IDempiere/FullMeeting20121212

From WikiQSS
Revision as of 11:36, 12 December 2012 by CarlosRuiz (talk | contribs) (full meeting)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Table of Contents | Full Meeting Minutes | Full Meeting 2012-12-12

CarlosRuiz: Good Morning
nmicoud: Bonjour
nmicoud: Did you have time to review tickets 448, 451, 459, 475, 492 & 526 ?
CarlosRuiz: starting with 448 ...
nmicoud: ok :)
red1: meanwhile i posted the first ever p2 site for plugin feature that can be installed remotely https://sourceforge.net/projects/red1/files/p2/
CarlosRuiz: wow red1 - that's great!
red1: I be doing more plugins such as Assets maintenance, HR Payroll, LCO...
CarlosRuiz: you just started the idempiere marketplace
red1: Jenkins can post such p2 plugins and this will open up the market - yep we got the right word here - marketplace..
red1: the 2Pack attachment worked like magic.. i did not see the prompts as it was not set to FINE.. but it appeared
CarlosRuiz: yes - at the end usually you still need to run "role access update" and that's all
red1: the TANG dream is reached today
red1: 12.12.12 :D
CarlosRuiz: I think we would better automate calling the role access update too
nmicoud: in other word, it is now possible to click on a 'link' to automatically install new plugins ?
red1: well quite so
red1: everything is in the p2 site remotely online
red1: you need to fetch it via the installer i posted
red1: just take that and put in your binary home root
a42niem: hi all
red1: and run it
red1: that is all
CarlosRuiz: Hi Dirk
red1: Hola Dirk
red1: i set the remote link inside the installer shell script
CarlosRuiz: we're getting close to that Nicolas - a one-click installer
red1: i will give a bat for windows later
nmicoud: like DV would say : impressive, most impressive ;)
red1: i did not get the UI to work
red1: spent 3 weeks tackling that and failed and gave up two days ago
red1: and did not expect the headless way to be that quick to resolve.. thanks to Hengsin
red1: basically copied his work
CarlosRuiz: how did you try the UI part?
red1: I put in the p2.ui projects
CarlosRuiz: I'm thinking on a zk form - downloading an index of marketplace extensions from a provided URL
red1: but they cannot deploy
red1: some app ID conflict
red1: the p2 UI has its own org.eclipse.ui and menus ext
red1: i wanted to attach to the big window item "Main"
red1: as an extra item > Update
CarlosRuiz: nmicoud, who is DV?
nmicoud: Darth Vader ;)
CarlosRuiz: :-D
tbayen_: Hi together! :-) I want to begin the official installation page in the german wiki. Is there something like an official download for the idempiere official release package for this? https://bitbucket.org/idempiere/idempiere/downloads is empty.
red1: well, if the shell can be run as a link, then that is all. Provided it can still install when the app is alive
red1: let me finish a few more addons
red1: i am also thinking, if migration scripts can be in 2Pack form
red1: u can just drop 2Packs in
red1: after they are installed, you just uninstall to remove the plugin luggage
CarlosRuiz: Hi Thomas
red1: but that kind of 2Pack does not register completely as the normal 2Pack import
CarlosRuiz: the installers were uploaded to sourceforge
red1: so unpacking may not work yet, but i seen some of the code, if i can solve that later..
CarlosRuiz: http://sourceforge.net/projects/idempiere/files/
red1: i choose sourceforge too as they got mirrors all over and will not jam one site
tbayen_: CarlosRuiz, thanks!
red1: this also means if we can get Jenkins to send to SF also
buildmaster: Project iDempiere build #576: SUCCESS in 6 min 41 sec: http://jenkins.idempiere.com/job/iDempiere/576/
buildmaster: * globalqss: hg merge e579d7ef4fa5
buildmaster: * elainetan: IDEMPIERE-60 - Bizidium must not save credit card number
CarlosRuiz: nmicoud, reviewing CopyFromBankStmt
CarlosRuiz: line 88
nmicoud: i'm opening eclipse ; what's the problem ?
CarlosRuiz: you're comparing a "SELECT C_BankStatementLine_ID" against zero
CarlosRuiz: I think you're checking that record doesn't exist - in such case it must compare with -1
CarlosRuiz: or maybe you were trying a count(*)?
nmicoud: right
nmicoud: just checking if the payment is used on another bank statement (draft, complete or close)
nmicoud: ==0 should become <0 or ==-1
CarlosRuiz: ok
CarlosRuiz: now - on line 98
CarlosRuiz: no - I think that's fine
nmicoud: :))
CarlosRuiz: testing ...
