opensync-users

[Opensync-users] Is opensync with irmc-sync meanwhile able to handle german umlauts?

[Achim S. <Ach...@mn...>]

Hi,

I had not time during the last months to test the development of the opensync 
framework with irmc-plugin and kdepim-plugin.

So has anybody successfully tested this combo with German Umlauts (i.e. UTF-8 
special characters)?

Last time I tested this (some months ago), all Umlauts were destroyed when I 
tried to sync between kdepim and my mobile phone (SonyEricsson K750i).

##
I looked at the checkin documentation at www.opensync.org - and found some 
changes that seem to tackle the UTF-8 problem both in core and irmc-sync. But 
I didn't find any success reports for German Umlauts in combination with 
irmc-sync and kdepim.

I saw also a changeset, which described a new irmc-sync development branch. 
Which features are developed there separatedly?



Thanks for any success report,
Achim

Re: [Opensync-users] Is opensync with irmc-sync meanwhile able to handle german umlauts?

[Christopher S. <ope...@ch...>]

Hi Achim,

Am Donnerstag, 22. Juni 2006 17:50 schrieb Achim Spangler:
> Hi,
>
> I had not time during the last months to test the development of the
> opensync framework with irmc-plugin and kdepim-plugin.
>
> So has anybody successfully tested this combo with German Umlauts
> (i.e. UTF-8 special characters)?
>
> Last time I tested this (some months ago), all Umlauts were destroyed
> when I tried to sync between kdepim and my mobile phone (SonyEricsson
> K750i).
> [...]

I fixed a lot of iusses by removing sync_vtype() from the irmc-plugin. 
Please test the combination with opensync (trunk) and irmc-development 
(branch).

I can sync vcards with german umlauts without any problems (also 
SonyEricsson K750i).

Regards,

	Christopher

Re: [Opensync-users] Is opensync with irmc-sync meanwhile able to handle german umlauts?

[Achim S. <Ach...@mn...>]

Hello Christopher,

hmmh - I was not able to verify the correct function so far, as my K750i is 
still crashing during sync (screen goes blank black,white ... and comes back 
after some seconds).

The output of the sync around the crash:
Sent a entry 000000000231 of size 399 to member 2. Changetype ADDED
Error writing entry 36651AE5BEB28DD2C1256FB700577E8B-Lotus_Notes_Generated to 
member 2: Timeout while waiting for a reply to message "COMMIT_CHANGE"

(all following entries are then also reporting the timeout)

How did I get this:
1) I started a fresh sync, where the standard reousrces for Kalendar and 
   contact pointed to fresh generated files (i.e. empty).
  --> calendar from mobile phone has been transfered fine

2) removed an obsolete calendar entry

3) restarted the sync
 --> several conflicts were detected - even if there was no real reason
     ( nearly all calendar entries --> the history of my konsole is no long
       enough; and I won't retry, as I had already a total permanent crash
       with my mobile during the last tries )
    ==> some of the control XML-tags like data source, ... were probably
        causing the conflict ( so the user visible parts were identic, but
        several other information was different in all reported calendar
        conflicts )
  --> after all "conflicts" were resolved by taking the entry from phone
      ( the entry, where the special XML-Tags indicated the phone as source ),
      the changes were being commited
      ---> during THIS flow, the mobile resetted itself

So maybe the syncing is disturbed by a calendar entry which is stemming from 
Lotus-Notes ( it was created on Lotus-Notes, sent by email to my 
email-account, and had been accepted/stored by kontact into KDE calendar ). 
Thus it might have contained some strange special XML tags, which are not 
handled by the mobile in the right way.

I hope that my description can help someway.

My configuration:
+ Kubuntu Dapper
+ opensync packages from the REPO at http://www.in.fh-merseburg.de/~jahn/
  ( today installed --> 0.18+svn20060605-1.2 )
+ Mobile: SonyEricsson K750i
+ newest KDE 3.5.3 from the Dapper update sources

My suggestion for solution:
Do not communicate the special XML-Tags which are not neccessary for real data 
and sync marks. KDE won't understand all K750i entries and vice versa.

