IRC logs of #tryton for Wednesday, 2008-08-06

chat.freenode.net #tryton log beginning Wed Aug 6 00:00:01 CEST 2008
-!- ChanServ(ChanServ@services.) has joined #tryton00:27
-!- ChanServ(ChanServ@services.) has joined #tryton00:38
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 807:b0f4ea618346 trytond/trytond/ (ir/translation.xml osv/orm.py):00:50
CIA-54tryton: Remove _error_messages in ORM object and put the translation in00:50
CIA-54tryton: translation.xml00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 808:a1e4e4b8ab27 trytond/trytond/ir/translation.py: Fix export translation for unicode string00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 809:a446034a0c53 trytond/trytond/ir/translation.py: Add test for ttype error and raise an exception for unknown type00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 810:8e7cf988d2be trytond/trytond/ (ir/fr_FR.csv res/fr_FR.csv webdav/fr_FR.csv): Update fr_FR translation00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 171:91ed02ac31ea account/ (__tryton__.py fr_FR.csv): Add fr_FR translation00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 91:d2008637a283 account_invoice/ (__tryton__.py fr_FR.csv): Add fr_FR translation00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 44:1c182bfaf52f company/ (__tryton__.py fr_FR.csv): Add fr_FR translation00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 11:680d7efc34c2 currency/ (__tryton__.py fr_FR.csv): Add fr_FR translation00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 66:8388184fd3d6 product/ (__tryton__.py fr_FR.csv): Add fr_FR translation00:50
CIA-54tryton: C?dric Krier <ced@b2ck.com> default * 118:5c51502203fb relationship/fr_FR.csv: Update fr_FR translation00:50
-!- betamax_(i=betamax@gateway/tor/x-fb6f1fbcdc6240a3) has joined #tryton01:27
-!- Timitos(n=Timitos@88.217.184.172) has joined #tryton07:46
-!- Gedd(n=ged@77.109.115.229) has joined #tryton09:04
-!- gadaga(n=gael@sednaco19320-gw.clients.easynet.fr) has joined #tryton09:10
gadagahi09:10
udonogadaga: hi09:32
gadagaudono: hi09:32
-!- markusleist(n=markus@212.14.79.210) has joined #tryton09:39
udonohello, someone knows how I can use the translation mechanism of tryton in the reports? I like to see the labels like Quantity, Description, amount,.. in the native language of the Party...10:22
udonoIn the reports are just the contents translated, but not the labels/headers...10:23
udonoanother question is, where is the invoice header defined? Its neither in the invoice template, nor in the company settings.10:24
-!- kultviech(n=kultviec@p54973F1D.dip.t-dialin.net) has joined #tryton10:34
-!- essich(n=essich@p4FCFAF67.dip0.t-ipconnect.de) has joined #tryton10:36
essichhi10:36
essichjust looking in after some weeks for a few minutes :-)10:37
CIA-54tryton: Timitos roundup * #227/translation of client does not work under windows: its a german xp pro. everything is german. so i think german language should be installed.11:12
gadagaWhy in product.category when we do a search, the result return the complete_name and not the name?12:28
gadagano a search a read ;)12:28
-!- nicoe(n=nicoe@ip-80-236-206-69.dsl.scarlet.be) has joined #tryton12:47
-!- kultviech(n=kultviec@p54973F1D.dip.t-dialin.net) has joined #tryton13:24
-!- gadaga(n=gael@sednaco19320-gw.clients.easynet.fr) has joined #tryton13:40
udonogadaga: Both are shown in the resultlist...14:37
-!- lionel_(n=lionel@pic33-1-87-88-179-8.dsl.club-internet.fr) has joined #tryton14:37
gadagaudono: thanks14:38
-!- Leyoda(n=Leyoda@pic33-1-87-88-179-8.dsl.club-internet.fr) has joined #tryton15:05
-!- nicoe(n=nicoe@ip-80-236-206-69.dsl.scarlet.be) has joined #tryton15:06
nicoeudono: You'll be happy to know that I successfully created a pivot table (called data pilot table) in openoffice calc15:07
udononicoe: The greatest man in universe :-)15:37
udononicoe: how did you do it?15:37
nicoeIt was complicated15:38
nicoeI'm commiting the changes to the hg right now15:38
nicoeBut it involved some customizing of the ods' content.xml file by hand15:39
