Re: calendar-backend API: CCalendar's entries empty

2010-11-06 Thread Filip-M. Brinkmann
Hi Nicolai,

thanks for the valuable hint. The problem, as it turned out, was a
classical PEBCAK (*). In Qt Developer, I used user "developer" to deploy
on my N900. Therefore, this user's calendar was queried, returning zero
entries ;) 
So switching to "user" solved the problem.

However, thanks for your answer!

Best,

Filip
---
(*) Problem Exists Between Chair And Keyboard ;)


On Fri, 2010-11-05 at 17:22 +0100, Nicolai Hess wrote:
> 
> 
> 2010/11/5 Filip-M. Brinkmann 
> Hi all,
> 
> I'd like to code a small Todo-list widget, which shows all due
> to-dos on
> the desktop.
> I started today, but I cannot access the to-dos because the
> CCalendar
> instances don't return any entries.
> I have 2 Calendars on my N900 and I get two calendar instances
> back,
> which is fine. They return the correct names and attributes,
> but are
> otherwise unresponsive. ;)
> so, for instance the following code returns 0 as to-do count:
> 
> ...
> 
> QString output;
> int count=0;
> int error;
> 
> CMulticalendar* mcal = CMulticalendar::MCInstance();
> vector cals = mcal->getListCalFromMc();
> for(std::vector::iterator it = cals.begin(); it !=
> cals.end(); ++it){
> // here, I get those 2 calendars correctly
> 
> // next line sums up to 0 all the time
> count +=
> (dynamic_cast(*it))->getTodos(error).size();
> 
> // next line shows the calendar names
> cerr << (dynamic_cast(*it))->getCalendarName();
> }
> 
> output= QString("Found %1 todos.").arg(count);
> // count is 0 here, although I have like 20 todos in one
> calendar
> return output;
> 
> --
> Same behaviour shows on getEvents() and so on...
> 
> Since documentation is...ehm... not very complete:
> does anyone know what happens there?
> 
> cheers,
> 
> filip
> 
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers 
> 
> 
> For debugging purpose you can create an empty file:
> /home/user/.calendar/log
> (reboot may necessary)
> 
> This file indicates the calendar-backend to log  ALL sql and results
> as  syslog messages.
> Now you can see which sql and parameters are called for example to
> retrive all todos with
> "getTodos()"
> 
> Btw pay attention, that methods like this one:
>   (dynamic_cast(*it))->getTodos(error).size();
> 
> create a vector with dynamically allocated objects, and you have
> to free them.
> 
> regards
> nicolai
> 
> 
> 
> 
> 
> 
> 



signature.asc
Description: This is a digitally signed message part
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend API: CCalendar's entries empty

2010-11-05 Thread Nicolai Hess
2010/11/5 Filip-M. Brinkmann 

> Hi all,
>
> I'd like to code a small Todo-list widget, which shows all due to-dos on
> the desktop.
> I started today, but I cannot access the to-dos because the CCalendar
> instances don't return any entries.
> I have 2 Calendars on my N900 and I get two calendar instances back,
> which is fine. They return the correct names and attributes, but are
> otherwise unresponsive. ;)
> so, for instance the following code returns 0 as to-do count:
>
> ...
>
> QString output;
> int count=0;
> int error;
>
> CMulticalendar* mcal = CMulticalendar::MCInstance();
> vector cals = mcal->getListCalFromMc();
> for(std::vector::iterator it = cals.begin(); it != cals.end();
> ++it){
> // here, I get those 2 calendars correctly
>
> // next line sums up to 0 all the time
> count += (dynamic_cast(*it))->getTodos(error).size();
>
> // next line shows the calendar names
> cerr << (dynamic_cast(*it))->getCalendarName();
> }
>
> output= QString("Found %1 todos.").arg(count);
> // count is 0 here, although I have like 20 todos in one calendar
> return output;
>
> --
> Same behaviour shows on getEvents() and so on...
>
> Since documentation is...ehm... not very complete:
> does anyone know what happens there?
>
> cheers,
>
> filip
>
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>


For debugging purpose you can create an empty file:
/home/user/.calendar/log
(reboot may necessary)

This file indicates the calendar-backend to log  ALL sql and results as
syslog messages.
Now you can see which sql and parameters are called for example to retrive
all todos with
"getTodos()"

Btw pay attention, that methods like this one:
  (dynamic_cast(*it))->getTodos(error).size();

create a vector with dynamically allocated objects, and you have
to free them.

regards
nicolai
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


calendar-backend API: CCalendar's entries empty

2010-11-05 Thread Filip-M. Brinkmann
Hi all,

I'd like to code a small Todo-list widget, which shows all due to-dos on
the desktop.
I started today, but I cannot access the to-dos because the CCalendar
instances don't return any entries.
I have 2 Calendars on my N900 and I get two calendar instances back,
which is fine. They return the correct names and attributes, but are
otherwise unresponsive. ;)
so, for instance the following code returns 0 as to-do count:

...

