about DataPilot

[Zhi Yu Yue]

Hello,

We (IBM) has done some improvements on DataPilot (improved UI, performance, cache support, ...), and would like to contribute the code back. I know there are also some DataPilot work ongoing here. What steps should we follow? or who can I contact? I have already read the guide here, http://contributing.openoffice.org/programming.html. It says I should come here. Any advise would be appreciated.
 
Best Regards,
Helen Yue
_________________________________________________________
Yue Zhi Yu  |  User Experience & Applications, Lotus Symphony |  yuezhiyu@...

[Kohei Yoshida-2]

On Wed, 2009-06-10 at 09:26 +0800, Zhi Yu Yue wrote:
> Hello,
>
> We (IBM) has done some improvements on DataPilot (improved UI,
> performance, cache support, ...), and would like to contribute the
> code back. I know there are also some DataPilot work ongoing here.
> What steps should we follow? or who can I contact? I have already read
> the guide here, http://contributing.openoffice.org/programming.html.
> It says I should come here. Any advise would be appreciated.

I've already sent you an email, but the first thing we need to do is to
coordinate our work to avoid duplication of effort.  I have several
DataPilot work in the pipeline, so we need to make sure we don't step on
each other's toes.

Can you provide the details of your enhancement?  Your description
(improved UI, performance, cache support, ...) is a bit vague.

Thanks!

Kohei
--
Kohei Yoshida - OpenOffice.org Engineer - Novell, Inc.
<kyoshida@...>


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Niklas Nebel]

On 06/10/09 04:15, Kohei Yoshida wrote:

> On Wed, 2009-06-10 at 09:26 +0800, Zhi Yu Yue wrote:
>> We (IBM) has done some improvements on DataPilot (improved UI,
>> performance, cache support, ...), and would like to contribute the
>> code back. I know there are also some DataPilot work ongoing here.
>> What steps should we follow? or who can I contact? I have already read
>> the guide here, http://contributing.openoffice.org/programming.html.
>> It says I should come here. Any advise would be appreciated.
>
> I've already sent you an email, but the first thing we need to do is to
> coordinate our work to avoid duplication of effort.  I have several
> DataPilot work in the pipeline, so we need to make sure we don't step on
> each other's toes.

I have to disagree here. If IBM wants to contribute their changes,
that's great and should be done right away, without waiting for go-oo
changes that are being held back instead of contributed.

Kohei, if you do your changes outside of openoffice.org, there's always
the risk of duplicated work. Remember the solver? It's your decision,
but don't try to inflict the same on others.

Helen, performance is on my agenda anyway, so we can work together on
that very quickly. For UI changes, we have to involve the User
Experience community. It would be best if we have a summary of the UI
changes that we can discuss on the discuss@... list.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Zhi Yu Yue]

Either way is OK for me. I just want to contribute the code quickly. :-)

Let me give my list first. We started this work mid of last year. Some work are based on OpenOffice code, while some are new. I list the new items as below. We may still need some time to migrate to new code base. I will soon send the UI related specification to user experience mail list.

-- New sidebar UI: Users can easily assign fields in sidebar by only drag and drop anytime to change the table structure and access the commands easily.
-- Empty table, default table style, and auto-resize: since we still do not support the AutoFormat, we improved the user experience by providing the instructions on empty table, well-defined table style by default, and also automatically resize the width when the table is created.
-- Duplicate items: users can drag the same fields multiple times, and set different field settings.
-- Cache/buffer support: MS Excel has this feature. With this, even the data source is deleted, the DataPilot table can still be updated.
-- Performance: roughly ~40% improvement on generating table and fields operation
-- Field items show/hide UI panel: provide a new filter panel to show/hide items easily. Users can also set Top N in this panel.
 
Best Regards,
Helen Yue
_________________________________________________________
Yue Zhi Yu  |  User Experience & Applications, Lotus Symphony |  yuezhiyu@...

From:	Niklas Nebel <Niklas.Nebel@...>
To:	dev@...
Date:	06/10/2009 03:48 PM
Subject:	Re: [sc-dev] about DataPilot
Sent by:	Niklas.Nebel@...