nicoeAnd I use a tab as a data repository only (no formula, just headers and data)15:40
-!- Gedd(n=ged@77.109.115.229) has joined #tryton15:40
nicoeudono: ok changes are pushed.15:42
nicoeAnother interesting thing is that now the table data should follow the data in placeholders even in the odt files.15:43
udononicoe: is it in the relatorio sources?15:44
nicoeudono: in relatorio's mercurial repository15:44
udononicoe: where to find?15:44
udononicoe: got it15:45
nicoehttp://hg.openhex.org/relatorio BTW we also have a trac instance to track bugs, add information in http://relatorio.openhex.org15:45
-!- gael_(n=gael@sednaco19320-gw.clients.easynet.fr) has joined #tryton15:48
udononicoe: is the change already as an egg available?16:03
udononicoe: or as mercurial repository?16:04
nicoeNot yet, I'll make a release soon16:04
nicoeBut it is already present in the mercurial repository16:04
nicoeudono: I released an egg: http://pypi.python.org/pypi/relatorio16:34
udononicoe: You are my man! Thanks a lot. So I just update my old eggs from eastern ,-)16:35
udononicoe: you dont change the version number? No prob, I just delete the old relatorio...16:36
udononicoe: easy install tells me its relatorio-0.2.0-py2.5.egg16:37
nicoeudono: What needs to be done is a way to create placeholder in the oocalc always going in the content.xml file is difficult for "normal" people :)16:37
nicoeon pypi it's 0.3.016:37
udononicoe: ok16:37
udononicoe: I first test it with your testscrips...16:42
nicoeudono: ok16:42
udononicoe: hmm, cant get the test called...16:45
nicoeudono: what's happening ?16:46
udononicoe: I installed via egg and easyinstall16:46
udononicoe: all fine16:46
udononicoe: the testscripts are in the repos, so I pull them16:47
nicoeudono: hu ok16:47
udononicoe: than I change into the relatorio folder:16:48
udononicoe:... and call python relatorio/tests/test_api.py16:48
udononicoe: but it raises an error: ImportError: No module named reporting16:49
udononicoe: maybe the eggs are broken :-)16:49
nicoeudono: there is one failed test there. But without any impact on the opendocument plugin16:49
nicoeudono: can you paste somewhere the complete traceback ?16:50
udononicoe:  http://paste.pocoo.org/show/81395/16:51
nicoeTry running the test from the relatorio directory16:53
nicoeBTW those are nosetests tests.16:53
nicoeWhat you are looking for is in the example directory16:53
udononicoe: nose I installed...16:53
udononicoe: the same result from relatorio folder...16:54
udononicoe: example works fine :-)16:55
nicoeDo you also have the pivot table ?16:55
nicoeYou need to refresh it16:55
nicoeit's strange that nose does not work16:55
udonoyeah, all there16:56
udononicoe: I show the Template of the ods, Thats the syntax I like to see. Great job!16:58
udononicoe: you name the table pivot, but its just a simple odt table isnt it?!16:59
udonosorrx odt=ods16:59
udononicoe what we have to manipulate inside the ods XML?17:00
nicoeNo, it's generated from the content of the data table17:00
udono:-)17:00
udononicoe: ok, I just recognize the data table...17:00
nicoeThere is no way to create a <text:placeholder> tag in the interface of openoffice for Spreadsheet17:00
nicoeSo I inserted them by hand17:01
nicoeI think the easyest way to create a pivot table is: (a) create your data table with the name of the fields (b) create the pivot table (c) edit content.xml to use placeholders17:02
nicoeyou could probably use an intermediate table to do computation on the data17:03
udononicoe: great! the pivot is running fine.17:03
udononicoe: is the manipulation needed just for parsing the data table? or on the pivot creation?17:04
nicoeThe creation of the pivot.ods file was a mess, but I would say that it is only need for the parsing data. I'll try to create a more complicated example without doing anything else but adding placeholder tags17:05
udononicoe: I will ask the openoffice guys if they have a solution.17:06
udononicoe: I c you put the placeholder tag inside the text:p <text:p><text:placeholder>&lt;l.item.price&gt;</text:placeholder></text:p>17:07
nicoeudono: please do :).17:07