CarlosRuiz: nmicoud,
CarlosRuiz: you just can copy lines from completed or closed statements (not reversed) right?
nmicoud: yes
nmicoud: and if there is a payment on a line, if this payment is not used elsewhere
CarlosRuiz: but payments can be copied from reversed
CarlosRuiz: is that right?
nmicoud: checking
nmicoud: yes
nmicoud: that's right
CarlosRuiz: ah
CarlosRuiz: voided bank statements clear the payment field
nmicoud: in order to be re-used elsewhere ?
CarlosRuiz: yes - so - in the end it cannot be copied :-)
nmicoud: hum...
nmicoud: i'm confused... i know i've tested it and succeeded to copy payment
CarlosRuiz: from a reversed?
nmicoud: did not remember
CarlosRuiz: maybe because of the == 0 on line 88
nmicoud: but actually, i don't see the interest to copy a line with a payment_id
nmicoud: probably :-/
CarlosRuiz: yes - you cannot cos the payment id is cleared
nmicoud: Should be : if (fromLine.getC_Payment_ID() > 0) { "do nothing" }
CarlosRuiz: committed into idempiere - your commit and then my peer review
nmicoud: ok
CarlosRuiz: thanks a lot for contributing that - checking next
nmicoud: fine
buildmaster: Project iDempiere build #577: SUCCESS in 6 min 50 sec: http://jenkins.idempiere.com/job/iDempiere/577/
buildmaster: * globalqss: IDEMPIERE-448 Add CopyFrom on Bank Statement / Peer review
buildmaster: * nmicoud: IDEMPIERE-448 - Add CopyFrom on Bank Statement
buildmaster: Project iDempiere build #578: SUCCESS in 6 min 29 sec: http://jenkins.idempiere.com/job/iDempiere/578/
buildmaster: * globalqss: hg merge f4f8ef565103
buildmaster: * nmicoud: Tree updates fails (swing mode) ; refer to https://groups.google.com/forum/#!topic/idempiere/ETF7fymVZRM
CarlosRuiz: nmicoud, checking IDEMPIERE-492
nmicoud: i've done it in a very ugly way
CarlosRuiz: :-) yes
CarlosRuiz: seems strange when you move from left to right
CarlosRuiz: but in my tests - the changes to the tree are not persisted
CarlosRuiz: not related to your patch - but is not saving
nmicoud: ah, never noticed it ; maybe another related trees bug !
CarlosRuiz: I created a new tree (not all nodes)
CarlosRuiz: and now in tree maintenance my changes are not persisted
nmicoud: swing?
CarlosRuiz: both
CarlosRuiz: lost here - sometimes persist - sometimes does not
nmicoud: maybe related to kind of tree ?
CarlosRuiz: this is what I did
CarlosRuiz: created two summary regions
CarlosRuiz: created a new tree "region 2"
CarlosRuiz: type sales region
CarlosRuiz: all nodes unchecked
nmicoud: when you said " and now in tree maintenance my changes are not persisted" : what do you mean ? the sequence of nodes or the fact that they are in the left panel ?
CarlosRuiz: ah
CarlosRuiz: I think is related to your other chage
CarlosRuiz: change
CarlosRuiz: AdempiereTreeModel
CarlosRuiz: the moving to left is saved
CarlosRuiz: I moved a region and a summary to left
CarlosRuiz: and then moved the region into the summary - that's not saved
CarlosRuiz: but seems is because of the change the AdempiereTreeModel
CarlosRuiz: hmm no - debugging
nmicoud: when i've done my tests of moving nodes, i did not see any problem using webUI ; only using swing
CarlosRuiz: ah yes - it's caused because of the change to AdempiereTreeModel - reverted that change and now is saved
nmicoud: :-/ 526 was too easy
tbayen_: rehi. I missed the meeting again. :-(
CarlosRuiz: yes - I need to check better the 526 case
CarlosRuiz: it broke tree maintenance on swing
tbayen_: Are there issues or proposals about the wiki pages? We will insert more content and define more structure in the next weeks.
CarlosRuiz: tbayen, haven't heard any issue or proposal yet
nmicoud: if i remember well : line 80 : for (int i = 0; i < to.getChildCount(); i++)
nmicoud: getChildCount was always superior to 1
nmicoud: Should be : getChildCount was always superior of 1 to the real value
nmicoud: gtg ; will try to come back later, so if you have questions, please update jira ticket; bye bye
CarlosRuiz: thanks Nicolas
CarlosRuiz: gtg too - thanks for the meeting