---

On 06/10/09 04:15, Kohei Yoshida wrote:
> On Wed, 2009-06-10 at 09:26 +0800, Zhi Yu Yue wrote:
>> We (IBM) has done some improvements on DataPilot (improved UI,
>> performance, cache support, ...), and would like to contribute the
>> code back. I know there are also some DataPilot work ongoing here.
>> What steps should we follow? or who can I contact? I have already read
>> the guide here, http://contributing.openoffice.org/programming.html.
>> It says I should come here. Any advise would be appreciated.
>
> I've already sent you an email, but the first thing we need to do is to
> coordinate our work to avoid duplication of effort.  I have several
> DataPilot work in the pipeline, so we need to make sure we don't step on
> each other's toes.

I have to disagree here. If IBM wants to contribute their changes,
that's great and should be done right away, without waiting for go-oo
changes that are being held back instead of contributed.

Kohei, if you do your changes outside of openoffice.org, there's always
the risk of duplicated work. Remember the solver? It's your decision,
but don't try to inflict the same on others.

Helen, performance is on my agenda anyway, so we can work together on
that very quickly. For UI changes, we have to involve the User
Experience community. It would be best if we have a summary of the UI
changes that we can discuss on the discuss@... list.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Niklas Nebel]

On 06/10/09 11:08, Zhi Yu Yue wrote:
> Either way is OK for me. I just want to contribute the code quickly. :-)

That really sounds very good.

> Let me give my list first. We started this work mid of last year. Some
> work are based on OpenOffice code, while some are new. I list the new
> items as below. We may still need some time to migrate to new code base.
> I will soon send the UI related specification to user experience mail list.

Find some comments on the list items below. The next steps could be two
things in parallel:
- Discussion of the UI changes with ux community.
- Looking at the performance improvements (see below).

> -- New sidebar UI: Users can easily assign fields in sidebar by only
> drag and drop anytime to change the table structure and access the
> commands easily.

I assume this depends on the general sidebar infrastructure?
The quickest way to get such an extra window into OOo is to use a
"docking window", like the Navigator or styles. See
ScFunctionChildWindow (Insert - Function List) for an example in Calc.
It's a VCL window, of course, so the interaction inside the window has
to be done with VCL controls.

> -- Empty table, default table style, and auto-resize: since we still do
> not support the AutoFormat, we improved the user experience by providing
> the instructions on empty table, well-defined table style by default,
> and also automatically resize the width when the table is created.

This sounds like something that can be applied to current code without
big changes.

> -- Duplicate items: users can drag the same fields multiple times, and
> set different field settings.

For data fields, right? It's mostly the dialog that currently doesn't
handle this.

> -- Cache/buffer support: MS Excel has this feature. With this, even the
> data source is deleted, the DataPilot table can still be updated.

Internally, OOo now uses the ScDPCacheTable. So in this area there will
probably be conflicts. But ScDPCacheTable still has some problems with
memory usage, maybe your solution is even better.

> -- Performance: roughly ~40% improvement on generating table and fields
> operation

Which parts were improved? We've had some improvements with issues 63686
and 72188. Do they overlap with your changes?

It would be great if you could isolate your performance improvements
from the other changes. If you have a diff against the old code, I would
like to look at it to see if the changes can be used with the current
version. This could then be the first thing to integrate.

> -- Field items show/hide UI panel: provide a new filter panel to
> show/hide items easily. Users can also set Top N in this panel.

This sounds similar to what Kohei has announced but not yet contributed.
Let's see if he wants to contribute his changes now, then we can try to
merge them.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[michael meeks]

Hi Niklas,

On Wed, 2009-06-10 at 09:45 +0200, Niklas Nebel wrote:
> On 06/10/09 04:15, Kohei Yoshida wrote:
> > I've already sent you an email, but the first thing we need to do is to
> > coordinate our work to avoid duplication of effort.  I have several
> > DataPilot work in the pipeline, so we need to make sure we don't step on
> > each other's toes.
>
> I have to disagree here. If IBM wants to contribute their changes,
> that's great and should be done right away, without waiting for go-oo
> changes that are being held back instead of contributed.

        What makes you think they are being held back ? the issues around
DataPilot that Kohei are working on have been communicated quite clearly
to Sun - eg. i#101328 - how clear can it be: the comment "accepting it
for 3.2" from Kohei; i#100619, i#22029 both assigned to
Kohei, all fitting the mould of work we send up-stream.

        It seems extraordinarily short sighted, and saddening to try and
avoid co-ordination on this; it makes it appear that you almost relish
destroying other people's work by duplication, even when they are trying
hard to work with you - which is a highly unfortunate and unpleasant
position. Strangely, sad to say, this is a friction that is not normally
experienced when dealing with your co-workers.