Bye,
Achim

Am Donnerstag, 22. Juni 2006 18:35 schrieb Christopher Stender:
> Hi Achim,
>
> Am Donnerstag, 22. Juni 2006 17:50 schrieb Achim Spangler:
> > Hi,
> >
> > I had not time during the last months to test the development of the
> > opensync framework with irmc-plugin and kdepim-plugin.
> >
> > So has anybody successfully tested this combo with German Umlauts
> > (i.e. UTF-8 special characters)?
> >
> > Last time I tested this (some months ago), all Umlauts were destroyed
> > when I tried to sync between kdepim and my mobile phone (SonyEricsson
> > K750i).
> > [...]
>
> I fixed a lot of iusses by removing sync_vtype() from the irmc-plugin.
> Please test the combination with opensync (trunk) and irmc-development
> (branch).
>
> I can sync vcards with german umlauts without any problems (also
> SonyEricsson K750i).
>
> Regards,
>
> 	Christopher
>
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job
> easier Download IBM WebSphere Application Server v.1.0.1 based on Apache
> Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Opensync-users mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensync-users

Re: [Opensync-users] Is opensync with irmc-sync meanwhile able to handle german umlauts?

[Daniel G. <dg...@su...>]

On Thursday 22 June 2006 20:14, Achim Spangler wrote:
> How did I get this:
> 1) I started a fresh sync, where the standard reousrces for Kalendar and
> =A0 =A0contact pointed to fresh generated files (i.e. empty).
> =A0 --> calendar from mobile phone has been transfered fine
>
> 2) removed an obsolete calendar entry
>
> 3) restarted the sync
> =A0--> several conflicts were detected - even if there was no real reason
> =A0 =A0 =A0( nearly all calendar entries --> the history of my konsole is=
 no long
> =A0 =A0 =A0 =A0enough; and I won't retry, as I had already a total perman=
ent crash
> =A0 =A0 =A0 =A0with my mobile during the last tries )
> =A0 =A0 =3D=3D> some of the control XML-tags like data source, ... were p=
robably
> =A0 =A0 =A0 =A0 causing the conflict ( so the user visible parts were ide=
ntic, but
> =A0 =A0 =A0 =A0 several other information was different in all reported c=
alendar
> =A0 =A0 =A0 =A0 conflicts )
> =A0 --> after all "conflicts" were resolved by taking the entry from phone
> =A0 =A0 =A0 ( the entry, where the special XML-Tags indicated the phone a=
s source

(msynctool --sync YYYY --conflict X  to let always member X win the conflic=
t)

> ), the changes were being commited
> =A0 =A0 =A0 ---> during THIS flow, the mobile resetted itself

Can you reproduce this with enabled OSYNC_TRACE?
( export OSYNC_TRACE=3D/tmp/osync_trace/; mkdir $OSYNC_TRACE; msynctool --s=
ync=20
YYYY )

Does this crash only happen with this lotus generated calendar entries?

Also interesting would be lotus generated vcalendar file (without sensitive=
=20
information ;) )

best regards,
Daniel

Re: [Opensync-users] Is opensync with irmc-sync meanwhile able to handle german umlauts?

[Christopher S. <ope...@ch...>]

Hi Achim

Am Donnerstag, 22. Juni 2006 20:14 schrieb Achim Spangler:
> Hello Christopher,
>
> hmmh - I was not able to verify the correct function so far, as my
> K750i is still crashing during sync (screen goes blank black,white
> ... and comes back after some seconds).

I had some crashes with old firmware versions, please provide yours.