QString output; 
int count=0;
int error;

CMulticalendar* mcal = CMulticalendar::MCInstance();
vector cals = mcal->getListCalFromMc();
for(std::vector::iterator it = cals.begin(); it != cals.end(); 
++it){
// here, I get those 2 calendars correctly

// next line sums up to 0 all the time
count += (dynamic_cast(*it))->getTodos(error).size(); 

// next line shows the calendar names
cerr << (dynamic_cast(*it))->getCalendarName();
}

output= QString("Found %1 todos.").arg(count);
// count is 0 here, although I have like 20 todos in one calendar
return output;

--
Same behaviour shows on getEvents() and so on...

Since documentation is...ehm... not very complete: 
does anyone know what happens there?

cheers,

filip

___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


calendar backend, CMulticalendar and importIcsFileData

2009-12-11 Thread Pierre Amadio
Hello.

I am trying to understand how the calendar api works:
http://maemo.org/api_refs/5.0/beta/calendar-backend/

I have several ical calendar exported from zimbra that i wish to import
in a calendar on the n900.

I manage to make something that seems to work (see attachement), but i
realise i have a problem with at least one entry.

In the zimbra ical output, the event starts the 17 of december at 10am
GMY and ends at 11 am GMT.

DTSTART;TZID="(GMT) Greenwich Mean Time - Dublin / Edinburgh / Lisbon / Lond 
on":20091217T10
DTEND;TZID="(GMT) Greenwich Mean Time - Dublin / Edinburgh / Lisbon / London 
":20091217T11

I am based in France (GMT+1). If i look in this event once imported in
the calendar, it is schedule for the 17 december, from 10am to 11 am instead of
11am to 12am.

The events that were set on GMT+1 (as the following, wich is supposed to
take place an hour before the previous one) are correctly set once
imported. 

DTSTART;TZID="(GMT+01.00) Amsterdam / Berlin / Bern / Rome / Stockholm / 
Vienna":20091217T10
DTEND;TZID="(GMT+01.00) Amsterdam / Berlin / Bern / Rome / Stockholm / 
Vienna":20091217T11

Am i doing something wrong (like forgetting some option about timezone
settings), or is this a bug i should open on bugzilla.maemo.org ?

Aside from this, has anybody idea what is the difference between 
CMulticalendar::importIcsFileData
and
CMulticalendar::importIcsFileDataProgressive
?




#include 
#include 
#include 
#include 

int get_target_cal_id(CMulticalendar *multical, std::string calname){
  CCalendar *curcal;
  std::vector  calvec = multical->getListCalFromMc  ();
  int out=0;

  for ( size_t j = 0, size = calvec.size(); j < size; ++j ) {
curcal=calvec[j];
if (curcal->getCalendarName () == calname){
  out=curcal->getCalendarId ();
}
  }

  multical->releaseListCalendars(calvec);

  
  return out;
}

int create_cal(CMulticalendar *multical,std::string calname){
  cout <<"Creating calendar "<addCalendar(
			calname,
			COLOUR_RED,
			0,
			1,
			(CalendarType) 0,
			"whot",
			"whot",
			perror
			);
  
  if(newcal==NULL) {
cout << "Problem creating a new calendar: "<getCalendarId();
}

int import_ical(CMulticalendar *multical,std::string filename,int cal_id) {
  cout <<"importing "<importIcsFileData( filename,
			   cal_id,
			   eventid,
			   iDuplicateCountTask,
			   iDuplicateCountEvent,
			   iDuplicateCountJournal,
			   perror
			   );
  cout<<"perror="<deleteCalendar(target_id,perror)) {
  cout<< "Problem deleting calendar error code="<<  perror << "\n";
  exit(1);
}
cout << "Deleting existing calendar "<getCalendarById(target_id,perror);
  if(perror!=CALENDAR_OPERATION_SUCCESSFUL){
cout <<"Error getting calendar by id. error="<___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-29 Thread Benoît HERVIER
Hum ... and a python one ?