        Despite your obviously offensive mail; to make it easier for IBM to
review the changes, I've asked Kohei to create a CWS and whack the
patches in there - please don't do the normal witheringly negative critique,
they may not be completely ready yet. FWIW, I'm always amused when I ask for
design review, people prefer instead to rave about commented (or not)
debugging fprintfs instead ;-)

        Then there is this gratuitous hair-curling bit:

> Kohei, if you do your changes outside of openoffice.org, there's always
> the risk of duplicated work. Remember the solver? It's your decision,
> but don't try to inflict the same on others.

        So - the context is that Kohei has asked, politely, and reasonably
asked, to co-ordinate this work so we don't waste effort here - he talks
of a pipeline of changes in data-pilot, of solidifying the code. And then
you turned him down: why all this aggressive talk of deliberate work
duplication ?

        If you choose to dig up the past, lets do it: sure - we remember
the solver, there was my code in there too; we remember your leading
role in destroying a volunteer's spare-time creation, your tone almost
sounds as if you enjoyed that. I had always assumed this appalling
initiative to abuse, divide and stifle the 'Open'Office.org project,
purely in Sun's proprietary interest, was forced by StarDivisions's
(visionary) senior management, rather than being cheer-lead by
individual Sun developers.

        Since I see almost nothing defensible about Sun's past action
here, I would (personally) recommend avoiding raking over it all again;
why don't we just get on with the technical issues around improving
the DataPilot implementation.

        Regards,

                Michael.

--
 michael.meeks@...  <><, Pseudo Engineer, itinerant idiot


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Niklas Nebel]

On 06/10/09 16:51, Michael Meeks wrote:

>         What makes you think they are being held back ? the issues around
> DataPilot that Kohei are working on have been communicated quite clearly
> to Sun - eg. i#101328 - how clear can it be: the comment "accepting it
> for 3.2" from Kohei; i#100619, i#22029 both assigned to
> Kohei, all fitting the mould of work we send up-stream.
>
>         It seems extraordinarily short sighted, and saddening to try and
> avoid co-ordination on this; it makes it appear that you almost relish
> destroying other people's work by duplication, even when they are trying
> hard to work with you - which is a highly unfortunate and unpleasant
> position. Strangely, sad to say, this is a friction that is not normally
> experienced when dealing with your co-workers.
>
>         Despite your obviously offensive mail; to make it easier for IBM to
> review the changes, I've asked Kohei to create a CWS and whack the
> patches in there - please don't do the normal witheringly negative critique,
> they may not be completely ready yet. FWIW, I'm always amused when I ask for
> design review, people prefer instead to rave about commented (or not)
> debugging fprintfs instead ;-)

There is code that the IBM people want to contribute now, and code that
you advertise as a go-oo feature but didn't contribute yet. If you're
changing your mind now, that's good. As I wrote: Then we can try to
merge them (the changes). Otherwise, what else can we do but use the
code that we can get? I am not going to delay integration of IBM's
changes until you decide you've kept your changes to yourself long enough.

>         So - the context is that Kohei has asked, politely, and reasonably
> asked, to co-ordinate this work so we don't waste effort here - he talks
> of a pipeline of changes in data-pilot, of solidifying the code. And then
> you turned him down: why all this aggressive talk of deliberate work
> duplication ?
>
>         If you choose to dig up the past, lets do it: sure - we remember
> the solver, there was my code in there too; we remember your leading
> role in destroying a volunteer's spare-time creation, your tone almost
> sounds as if you enjoyed that. I had always assumed this appalling
> initiative to abuse, divide and stifle the 'Open'Office.org project,
> purely in Sun's proprietary interest, was forced by StarDivisions's
> (visionary) senior management, rather than being cheer-lead by
> individual Sun developers.
>
> Since I see almost nothing defensible about Sun's past action
> here, I would (personally) recommend avoiding raking over it all again;
> why don't we just get on with the technical issues around improving
> the DataPilot implementation.