(without the double-quotes, joystick left = <, joystick right = >)
">*<<*<*  Service Info -> Software Info -> RXXXXX"

> The output of the sync around the crash:
> Sent a entry 000000000231 of size 399 to member 2. Changetype ADDED
> Error writing entry
> 36651AE5BEB28DD2C1256FB700577E8B-Lotus_Notes_Generated to member 2:
> Timeout while waiting for a reply to message "COMMIT_CHANGE"
>
> (all following entries are then also reporting the timeout)

Do you now which vcal cause this problem? What happens if you delete 
this calender entry?

> [...]

Best regards

	Christopher

Re: [Opensync-users] Is opensync with irmc-sync meanwhile able to handle german umlauts?

[Xin Z. <xi...@gm...>]

在 2006-06-22四的 18:35 +0200，Christopher Stender写道：
> I fixed a lot of iusses by removing sync_vtype() from the irmc-plugin. 
> Please test the combination with opensync (trunk) and irmc-development 
> (branch).
> 
> I can sync vcards with german umlauts without any problems (also 
> SonyEricsson K750i).
> 
This branch of irmc-development is really working. With it I finally got
my Chinese contacts stored and displayed correctly in my SonyEricsson
T68. I was very excited that charsets conversion was no longer a
problem.

And there is a little issue: Everytime I access the "vnote" of my phone,
it stops responding any further IrMC request unless I reboot it.

It seems a bug of my phone, but it really blocks me. I have to comment
out some code in irmc_sync.c to disable "vnote". So I hope there could
be a decent option to choose which datatype I want to sync (cal, pb or
nt). I'll be glad to help. Thanks for your great work.

> Regards,
> 
> 	Christopher

Re: [Opensync-users] =?utf-8?q?Is_opensync_with_irmc-sync_meanwhile_a?= =?utf-8?q?ble_to=09handle_german_umlauts=3F?=

[Christopher S. <ope...@ch...>]

Am Sonntag, 25. Juni 2006 21:42 schrieb Xin Zhen:
> =E5=9C=A8 2006-06-22=E5=9B=9B=E7=9A=84 18:35 +0200=EF=BC=8CChristopher St=
ender=E5=86=99=E9=81=93=EF=BC=9A
>
> > I fixed a lot of iusses by removing sync_vtype() from the
> > irmc-plugin. Please test the combination with opensync (trunk) and
> > irmc-development (branch).
> >
> > I can sync vcards with german umlauts without any problems (also
> > SonyEricsson K750i).
>
> This branch of irmc-development is really working. With it I finally
> got my Chinese contacts stored and displayed correctly in my
> SonyEricsson T68. I was very excited that charsets conversion was no
> longer a problem.

Great.

> And there is a little issue: Everytime I access the "vnote" of my
> phone, it stops responding any further IrMC request unless I reboot
> it.
>
> It seems a bug of my phone, but it really blocks me. I have to
> comment out some code in irmc_sync.c to disable "vnote". So I hope
> there could be a decent option to choose which datatype I want to
> sync (cal, pb or nt). I'll be glad to help. Thanks for your great
> work.

If you're using msynctool, please try:
msynctool --sync $GROUP --filter-objtype note


Best Regards

	Christopher

Re: [Opensync-users] Is opensync with irmc-sync meanwhile able to handle german umlauts?

[Xin Z. <xi...@gm...>]

在 2006-06-26一的 10:20 +0200，Christopher Stender写道：
> 
> > And there is a little issue: Everytime I access the "vnote" of my
> > phone, it stops responding any further IrMC request unless I reboot
> > it.
> >
> > It seems a bug of my phone, but it really blocks me. I have to
> > comment out some code in irmc_sync.c to disable "vnote". So I hope
> > there could be a decent option to choose which datatype I want to
> > sync (cal, pb or nt). I'll be glad to help. Thanks for your great
> > work.
> 
> If you're using msynctool, please try:
> msynctool --sync $GROUP --filter-objtype note
> 

It is in irmcConnect():

  if ( !detect_slowsync( config->notebook_changecounter, "nt",
&(config->notebook_dbid),

This line seems to shutdown my phone. Since the phone won't respond,
this function never returns until timeout, while the phone keeps showing
"Synchroninzing". This code is not affected by --filter-objtype note.

