IDempiere/FullMeeting20140205

⇐ Table of Contents | Full Meeting Minutes | Full Meeting 2014-02-05

CarlosRuiz: Good morning
tbayen: Daarestiet! :-)
nmicoud: Bonjour
JanThielemann: hi
nmicoud: I'm stucked with https://idempiere.atlassian.net/browse/IDEMPIERE-1694 for days. Any idea for fixing it ?
tbayen: Hi JanThielemann . You said you wrote "GV" classes for SEPA. Did you do that in your plugin or in the upstream HBCI4Java library?
JanThielemann: hbci4java library
tbayen: Can you give me links where you got the know-how to do that? I would like to understand these classes and have not worked with XML much in the past (only reading and writing DOM, but your classes do not look like DOM code...).
CarlosRuiz: nmicoud, RequestEMailProcessor line 550 - maybe you need to do some format on the part.getFileName ?
JanThielemann: i mainly used https://groups.google.com/forum/#!forum/hbci4java and had some mails with olaf willuhn
JanThielemann: also i used the sepa specifications
JanThielemann: but i currently can't find the link :/
JanThielemann: tbayen, have you taken a look at the test classes?
JanThielemann: TestGVUebSEPA and so on
nmicoud: CarlosRuiz : i've tried that : eg String fileName = MimeUtility.decodeText(part.getFileName()); or fileName=Normalizer.normalize(fileName, Normalizer.Form.NFC); but at the end there is still exotic characters
nmicoud: the part.getFileName is a string and it seems that you cannot change the encoding of it (it's too late)
tbayen: JanThielemann, yes, I have. So you used the code of Olaf and changed it for the new Multi GV?
CarlosRuiz: nmicoud, is it possible that the problem is in the message? does thunderbird receives the same attachment with wrong name?
JanThielemann: no, i refactored the code to use jaxb instead of the old xmlcreator
JanThielemann: and i created a factory
nmicoud: i haven't tested with thunderbird with the mailbox i used for the tests. But my profesional email has same parameters and i receive attachments with the 'good' names
JanThielemann: see my posts in https://groups.google.com/forum/#!forum/hbci4java (all the way down to the end)
CarlosRuiz: nmicoud - possible clue here -> http://www.oracle.com/technetwork/java/faq-135477.html#encodefilename
nmicoud: idempiere is running on a linux CentOS, so encoding is UTF-8. But if the email is sent from a windows, the encoding will be ISO-8859
CarlosRuiz: MimeUtility.decodeText(part.getFileName())
tbayen: Jan. Thanks! Your thread in the hbci4java forum is exactly what I was looking for. :-)
nmicoud: i've tried that, and if i remember well, there was still some "?" in the filename
CarlosRuiz: maybe then try the System properties "mail.mime.encodefilename" and "mail.mime.decodefilename" -> as suggested on that link
nmicoud: will try it. thanks ; do you know where i should set this system property ?
nmicoud: in the getSession() method ?
CarlosRuiz: you could try first as a JVM parameter
CarlosRuiz: like -> java -Dmail.mime.decodefilename=true
nmicoud: where should i put it on a running instance ?
Not-001: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to release-1.0c [+0/-0/±2] https://bitbucket.org/idempiere/idempiere/commits/
Not-001: [iDempiere] globalqss e23e3b8 - IDEMPIERE-1416 Pack-in issue when an attachment has been changed
CarlosRuiz: you could try adding that parameter on the idempiere-server.sh and see if it works
nmicoud: idempiere.ini ?
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-1416 status set to "Resolved" -resolution set to "Fixed"
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-1416
nmicoud: ok will try that ; thanks
nmicoud: it changes nothing :(
nmicoud: the "é" is 'translated' to "?"
CarlosRuiz: so, it drops all the "ISO-8859" stuff - but doesn't translate correctly the characters?
nmicoud: it seems
nmicoud: with MimeUtility.decodeText(part.getFileName() => "coté" become "cot=E9"
CarlosRuiz: and with "mail.mime.decodefilename" becomes "cot?" ?
nmicoud: there is still mail.mime.decodefilename=true in the idempiere-server.sh
nmicoud: just found that link : http://stackoverflow.com/questions/18352605/decoding-e9-to-utf8-fails will try it
tbayen: I got to go in some minutes. I hope you keep working much longer. ;-) cu
hengsin: nmicoud, everywhere I read, it say String filename = MimeUtility.decodeText(part.getFileName()) should work
nmicoud: i agree, but in my case, it doesn't Could it be related to a wrong encoding style in the idempiere server ?
nmicoud: that mean the problem is not in idempiere, but relates to something else, right ?
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-1416 status set to "Reopened" -resolution set to "None"
Not-001: [IDEMPIERE] Commit https://bitbucket.org/idempiere/idempiere/changeset/e23e3b8 solves a problem that attachment packin was just working for DB and not other attachment methods (like filesystem). The original problem described here is caused because the packout_example_2.zip has a different UUID than the packout_example.zip So, the POFinder.findPO doesn't find the attachment (because the UUID changed) and tries to create a new one. Will keep this open
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-1416
aguerra: Buenos dias people!!!!
CarlosRuiz: Hi Alejandro
aguerra: CarlosRuiz, hi!!!!
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-976 status set to "Open" -assignee set to "None"
Not-001: [IDEMPIERE] Hi Nicolas, unfortunately this patch is not valid with actual sources as it collided with the commit from IDEMPIERE-980. Would need to work on an actual patch. Regards, Carlos Ruiz
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-976
CarlosRuiz: now reviewing IDEMPIERE-1732
Not-001: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to release-1.0c [+0/-0/±1] https://bitbucket.org/idempiere/idempiere/commits/
Not-001: [iDempiere] globalqss 656c295 - IDEMPIERE-1732 two users can complete same M_InOut and data is corrupted / integrate patch from hengsin
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-1732 status set to "Resolved" -assignee set to "Heng Sin Low" -resolution set to "Fixed"
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-1732
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-1736
Not-001: [IDEMPIERE] Anna, the UUID idx is created automatically when you create the table, you can run the process "Create Table Index" to fill the index definition. I'm going to change the description of this ticket to "Create table must autofill uuidx definition". Regards, Carlos Ruiz
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-1736
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-1736 summary set to "Create table must autofill uuidx definition"
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-1736
CarlosRuiz: nmicoud, still there?
nmicoud: yes
CarlosRuiz: I don't remember if you had a similar ticket like this https://idempiere.atlassian.net/browse/IDEMPIERE-1729
CarlosRuiz: I think you did something similar - but I can't find the ticket
nmicoud: i guess. something related to orgs on login panel ; don't remember exactly
nmicoud: let me check
nmicoud: perhaps https://idempiere.atlassian.net/browse/IDEMPIERE-449 ?
CarlosRuiz: ah - got it
CarlosRuiz: https://idempiere.atlassian.net/browse/IDEMPIERE-449
nmicoud: :D
CarlosRuiz: revisiting this - I think a user preference could do the trick
CarlosRuiz: we could check if the user has a global preference set for the Org field (I mean on all windows)
nmicoud: how could he define it ?
nmicoud: directly in the preference window ?
CarlosRuiz: right click -> preference on org on any window
nmicoud: not sure as it is designed to fill some fields when already logged in
nmicoud: perhaps a new field on ad_user or in the WPreference ?
nmicoud: could be hard to explain that the default org for login has to be defined with a right click on any AD_Org_ID field, no ?
CarlosRuiz: and the idea is to set that one when the user doesn't check the "Select Role" on login?
nmicoud: yes or if he choose to "Select Role", the org field will be filled with this default org
CarlosRuiz: for me it sounds like is the org preference for the user
CarlosRuiz: but I thought 449 and 1729 were the same - now I see are different
nmicoud: 1729 is to limit available orgs, right ?
nmicoud: whereas 449 is to "force" a specific org for login
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-1729
Not-001: [IDEMPIERE] Thanks [~sureeraya], Code seems right but the migration script doesn't: * The migration script seems to be generated with Adempiere as it doesn't have the UUID columns (this could be solved running after that the fill UUID process, but also if you generate the script directly from iDempiere) * the IDs are not centralized which would cause conflict if integrated - if you want centralized ID just let me know and I'll generate a password for
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-1729
Not-001: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-1729 labels set to "+Patch"
Not-001: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-1729
Not-001: [iDempiere] CarlosRuiz_globalqss pushed 2 commits to development [+0/-0/±3] https://bitbucket.org/idempiere/idempiere/commits/
Not-001: [iDempiere] globalqss 6460a9e - hg merge release-1.0c (merge release1 into release2)
Not-001: [iDempiere] globalqss 9eea5bb - hg merge release-2.0 (merge release2 into development)
CarlosRuiz: gtg - thanks for the meeting