For at least two years (2005 to 2007) we didn't start a solver
component, although there was demand and the approach was sketched out,
because I believed Kohei was going to make (and contribute) one. I don't
want the same situation with DataPilot enhancements.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Kohei Yoshida-2]

On Wed, 2009-06-10 at 17:42 +0200, Niklas Nebel wrote:

> On 06/10/09 16:51, Michael Meeks wrote:
> >         What makes you think they are being held back ? the issues around
> > DataPilot that Kohei are working on have been communicated quite clearly
> > to Sun - eg. i#101328 - how clear can it be: the comment "accepting it
> > for 3.2" from Kohei; i#100619, i#22029 both assigned to
> > Kohei, all fitting the mould of work we send up-stream.
> >
> >         It seems extraordinarily short sighted, and saddening to try and
> > avoid co-ordination on this; it makes it appear that you almost relish
> > destroying other people's work by duplication, even when they are trying
> > hard to work with you - which is a highly unfortunate and unpleasant
> > position. Strangely, sad to say, this is a friction that is not normally
> > experienced when dealing with your co-workers.
> >
> >         Despite your obviously offensive mail; to make it easier for IBM to
> > review the changes, I've asked Kohei to create a CWS and whack the
> > patches in there - please don't do the normal witheringly negative critique,
> > they may not be completely ready yet. FWIW, I'm always amused when I ask for
> > design review, people prefer instead to rave about commented (or not)
> > debugging fprintfs instead ;-)
>
> There is code that the IBM people want to contribute now, and code that
> you advertise as a go-oo feature but didn't contribute yet.

So, what you mean exactly by 'contribute'?  I have created issues for my
work, has set the status to STARTED, and expressed my interest on
contributing.  How does that not constitute as a desire to contribute?
Obviously that's not too different from the way IBM has expressed their
interest in "contributing". Besides, what's wrong with advertising
features as go-oo?  IBM obviously advertises their features as features
of Symphony.  How is that different?  Please enlighten me.

> If you're
> changing your mind now, that's good. As I wrote: Then we can try to
> merge them (the changes). Otherwise, what else can we do but use the
> code that we can get? I am not going to delay integration of IBM's
> changes until you decide you've kept your changes to yourself long enough.

I have no mind to change here; I have always expressed interest in
contributing the DataPilot code back once the code matures & becomes in
good shape.  Perhaps it is your mind that needs to change; the way you
seem to see us (or see me), in a very twisted way, and unwillingness to
communicate with us (or me, I don't care the distinction).  If you had
asked me on the update, and asked me to hurry up on the integration, I
would have gladly done so as I have done so in the past when I was asked
to do it.

>
> >         So - the context is that Kohei has asked, politely, and reasonably
> > asked, to co-ordinate this work so we don't waste effort here - he talks
> > of a pipeline of changes in data-pilot, of solidifying the code. And then
> > you turned him down: why all this aggressive talk of deliberate work
> > duplication ?
> >
> >         If you choose to dig up the past, lets do it: sure - we remember
> > the solver, there was my code in there too; we remember your leading
> > role in destroying a volunteer's spare-time creation, your tone almost
> > sounds as if you enjoyed that. I had always assumed this appalling
> > initiative to abuse, divide and stifle the 'Open'Office.org project,
> > purely in Sun's proprietary interest, was forced by StarDivisions's
> > (visionary) senior management, rather than being cheer-lead by
> > individual Sun developers.
> >
> > Since I see almost nothing defensible about Sun's past action
> > here, I would (personally) recommend avoiding raking over it all again;
> > why don't we just get on with the technical issues around improving
> > the DataPilot implementation.
>
> For at least two years (2005 to 2007) we didn't start a solver
> component, although there was demand and the approach was sketched out,
> because I believed Kohei was going to make (and contribute) one. I don't
> want the same situation with DataPilot enhancements.

So, your mind is unfairly fixed on the past solver issue.  Have you
checked my track record since then?  When did I explicitly hold back my
contribution to Calc core ever since that solver incident?  Please tell
me.

I'm confused.

Kohei



---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Kohei Yoshida-2]

On Wed, 2009-06-10 at 14:51 +0000, Michael Meeks wrote:
> I've asked Kohei to create a CWS and whack the
> patches in there

There is one problem.  I can't seem to create a CWS.  I've tried:

$ cws create DEV300 kohei03 -m m49
.. check cws path: 'svn
+ssh://svn@.../ooo/cws/kohei03'
svn+ssh://svn@.../ooo/cws/kohei03:  (Not a valid
URL)

svn: A problem occurred; see other errors for details
cws: ERROR: The subversion command line client failed with exit status
'1'

FAILURE: cws aborted.

