because he is not honoring what ha be done with not even a word. HGe is just nit-picking on unimportand stuff.21:35
essich_bechamel: and his coding style is odd in some places (like the header comments in each file)21:36
bechamelessich_: it works both ways: if it's unimportant, why you wanna fight so hard for it ?21:36
essich_bechamel: because it's unimportant :-) Why is cedk riding about it instead of just taking it?21:39
essich_bechamel: even writing a comment "remove ..." in rietveld is a wast of time21:39
essich_bechamel: (I know I'm iding on it hre, too)21:40
vengfulsquirrelWell probably because of the inconsistent code base.21:40
bechamelessich_: I agree that a "nice work" or something like that on reitveld comment would be better, but now you know it, comments (even negatives) means that people are interested by your work21:41
vengfulsquirrelCan someone explain to me how to search a many2many like this:
cedkvengfulsquirrel: the best is to search on res.user-res.group21:50
cedkvengfulsquirrel: but ('groups', =, group_id) should work21:51
cedkvengfulsquirrel: or ('groups', 'in', [group_id])21:52
CIA-5htgoebel roundup * #1389/wrong message if importing many2many fails: [new] When importing a many2many field, the filed get split at the comma. If the lookup fails, the error/warning message shows the cell content, n ...21:53
essich_bechamel: in your team you ar obviously the one responsibe for social competence ;-)22:03
-!- FWiesing( has joined #tryton22:21
CIA-5vengfulsquirrel roundup * #1390/Cannot receive customer return shipment.: [new] This traceback occurs when I am creating a manual return shipment with one entry and have clicked the "Received" button. Should return ship ...22:30
CIA-5vengfulsquirrel roundup * #1390/Cannot receive customer return shipment.: [chatting] Oops, forgot to add people other than myself.22:30
-!- paepke( has joined #tryton23:02
CIA-5ced roundup * #1388/search_rec_name fails for ir.lang: [invalid] There is no search_rec_name on ir.lang23:50
CIA-5ced roundup * #1390/Cannot receive customer return shipment.: [closed] Duplicate issue137923:53

Generated by 2.11.0 by Marius Gedminas - find it at!