2009/6/13 Sebastian 'CrashandDie' Lauwers :
> It is very easy to write C++ code that is usable by C compilers.
> Considering there is apparently a notable demand for a C interface,
> pushing for an enhancement request early in the project may be an
> interesting thing to do. The code change is relatively minimal if
> architectured correctly and would be beneficial to quite a few people
> apparently.
>
> I'm not teaching any experienced coder anything new here, and I don't
> mean to preach to the choir, but for future reference:
>
> #ifdef __cplusplus
>  extern "C"
> #endif
> int foobar(int i);
>
> http://developers.sun.com/solaris/articles/mixing.html#cpp_from_c
>
>
> On 11/06/2009, Hubert Figuiere  wrote:
>> On 06/11/2009 05:01 PM, Jamie Bennett wrote:
>>> Forgive my ignorance but is it 10x longer to compile or 10x increment
>>> in speed? Compilation speed is a matter of grabbing two coffee's
>>> instead of one, run time speed is a matter of giving up completely.
>>
>> C code compiled by g++ isn't slower or faster. It should be the same.
>> (I'm talking runtime speed here)
>> Just because there seems to be another myth being entertained.
>>
>> Hub
>> ___
>> maemo-developers mailing list
>> maemo-developers@maemo.org
>> https://lists.maemo.org/mailman/listinfo/maemo-developers
>>
>
> --
> Sent from my mobile device
>
> question = ( to ) ? be : ! be;
>      -- Wm. Shakespeare
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>
>



-- 
Benoît HERVIER - http://khertan.net/
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-12 Thread Sebastian 'CrashandDie' Lauwers
It is very easy to write C++ code that is usable by C compilers.
Considering there is apparently a notable demand for a C interface,
pushing for an enhancement request early in the project may be an
interesting thing to do. The code change is relatively minimal if
architectured correctly and would be beneficial to quite a few people
apparently.

I'm not teaching any experienced coder anything new here, and I don't
mean to preach to the choir, but for future reference:

#ifdef __cplusplus
  extern "C"
#endif
int foobar(int i);

http://developers.sun.com/solaris/articles/mixing.html#cpp_from_c


On 11/06/2009, Hubert Figuiere  wrote:
> On 06/11/2009 05:01 PM, Jamie Bennett wrote:
>> Forgive my ignorance but is it 10x longer to compile or 10x increment
>> in speed? Compilation speed is a matter of grabbing two coffee's
>> instead of one, run time speed is a matter of giving up completely.
>
> C code compiled by g++ isn't slower or faster. It should be the same.
> (I'm talking runtime speed here)
> Just because there seems to be another myth being entertained.
>
> Hub
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>

-- 
Sent from my mobile device

question = ( to ) ? be : ! be;
  -- Wm. Shakespeare
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-11 Thread Hubert Figuiere
On 06/11/2009 05:01 PM, Jamie Bennett wrote:
> Forgive my ignorance but is it 10x longer to compile or 10x increment
> in speed? Compilation speed is a matter of grabbing two coffee's
> instead of one, run time speed is a matter of giving up completely.

C code compiled by g++ isn't slower or faster. It should be the same.
(I'm talking runtime speed here)
Just because there seems to be another myth being entertained.

Hub
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-11 Thread gary liquid
heh Jamie,

its 10x (might be slight exag) longer to compile.
i normally make code changes and compile the 2 or 3 modules I've been
working on.
its quick and simple.
with c++ even the simplest files are slow.
its like waiting for visual studio to run, an age.
and my method of on device compilation is FASTER for normal day to day
coding than scratchbox and copy and execute :)
its got obvious serious drawbacks tho lol

gary


On Thu, Jun 11, 2009 at 10:01 PM, Jamie Bennett wrote:
>
> On 11 Jun 2009, at 21:08, gary liquid wrote:
>
>> Hubert,
>>
>> Compiling my pure C in a C++ compiler is indeed possible.
>> However, it takes 10x longer on my tablet to compile even simple
>> programs using a c++ compiler.
>>
>> I don't mind if the libs aren't available, it was just a nicety for me
>> - there will be other libraries that are usable as required I am sure.
>
> Forgive my ignorance but is it 10x longer to compile or 10x increment in
> speed? Compilation speed is a matter of grabbing two coffee's instead of
> one, run time speed is a matter of giving up completely. (oh and compiling
> on the tablet is so 1980's ;) )
>
>> Gary
>
> Regards,
> Jamie.
> --
> http://www.linuxuk.org
>
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-11 Thread Jamie Bennett

On 11 Jun 2009, at 21:08, gary liquid wrote:

> Hubert,
>
> Compiling my pure C in a C++ compiler is indeed possible.
> However, it takes 10x longer on my tablet to compile even simple
> programs using a c++ compiler.
>
> I don't mind if the libs aren't available, it was just a nicety for me
> - there will be other libraries that are usable as required I am sure.