nicoeudono, yes I did minimal changes on the generated content17:08
udononicoe: and you parse the ods the same like the odt?!17:08
nicoeit's the same norm17:08
udononicoe: ok, I will ask them in some minutes...17:08
nicoeDoing it the same way allows you to use the format number stuff from calc in writer17:09
nicoeeg you do not have to use a py:if just to highlight negative numbers in red. OO will do it for you. even in odt document17:10
udonoACTION can't get enough from dreaming ...17:14
udononicoe: great ideas!17:14
nicoeIn fact it  was a nice plus ... I realized it would work only after having done it17:15
udononicoe: there is no reaction untill now... I requested the following:18:17
udonoHello, we have a problem with parsing ODS Content.xml files. We are working on relatorio (http://relatorio.openhex.org/) an open source template parser for Open-Office dokuments. With this parser it's possible to replace special tags in ODT files with Python variable values. We like to extend the functionality of our parser to ODS documents. But there are some Problems.How we do in ODT: The user create his document via OpenOffice. Spe18:17
udonoield function to be inserted. This function generates the following code in the content.xml:<text:p text:style-name="P4"><text:placeholder text:placeholder-type="text">My Variable to be parsed</text:placeholder></text:p>The same we like to do in ODS. But there is no eqivalent function like this. The ODS cell content looks like:<table:table-cell office:value-type="float"><text:p>My Variable to be parsed</text:p></table:table-cell>But 18:17
udonocontent is just normal text or a variable to be parsed.If we could put a <text:placeholder> tag inside, we even do not neet to rewrite our parser. In a testcase with manualy inserting this tag we can parse ods without problems.Is there a way to put a tag inside the <text:p></text:p> from userside?Where I can start to solve this problem?18:17
udonoHey nicoe, have a soulution...18:34
udononicoe: if you insert a hyperlink with an own protocol like "relatorio://<and the tags inside />" than you get the structure:<table:table-cell office:value-type="string"><text:p><text:a xlink:href="relatorio://&lt;test&gt;&lt;/test&gt;">relatorio://&lt;test&gt;&lt;/test&gt;</text:a></text:p></table:table-cell>18:44
udononicoe: a comparisation between ODT and ODS results in the same structure: http://paste.pocoo.org/show/81403/18:56
udononicoe: so I would propose to use Hyperlinks with a private Protokoll to insert the genshi structures and no longer ODT only placeholders...18:59
udononicoe: The Hyperlink solution is valid ODF.19:01
udononicoe: We can hide complex structures behind a simple Name, if we use the xlink:href="relatorio://&lt;test&gt;&lt;/test&gt;" expression for the parsing things and the content between for the Name...19:02
nicoeudono I'm back19:04
udononicoe: welcome I received a good ides...19:04
udonoidea19:04
nicoeI did not understand your last message could you show me the url you'd like to create (but withtout escaping it for the sake of clarity)19:05
udonorelatorio://<test></test>19:06
udonoits a private protokol and genshi after this...19:06
nicoeIn the url idee we do not need the <> ... They are inserted by openoffice but I dont like them19:06
nicoeWe could have relatorio://(a + b)19:07
nicoerelatorio://for each="foo in bar"19:07
udononicoe: this looks not bad19:08
nicoeI don't know ...19:08
nicoeI'll rewrite the example documents and see what it looks like19:09
nicoeOf course if we go this way, cedk will have to rewrite their documents19:10
udononicoe: did you recognize the sentence with xlink:href?19:10
nicoeOr I could support the old syntax and issue warning19:10
nicoeudono it's <test></test>19:11
udononicoe: It could be possible to hide complex structure behind a simple name...19:11
nicoeudonon ?19:11
udononicoe: The user see the simple name, and behind this is the complex hyperlink...19:12
udononicoe: For me the old syntax is only good for ODT, but unusable for ods, because Openoffice filter them out after saving...19:13