Re: [Opensync-users] =?utf-8?q?Is_opensync_with_irmc-sync_meanwhile_a?= =?utf-8?q?ble=09to=09handle_german_umlauts=3F?=

[Christopher S. <ope...@ch...>]

Hi,

Am Montag, 26. Juni 2006 20:49 schrieb Xin Zhen:
> =E5=9C=A8 2006-06-26=E4=B8=80=E7=9A=84 10:20 +0200=EF=BC=8CChristopher St=
ender=E5=86=99=E9=81=93=EF=BC=9A
>
> > > And there is a little issue: Everytime I access the "vnote" of my
> > > phone, it stops responding any further IrMC request unless I
> > > reboot it.
> > >
> > > It seems a bug of my phone, but it really blocks me. I have to
> > > comment out some code in irmc_sync.c to disable "vnote". So I
> > > hope there could be a decent option to choose which datatype I
> > > want to sync (cal, pb or nt). I'll be glad to help. Thanks for
> > > your great work.
> >
> > If you're using msynctool, please try:
> > msynctool --sync $GROUP --filter-objtype note
>
> It is in irmcConnect():
>
>   if ( !detect_slowsync( config->notebook_changecounter, "nt",
> &(config->notebook_dbid),
>
> This line seems to shutdown my phone. Since the phone won't respond,
> this function never returns until timeout, while the phone keeps
> showing "Synchroninzing". This code is not affected by
> --filter-objtype note.

Yes, you're right. It looks like new Sony Ericsson cell phones have a=20
bug with irmc+notes. Maybe we can add a blacklist for note=20
synchronization to the irmc-plugin.

Best regards

	Christopher

Re: [Opensync-users] Slow-sync conflict?

[Xin Z. <xi...@gm...>]

在 2006-06-30五的 16:23 +0200，Christopher Stender写道： 
> Am Freitag, 30. Juni 2006 10:57 schrieb Christopher Stender:
> > Am Freitag, 30. Juni 2006 10:11 schrieb Christopher Stender:
> > > Hi Xin,
> > > >
> > > > Entry 2:
> > > > UID: pas-id-44A21F92000001E1
> > > > <?xml version="1.0"?>
> > > > <contact>
> > > > <Name>
> > > > <FirstName>Cindy</FirstName>
> > > > </Name>
> > > > <Telephone>
> > > > <Content>13817665288</Content>
> > > > <Type>CELL</Type>
> > > > </Telephone>
> > > > <UnknownNode>
> > > > <NodeName>X-IRMC-LUID</NodeName>
> > > > <Content>000042000000</Content>
> > > > </UnknownNode>
> > > > <FormattedName>
> > > > <Content>Cindy</Content>
> > > > </FormattedName>
> > > > <Uid>
> > > > <Content>pas-id-44A21F92000001E1</Content>
> > > > </Uid>
> > > > <Revision>
> > > > <Content>2006-06-28T06:20:02Z</Content>
> > > > </Revision>
> > > > <FormattedName>
> > > > <Content>Cindy</Content>
> > > > </FormattedName>
> > > > </contact>
> >
> > Hmm, I recognized that there are two "FormattedName" attributes.
> > Maybe this is the reason for the conflict, but I need the TRACE-Files
> > to verify this. I wrote a patch for opensync which genereates the
> > FN-Attribute because the K750i/D750i doesn't do that. Maybe Evolution
> > also genereate one. I try to reproduce this issue.
> 
> I committed a patch (r1077) which fixed this issue. Please check out svn 
> and try it again.

Great! All the slow-sync conflicts are gone now. Some other issues, like
no-name contact (<FormattedName/>) in phone book, disappeared either.

Then I found another issue: the number entered in "Home" field of phone
will go to "Business Phone" field of Evolution. While the evo-to-irmc
mapping is OK ("Business Phone" to "Work", "Home Phone" to "Home"). Is
this known before?

> Regards
> 
> 	Christopher