Forgive my ignorance but is it 10x longer to compile or 10x increment  
in speed? Compilation speed is a matter of grabbing two coffee's  
instead of one, run time speed is a matter of giving up completely.  
(oh and compiling on the tablet is so 1980's ;) )

> Gary

Regards,
Jamie.
--
http://www.linuxuk.org
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-11 Thread gary liquid
Hubert,

Compiling my pure C in a C++ compiler is indeed possible.
However, it takes 10x longer on my tablet to compile even simple
programs using a c++ compiler.

I don't mind if the libs aren't available, it was just a nicety for me
- there will be other libraries that are usable as required I am sure.

Gary


On Thu, Jun 11, 2009 at 5:15 PM, Hubert Figuiere wrote:
> On 06/10/2009 07:32 AM, Sampo Savola wrote:
>> There seems to be only C++ interface for this.
>  >
>> How should this be used in C application,
>> is there going to be C interface also?
>
> You can mix C and C++ easily. You can recompile your C code with a C++
> compiler[1].
>
>
> Hub
>
> [1] possibly with a few changes that will keep the code compilable with
> a C compiler. This was part of the C++ design.
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-11 Thread Jamie Bennett
On 11 Jun 2009, at 17:15, Hubert Figuiere wrote:

> On 06/10/2009 07:32 AM, Sampo Savola wrote:
>> There seems to be only C++ interface for this.
>>
>> How should this be used in C application,
>> is there going to be C interface also?
>
> You can mix C and C++ easily. You can recompile your C code with a C++
> compiler[1].

Anyway, C is a just a C++ revolutionists realization of a programming  
language. Some day they will come back and say, what was I thinking.

Hey I should print that on a t-shirt.

> Hub

Regards,
Jamie.
--
http://www.linuxuk.org
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-11 Thread Hubert Figuiere
On 06/10/2009 07:32 AM, Sampo Savola wrote:
> There seems to be only C++ interface for this.
 >
> How should this be used in C application,
> is there going to be C interface also?

You can mix C and C++ easily. You can recompile your C code with a C++ 
compiler[1].


Hub

[1] possibly with a few changes that will keep the code compilable with 
a C compiler. This was part of the C++ design.
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-11 Thread Soumya Bijjal
Hello!

I asked the guys responsible for calendar area.. seems that having a C
API is not very likely. Adding Ilias in CC who is the owner of this
area. 

-Soumya


On Wed, 2009-06-10 at 13:36 +0200, ext gary liquid wrote:
> Sampo,
> 
> I would also be interested in a C api to the same calendar backend.
> Obviously my calendar is very simplistic and sketch based, but it
> would be of obvious benefit to allow system calendar entries to be
> visible there :)
> 
> 
> Gary
> 
> 
> 
> On Wed, Jun 10, 2009 at 12:32 PM, Sampo Savola wrote:
> > Hey
> >
> > With Fremantle beta sdk there comes this, calendar-backend :
> > http://maemo.org/api_refs/5.0/beta/calendar-backend/
> >
> > There seems to be only C++ interface for this.
> >
> > How should this be used in C application,
> > is there going to be C interface also?
> >
> > //Sampo
> >
> > ___
> > maemo-developers mailing list
> > maemo-developers@maemo.org
> > https://lists.maemo.org/mailman/listinfo/maemo-developers
> >
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers

___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


Re: calendar-backend

2009-06-10 Thread gary liquid
Sampo,

I would also be interested in a C api to the same calendar backend.
Obviously my calendar is very simplistic and sketch based, but it
would be of obvious benefit to allow system calendar entries to be
visible there :)


Gary



On Wed, Jun 10, 2009 at 12:32 PM, Sampo Savola wrote:
> Hey
>
> With Fremantle beta sdk there comes this, calendar-backend :
> http://maemo.org/api_refs/5.0/beta/calendar-backend/
>
> There seems to be only C++ interface for this.
>
> How should this be used in C application,
> is there going to be C interface also?
>
> //Sampo
>
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers


calendar-backend

2009-06-10 Thread Sampo Savola
Hey

With Fremantle beta sdk there comes this, calendar-backend :
http://maemo.org/api_refs/5.0/beta/calendar-backend/

There seems to be only C++ interface for this.

How should this be used in C application,
is there going to be C interface also?

//Sampo

___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers