CarlosRuiz: full meeting

2013-02-13T19:20:46Z

full meeting

New page

⇐
[[IDempiere|Table of Contents]] |
[[IDempiere/Full Meeting Minutes|Full Meeting Minutes]] |
Full Meeting 2013-02-13


'''''CarlosRuiz''''': Good Morning 
'''''tbayen''''': Hi 
'''''nmicoud''''': Bonjour 
'''''fcojvanninir''''': hello 
'''''Deepak''''': Good Evening 
'''''red1''''': ha! I thought it's Thursday. 
'''''tbayen''''': red1, you have not run away fast enough! 
'''''red1''''': Peace! 
'''''red1''''': i can't I m on Android. 
'''''red1''''': thanks to Heng Sin i made the plugins more compact now. 
'''''tbayen''''': Is it true that I can't access the table import from Swing? 
'''''red1''''': i just posted some notes in my forum if anyone wants to know more. 
'''''CarlosRuiz''''': what is table import? the import button? 
'''''tbayen''''': Yes, the import button that is in zk in the toolbar at the very right side (near export). I can call export via menu in Swing but I can't find import. 
'''''CarlosRuiz''''': yes, I think it has not been developed for swing - must be something easy to port 
'''''tbayen''''': ok :-( 
'''''hengsin''''': @carlos, any deployment for demo.idempiere.com today ? 
'''''CarlosRuiz''''': I'll do one - saw you committed lot of things 
'''''tbayen''''': While doing IDEMPIERE-607 (restrict Login languages for Swing) I realized that "English (US)" is not a Standard Login Language. Would'nt it be better to change that? How is this resolved in zk? Is it only spanish when fresh installed?!? 
'''''CarlosRuiz''''': tbayen, to port the import button I think is just to port the class org.adempiere.ui.zk/WEB-INF/src/org/adempiere/webui/panel/action/FileImportAction.java 
'''''tbayen''''': CarlosRuiz, yes, it would not be a big issue. I hate this because it remembers me that the best solution would we to implement the configurable toolbar (IDEMPIERE-234) in Swing. 
'''''tbayen''''': ;-) 
'''''tbayen''''': As ever in life it is much work to do it the right way. 
'''''CarlosRuiz''''': hengsin, WDYT about last comments on http://jira.idempiere.com/browse/IDEMPIERE-607 
'''''CarlosRuiz''''': I think maybe is clearer and easier to show just the languages marked as login - actually we're showing login and base 
'''''tbayen''''': I could do that in Swing too. Which way is the better? 
'''''nmicoud''''': but if there are no login language, we should show at least one ? 
'''''tbayen''''': nmicoud, if we do that it would be the same effort to do it like before in zk. Is the a usage case where anyone does not want the base language shown? 
'''''tbayen''''': But if someone screws up the database by deactivating all loginm languages he could even deactivate the base language. 
'''''CarlosRuiz''''': well - related news - Juliana is working on this one http://jira.idempiere.com/browse/IDEMPIERE-136 
'''''CarlosRuiz''''': and that will make the base configurable, not always en_US 
'''''nmicoud''''': correct tbayen 
'''''nmicoud''''': but we should avoid having a language box empty 
'''''CarlosRuiz''''': so - IDEMPIERE-136 would solve that - show login languages and base - and base can be configurable (not always en_US) 
'''''tbayen''''': So I will make the Swing login behave like the zk does today and this will work even better when IDEMPIERE-136 is ready. 
'''''nmicoud''''': ok 
'''''CarlosRuiz''''': banym, AYT? can I reboot demo.idempiere.com? I just upgraded with apt-get 
'''''tbayen''''': Thanks nmicoud for pointing to that at first. ;-) 
'''''nmicoud''''': héhé :) you're welcome 
'''''tbayen''''': Has someone looked into my last comment to IDEMPIERE-604? Does this code break anything? 
'''''CarlosRuiz''''': Thomas the patch looks wrong 
'''''CarlosRuiz''''': you're setting the current row to 0 or 1 - before it was setting it to count-1 
'''''tbayen''''': :-( It works for me but I do not know why this code was there before. That's why I am a bit afraid breaking anything. 
'''''tbayen''''': "count-1" meant: "the last record of the list or -1 if the list is empty". "0:-1" means: "always 0 or -1 if the list is empty" 
'''''tbayen''''': so I activate the first and not the last record 
'''''CarlosRuiz''''': but (m_currentRow != verifyRow(m_currentRow)) doesn't mean the list is empty 
'''''CarlosRuiz''''': it means the required row is still not loaded - so as I see it tries to set in the last loaded 
'''''CarlosRuiz''''': but my assumptions are just reading the code - have not debugged or tested 
'''''CarlosRuiz''''': hengsin - demo is updated and ready 
'''''tbayen''''': getRowCount()>0 does mean the list is (not) empty. 
'''''CarlosRuiz''''': ah, so you're setting current row on first 
'''''CarlosRuiz''''': instead of last - that's the change 
'''''tbayen''''': Yes, that was my purpose. Sorry when I did not explain this well. 
'''''CarlosRuiz''''': ok, understood, will need to test the case you described on the ticket 
'''''tbayen''''': The question is: Which other code (or user) relies on the fact that the last line is activated here? (Regarding the fact that in my use case "search" the active column is changed again later) 
'''''tbayen''''': You can test better if you set the timeout to 10000 like I described. Or you need some tenthousand entries. 
'''''CarlosRuiz''''': yep - that's my concern - gridTab.getCurrentRow() is referenced in 47 calls 
'''''JanThielemann''''': ll 
'''''buildmaster''''': Project iDempiere build #757: SUCCESS in 5 min 57 sec: http://jenkins.idempiere.com/job/iDempiere/757/ 
'''''buildmaster''''': globalqss: IDEMPIERE-166 Rebranding of logo and product name / fix demo server not showing login and vendor info 
'''''tbayen''''': nmicoud, have you review your last issue with IDEMPIERE-90 and Carlos patch? Can we close this one as resolved? 
'''''nmicoud''''': yes, it was resolved, but there are stille something to do : add the padlock icon for zk 
'''''nmicoud''''': i look at it but i did not manage to do it 
'''''nmicoud''''': seems that table/table direct field can't have img 
'''''tbayen''''': OK, so we will leave it open. 
'''''nmicoud''''': i think it's better 
'''''nmicoud''''': as it is not totally completed 
'''''hengsin''''': nmicoud, you will need to create a new composite component for that. 
'''''nmicoud''''': that is why i'm afraid. Have no idea on how to do it :( 
'''''hengsin''''': tbayen, I need to test the scenario you posted but I've a feeling that your patch could miss the root cause of the issue. 
'''''tbayen''''': hengsin, Yes. I thought the same... In my usecase this function is called by a refresh event. I don't think that in a refresh there should be made a decision where the active column is. My thought is that my patch is the right one (or we even have to set the active column fixed to -1) and we should fix the other places where code relies on the refreshing code to set the right column. WDYT? 
'''''tbayen''''': But I have not nearly a deep understanding as you... 
'''''tbayen''''': gtg, bye and thanks! 
'''''tbayen''''': cul8er 
'''''nmicoud''''': bye 
'''''buildmaster''''': Project iDempiere build #758: SUCCESS in 5 min 37 sec: http://jenkins.idempiere.com/job/iDempiere/758/ 
'''''buildmaster''''': globalqss: IDEMPIERE-77 Start on the left, Cancel on the right / fix the link to the wiki login help page - make it language friendly 
'''''CarlosRuiz''''': bye 
'''''a42niem''''': hi all 
'''''a42niem''''': i am a bit frustrated currently as i seem not to be able to materialize the swing repo without errors 
'''''a42niem''''': i am using indigo. is it better to use juno perhaps? 
'''''a42niem''''': any other recommendation? 
'''''CarlosRuiz''''': I'm using Version: 3.7.2.v20120207 
'''''CarlosRuiz''''': I think Juno has problemas 
'''''hengsin''''': no, indigo is better. 
'''''a42niem''''': hm, ok, so stick to indigo 
'''''a42niem''''': my problem was that jsp was available in version 2.2 iirc but was required as 2.1 somewhere 
'''''a42niem''''': and i am still not experienced enough to know wher to change some entry :( 
'''''a42niem''''': so i deleted the target platform and created it new and materialized again and got hundreds of errors 
'''''red1''''': during materialize there are steps. first it shows what isn't resolved. did u see? 
'''''red1''''': also about spam in wiki. Can we block like u did in ADempiere? 
'''''a42niem''''': i am currently trying again with a freshly cloned repo. still working 
'''''a42niem''''': concerning spam, i think so. it was using a specific extension. i have to look it up 
'''''a42niem''''': argh. this time it worked. there is only one java error. thanks for all the good vibrations :) 
'''''red1''''': Are you using 64 not 
'''''a42niem''''': linux 64bit 
'''''CarlosRuiz''''': red1 - the spam is on acceptable levels I think - I would prefer allow users to self-register without needing external permissions 
'''''red1''''': 64 bit Eclipse. 
'''''red1''''': i had trouble too and switch around alot 
'''''a42niem''''': maybe it is a bit old? eclipse.buildId=M20120208-0800 
'''''red1''''': Presently using 64 bit Indigo on Mac 
'''''banym''''': CarlosRuiz, yes feel free to reboot the server 
'''''a42niem''''': i think it is 64bit version but it does not show me that explicitely anywhere 
'''''banym''''': hi red1, how are you? Did you had some time to sign the letter for Caroline? 
'''''red1''''': Also had trouble with Java and i suspect maybe related. 
'''''red1''''': i settled for Java 6 
'''''a42niem''''': i use openjdk 6 
'''''red1''''': banym: I cannot sign anything that has ADempiere on it. There are also other matters but i discussed with and leaving it to TBayen 
'''''CarlosRuiz''''': guys - gtg - thanks for attending the meeting

IDempiere/FullMeeting20130213 - Revision history

CarlosRuiz: full meeting