So, I've created one via EIS, which seems to have succeeded.  But I
can't seem to set the status to 'new'.  But I went ahead and tried to
check out the cws:

$ svn co svn+ssh://svn@.../ooo/cws/kohei03
svn: URL 'svn+ssh://svn@.../ooo/cws/kohei03'
doesn't exist

Maybe I have forgotten to perform some magical dance or whatnot, maybe I
have forgotten the proper way of creating a CWS, or maybe my mind is not
worthy of new CWS. ;-)

Anyway, I'm still investigating...

Just FYI...

Kohei


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Daniel Rentz]

Kohei Yoshida schrieb:

> On Wed, 2009-06-10 at 14:51 +0000, Michael Meeks wrote:
>> I've asked Kohei to create a CWS and whack the
>> patches in there
>
> There is one problem.  I can't seem to create a CWS.  I've tried:
>
> $ cws create DEV300 kohei03 -m m49
> .. check cws path: 'svn
> +ssh://svn@.../ooo/cws/kohei03'
> svn+ssh://svn@.../ooo/cws/kohei03:  (Not a valid
> URL)
>
> svn: A problem occurred; see other errors for details
> cws: ERROR: The subversion command line client failed with exit status
> '1'

Weird. Why should that be an invalid URL? Maybe some SSH problems?


> So, I've created one via EIS, which seems to have succeeded.  But I
> can't seem to set the status to 'new'.  But I went ahead and tried to
> check out the cws:

Using the EIS 'create CWS' function does only create the database entry
in EIS, not a new directory in SVN. The only way to change the CWS from
'planned' to 'new' (or to create a 'new' CWS from scratch) is to call
'cws create' which creates the SVN copy and updates/creates the database
entry in EIS.



Daniel

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Kohei Yoshida-2]

On Wed, 2009-06-10 at 19:50 +0200, Daniel Rentz wrote:

> Kohei Yoshida schrieb:
> > On Wed, 2009-06-10 at 14:51 +0000, Michael Meeks wrote:
> >> I've asked Kohei to create a CWS and whack the
> >> patches in there
> >
> > There is one problem.  I can't seem to create a CWS.  I've tried:
> >
> > $ cws create DEV300 kohei03 -m m49
> > .. check cws path: 'svn
> > +ssh://svn@.../ooo/cws/kohei03'
> > svn+ssh://svn@.../ooo/cws/kohei03:  (Not a valid
> > URL)
> >
> > svn: A problem occurred; see other errors for details
> > cws: ERROR: The subversion command line client failed with exit status
> > '1'
>
> Weird. Why should that be an invalid URL? Maybe some SSH problems?

Not as far as I know.  As an example, I can check out existing CWS'es
via ssh tunnel just fine.  And I hadn't had any problems with CWS'es in
the past either.

> > So, I've created one via EIS, which seems to have succeeded.  But I
> > can't seem to set the status to 'new'.  But I went ahead and tried to
> > check out the cws:
>
> Using the EIS 'create CWS' function does only create the database entry
> in EIS, not a new directory in SVN. The only way to change the CWS from
> 'planned' to 'new' (or to create a 'new' CWS from scratch) is to call
> 'cws create' which creates the SVN copy and updates/creates the database
> entry in EIS.

Gotcha.  So, to create a new cws I can commit to, I have to run the
command, which doesn't seem to work at the moment.  :-(

Oh well.

Kohei


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Kohei Yoshida-2]

On Wed, 2009-06-10 at 15:49 -0400, Kohei Yoshida wrote:

> On Wed, 2009-06-10 at 19:50 +0200, Daniel Rentz wrote:
> > Kohei Yoshida schrieb:
> > > On Wed, 2009-06-10 at 14:51 +0000, Michael Meeks wrote:
> > >> I've asked Kohei to create a CWS and whack the
> > >> patches in there
> > >
> > > There is one problem.  I can't seem to create a CWS.  I've tried:
> > >
> > > $ cws create DEV300 kohei03 -m m49
> > > .. check cws path: 'svn
> > > +ssh://svn@.../ooo/cws/kohei03'
> > > svn+ssh://svn@.../ooo/cws/kohei03:  (Not a valid
> > > URL)
> > >
> > > svn: A problem occurred; see other errors for details
> > > cws: ERROR: The subversion command line client failed with exit status
> > > '1'
> >
> > Weird. Why should that be an invalid URL? Maybe some SSH problems?
>
> Not as far as I know.  As an example, I can check out existing CWS'es
> via ssh tunnel just fine.  And I hadn't had any problems with CWS'es in
> the past either.

Ok.  I've tried it on a different machine with svn 1.5.5, and it worked.
The version of svn that failed is 1.6.2.  So, apparently it was because
of the svn client.

Thanks for your help, Daniel.

Kohei

--
Kohei Yoshida - OpenOffice.org Engineer - Novell, Inc.
<kyoshida@...>


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Zhi Yu Yue]

Glad to know we are good now. I did not hope my thread becomes a fighting. :-) Thanks everyone, Kohei, Niklas, Michael, and Daniel, for your kind help.

I'm new here, so I list my ToDos as below. Please help to advise.
1. Send our UX spec on user experience community for review.
  - I compared Kohei's list with mine. Most are different and should be fine. One may have some overlap. Kohei, I would appreciate if you can also review the UX spec after I post it. We can solve the conflict at the UX level.
2. Check in the code after UX spec is agreed. I may have more questions at that time.
 
Best Regards,
Helen Yue
_________________________________________________________
Yue Zhi Yu  |  Manager of User Experience & Applications, Lotus Symphony |  yuezhiyu@...

From:	Kohei Yoshida <kyoshida@...>
To:	dev@...
Date:	06/11/2009 04:38 AM
Subject:	Re: [sc-dev] about DataPilot

---

On Wed, 2009-06-10 at 15:49 -0400, Kohei Yoshida wrote:
> On Wed, 2009-06-10 at 19:50 +0200, Daniel Rentz wrote:
> > Kohei Yoshida schrieb:
> > > On Wed, 2009-06-10 at 14:51 +0000, Michael Meeks wrote:
> > >> I've asked Kohei to create a CWS and whack the
> > >> patches in there
> > >
> > > There is one problem.  I can't seem to create a CWS.  I've tried:
> > >
> > > $ cws create DEV300 kohei03 -m m49
> > > .. check cws path: 'svn
> > > +ssh://svn@.../ooo/cws/kohei03'
> > > svn+ssh://svn@.../ooo/cws/kohei03:  (Not a valid
> > > URL)
> > >
> > > svn: A problem occurred; see other errors for details
> > > cws: ERROR: The subversion command line client failed with exit status
> > > '1'
> >
> > Weird. Why should that be an invalid URL? Maybe some SSH problems?
>
> Not as far as I know.  As an example, I can check out existing CWS'es
> via ssh tunnel just fine.  And I hadn't had any problems with CWS'es in
> the past either.

Ok.  I've tried it on a different machine with svn 1.5.5, and it worked.
The version of svn that failed is 1.6.2.  So, apparently it was because
of the svn client.

Thanks for your help, Daniel.

Kohei

--
Kohei Yoshida - OpenOffice.org Engineer - Novell, Inc.
<kyoshida@...>


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Niklas Nebel]

On 06/11/09 05:41, Zhi Yu Yue wrote:
> I'm new here, so I list my ToDos as below. Please help to advise.
> 1. Send our UX spec on user experience community for review.
>   - I compared Kohei's list with mine. Most are different and should be
> fine. One may have some overlap. Kohei, I would appreciate if you can
> also review the UX spec after I post it. We can solve the conflict at
> the UX level.
> 2. Check in the code after UX spec is agreed. I may have more questions
> at that time.

That's a good plan for the UI changes.

One thing that doesn't need UX review is the performance improvements.
Can you separate the performance changes from the rest? Then we could
work to integrate them already in parallel with UX discussion.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Zhi Yu Yue]

Hi Niklas,

I need to check my colleague (Xu Ming Wang) if we can separate the code out. It is possibly a little difficult, but I'm not quite sure. I will update you later once I get the result. However I think the least we can do is to post some of our idea/or code here so that we can start some discussion. How do you think?
 
Best Regards,
Helen Yue
_________________________________________________________
Yue Zhi Yu  |  Manager of User Experience & Applications, Lotus Symphony |  yuezhiyu@...

From:	Niklas Nebel <Niklas.Nebel@...>
To:	dev@...
Date:	06/11/2009 04:04 PM
Subject:	Re: [sc-dev] about DataPilot
Sent by:	Niklas.Nebel@...