nicoein your pasted code the href contains the same thing as the a19:13
udononicoe: yes, but it mustn't be19:13
nicoeyou're right19:13
udononicoe: better to use the xlink:href attribute.19:13
nicoeWe should use the xlink:href and let the developper the opportunity to add "comments" in the TEXT node19:14
nicoeThat seems quite good19:14
nicoeWhere there any other ideas on the opeoffice channel ?19:15
udononicoe: I tested the href entry: I cant see an end, It slurps more than 4500 characters...19:15
udononicoe: no, the gui who helps me (seems a sun-boy) meant that the hyperlink is the only way to get valid odf, even for odt and ods19:17
nicoeIf guys at Sun says so ...19:17
nicoeThey know better :)19:17
nicoeI'll try that.19:18
udononicoe: One guy with sun in his name :-) http://www.jroller.com/erAck/ thats his blog19:18
udononicoe: He meant at the end of our conversation: thinking further: it might even be possible to create an extension that handles a private protocol during runtime [/me doesn't know how though]19:21
nicoeudono I suppose he meant an extension to OO19:21
udononicoe: yes19:22
udononicoe: I understand the same19:22
nicoeudono I don't feel like writing an OO extension (Java, something to install, etc)19:23
nicoeudono one day who knows19:23
udononicoe: me not, too. But we must not do this. The only thing not so nice is, that the Hyperlink is a single entryline...19:24
nicoeUsually stuffs replacing the code is just one line this does not seems like a problem to me19:25
udononicoe: ok19:26
nicoeIt seems more like a good thing in fact19:26
udononicoe: yes, maybe. Iam new to the genshi syntax, but in the past when we used python as the template language, it was not so easy to put complex structures into one single line...19:27
udononicoe: but you may be right19:28
nicoeoh ok I understand19:28
udononicoe: but now everything is going easier, thanks to relatorio...19:29
nicoeI think the more processing you put in the python code before rendering the template the better19:31
udononicoe: of course, even when using python as templating language, and eval="it" you get a security hole...19:34
nicoeIndeed19:35
udonoFWiesing: Hello20:50
FWiesingudono: hello!21:07
udonoFWiesing: how are you?21:07
FWiesingudono: all ok21:08
FWiesingudono: thx21:08
udonoFWiesing: great! I like to ask, if we can put the translations into the modules?21:08
FWiesingudono: please give me 5 minutes21:08
udonoFWiesing: You have 1021:08
FWiesingI think you could21:09
udono;-)21:09
FWiesing:-)21:09
FWiesingback21:19
FWiesingudono: If I update my trytond - can I start the trytond with the options -d <databasenam> and -u all ?21:36
TimitosFWiesing: the server will stop after the update. you need to start the server in normal mode after db update21:37
FWiesingudono: Ok - First step is to start trytond with the option -d <databasename> and then in restart in normal mode21:38
TimitosFWiesing: I am Timitos ;-) and yes you are right21:39
FWiesingthx :-)21:39
FWiesinggerman language - I think you can use the translation - but I'm sure - there are one or two things which needs an update ...21:41
FWiesinglast week I didn't could test very much21:41
Timitosi think we will find these points easier when using the translation21:41
TimitosFWiesing: me too. had to much other work.21:42
FWiesingthis month and september I'm in a terrible hurry and only in the evening available21:42
TimitosdonĀ“t worry. we have enough to do. i am sitting on account chart these days. i hope that it will be available in the end of august.21:44
FWiesingshould I send a databasedump?21:44
Timitosthis is not nessecary in the moment as i go a different way from yours. if i succeed i will be able to parse a account chart for AT on the fly21:45
FWiesingTimitos: you have time to take a look of my account-chart for Austria21:45
FWiesingOK21:45
FWiesinggirl-friend is calling - I need to go ... :-)21:49
FWiesingcu21:49
udonobye21:49
Timitoscu21:49
FWiesingcome back a later!21:50
udonogood night all, need sleep.22:11
Timitosgute n822:11

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