---

On 06/11/09 05:41, Zhi Yu Yue wrote:
> I'm new here, so I list my ToDos as below. Please help to advise.
> 1. Send our UX spec on user experience community for review.
>   - I compared Kohei's list with mine. Most are different and should be
> fine. One may have some overlap. Kohei, I would appreciate if you can
> also review the UX spec after I post it. We can solve the conflict at
> the UX level.
> 2. Check in the code after UX spec is agreed. I may have more questions
> at that time.

That's a good plan for the UI changes.

One thing that doesn't need UX review is the performance improvements.
Can you separate the performance changes from the rest? Then we could
work to integrate them already in parallel with UX discussion.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Niklas Nebel]

On 06/11/09 10:37, Zhi Yu Yue wrote:
> I need to check my colleague (Xu Ming Wang) if we can separate the code
> out. It is possibly a little difficult, but I'm not quite sure. I will
> update you later once I get the result. However I think the least we can
> do is to post some of our idea/or code here so that we can start some
> discussion. How do you think?

Yes, sure, that would help.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Jens-Heiner Rechtien]

Hi Kohei,

Kohei Yoshida wrote:

> On Wed, 2009-06-10 at 15:49 -0400, Kohei Yoshida wrote:
>> On Wed, 2009-06-10 at 19:50 +0200, Daniel Rentz wrote:
>>> Kohei Yoshida schrieb:
>>>> On Wed, 2009-06-10 at 14:51 +0000, Michael Meeks wrote:
>>>>> I've asked Kohei to create a CWS and whack the
>>>>> patches in there
>>>> There is one problem.  I can't seem to create a CWS.  I've tried:
>>>>
>>>> $ cws create DEV300 kohei03 -m m49
>>>> .. check cws path: 'svn
>>>> +ssh://svn@.../ooo/cws/kohei03'
>>>> svn+ssh://svn@.../ooo/cws/kohei03:  (Not a valid
>>>> URL)
>>>>
>>>> svn: A problem occurred; see other errors for details
>>>> cws: ERROR: The subversion command line client failed with exit status
>>>> '1'
>>> Weird. Why should that be an invalid URL? Maybe some SSH problems?
>> Not as far as I know.  As an example, I can check out existing CWS'es
>> via ssh tunnel just fine.  And I hadn't had any problems with CWS'es in
>> the past either.
>
> Ok.  I've tried it on a different machine with svn 1.5.5, and it worked.
> The version of svn that failed is 1.6.2.  So, apparently it was because
> of the svn client.

Um, not good. Apparently they changed some detail in the SVN client.
Care to file a bug against cws.pl assigned to me? Probably can't fix it
immediately, because of a vacation, but I will do so after my return (
July, 6th)

Heiner


--
Jens-Heiner Rechtien
rechtien@...

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Niklas Nebel]

On 06/11/09 10:37, Zhi Yu Yue wrote:
> I need to check my colleague (Xu Ming Wang) if we can separate the code
> out. It is possibly a little difficult, but I'm not quite sure. I will
> update you later once I get the result. However I think the least we can
> do is to post some of our idea/or code here so that we can start some
> discussion. How do you think?

Are there any updates on this yet?
Like you say, it would already help to have a rough overview about your
performance changes.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Zhi Yu Yue]

It spent a little more time than I thought to organize it. Where should I post it? should I create a new wiki page? Thanks.
 
Best Regards,
Helen Yue
_________________________________________________________
Yue Zhi Yu  |  Manager of User Experience & Applications, Lotus Symphony |  yuezhiyu@...

From:	Niklas Nebel <Niklas.Nebel@...>
To:	dev@...
Date:	06/18/2009 11:58 PM
Subject:	Re: [sc-dev] about DataPilot
Sent by:	Niklas.Nebel@...

---

On 06/11/09 10:37, Zhi Yu Yue wrote:
> I need to check my colleague (Xu Ming Wang) if we can separate the code
> out. It is possibly a little difficult, but I'm not quite sure. I will
> update you later once I get the result. However I think the least we can
> do is to post some of our idea/or code here so that we can start some
> discussion. How do you think?

Are there any updates on this yet?
Like you say, it would already help to have a rough overview about your
performance changes.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...

[Niklas Nebel]

On 06/19/09 10:58, Zhi Yu Yue wrote:
> It spent a little more time than I thought to organize it. Where should
> I post it? should I create a new wiki page? Thanks.

Yes, a wiki page would be good.

Niklas

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@...
For additional commands, e-mail: dev-help@...