A question about Time format

[Zhu Lihua-3]

Hi all,

There is a time format: [HH]:mm:ss

I found it has a value scope from -49710 to 49710, which means from
Nov/23/1763 00:00:00 to Feb/6/2036 00:00:00. But this format show hour
value as hours passed since Dec/30/1899, So it's value is 1193040:00:00.

Input 49711 with this time format will display "###" in the cell. I
guess it must be out side of the permitted scope. But why it has this
limitation? I guess it must have something to do with the storage of the
time in the memory. I wonder how it stored? and why it has this
limitation? Could anyone give me some hint?

Thanks and best regards,
--
Lihua


北京红旗贰仟软件技术有限公司
Beijing Redflag 2000 Software Co., Ltd.
Building No.2, Block A, Huilongsen, 18 Xihuan Nanlu
Beijing Economic-Technological Development Area
Beijing - P.R.China

Tel：[phone] ext.6141
http://www.RedOffice.com



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

[Joost Andrae]

Hi,

> There is a time format: [HH]:mm:ss

this time format allows to count hours > 24 hours

A1      03:00
A2      23:00
A3      =sum(a1:a2)  ==> 02:00:00

if you format A3 as [HH]:mm:ss then A3 gives 26:00:00   (26 hours)

>
> I found it has a value scope from -49710 to 49710, which means from

the value scope must be higher because I can create a date of 2036-10-10
  having the date value of 49958

> Nov/23/1763 00:00:00 to Feb/6/2036 00:00:00. But this format show hour
> value as hours passed since Dec/30/1899, So it's value is 1193040:00:00.
>

the default value for the zero point for date values is 1899-12-30. Calc
can handle both negative and positive values for a date value. For the
value -1 Calc would display 1899-12-29. The default value can be changed
(in /tools/options/calc/calculate)  to 1900-1-1 (to be compatible to
StarCalc 1.0 documents) and to 1904-1-1 to stay compatible to older
Excel documents.

> Input 49711 with this time format will display "###" in the cell. I
> guess it must be out side of the permitted scope. But why it has this
> limitation? I guess it must have something to do with the storage of the
> time in the memory. I wonder how it stored? and why it has this
> limitation? Could anyone give me some hint?

change the cell width. The value just doesn't fit into the cell boundaries.

Kind regards, Joost

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

[Zhu Lihua-3]

Hi Joost,

Thank you very much for the reply.

I tried in OOo-DEV3.2 just now, input 49958 in a cell and format it to
that format will show "###", and change the width doesn't work.

Best regards,
Lihua

在 2009-10-14三的 13:07 +0200，Joost Andrae写道：

> Hi,
>
> > There is a time format: [HH]:mm:ss
>
> this time format allows to count hours > 24 hours
>
> A1      03:00
> A2      23:00
> A3      =sum(a1:a2)  ==> 02:00:00
>
> if you format A3 as [HH]:mm:ss then A3 gives 26:00:00   (26 hours)
>
> >
> > I found it has a value scope from -49710 to 49710, which means from
>
> the value scope must be higher because I can create a date of 2036-10-10
>   having the date value of 49958
>
> > Nov/23/1763 00:00:00 to Feb/6/2036 00:00:00. But this format show hour
> > value as hours passed since Dec/30/1899, So it's value is 1193040:00:00.
> >
>
> the default value for the zero point for date values is 1899-12-30. Calc
> can handle both negative and positive values for a date value. For the
> value -1 Calc would display 1899-12-29. The default value can be changed
> (in /tools/options/calc/calculate)  to 1900-1-1 (to be compatible to
> StarCalc 1.0 documents) and to 1904-1-1 to stay compatible to older
> Excel documents.
>
> > Input 49711 with this time format will display "###" in the cell. I
> > guess it must be out side of the permitted scope. But why it has this
> > limitation? I guess it must have something to do with the storage of the
> > time in the memory. I wonder how it stored? and why it has this
> > limitation? Could anyone give me some hint?
>
> change the cell width. The value just doesn't fit into the cell boundaries.
>
> Kind regards, Joost
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@...
> For additional commands, e-mail: dev-help@...
>

--
Lihua


北京红旗贰仟软件技术有限公司
Beijing Redflag 2000 Software Co., Ltd.
Building No.2, Block A, Huilongsen, 18 Xihuan Nanlu
Beijing Economic-Technological Development Area
Beijing - P.R.China

Tel：[phone] ext.6141
http://www.RedOffice.com



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

[Uwe Altmann-3]

Hi

Zhu Lihua schrieb:
> I tried in OOo-DEV3.2 just now, input 49958 in a cell and format it to
> that format will show "###", and change the width doesn't work.

Using 3.2m1 on a Mac, this gives me 00:00:00 when formating as HH:MM:SS
and 10.10.2036 when using DD.MM.YYYY as format. Changing row width works
well too in both ways (changing the numeric value in cotext menu or
dragging cell limiter with mouse).
--
Mit freundlichen Grüßen

Uwe Altmann

OpenOffice.org auf dem Mac:
http://porting.openoffice.org/mac/

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

[Zhu Lihua-3]

Is there any reasonable explanation in technique? Or is it a bug?

在 2009-10-14三的 17:48 +0800，Zhu Lihua写道：

> Hi all,
>
> There is a time format: [HH]:mm:ss
>
> I found it has a value scope from -49710 to 49710, which means from
> Nov/23/1763 00:00:00 to Feb/6/2036 00:00:00. But this format show hour
> value as hours passed since Dec/30/1899, So it's value is 1193040:00:00.
>
> Input 49711 with this time format will display "###" in the cell. I
> guess it must be out side of the permitted scope. But why it has this
> limitation? I guess it must have something to do with the storage of the
> time in the memory. I wonder how it stored? and why it has this
> limitation? Could anyone give me some hint?
>
> Thanks and best regards,

--
Lihua


北京红旗贰仟软件技术有限公司
Beijing Redflag 2000 Software Co., Ltd.
Building No.2, Block A, Huilongsen, 18 Xihuan Nanlu
Beijing Economic-Technological Development Area
Beijing - P.R.China

Tel：[phone] ext.6141
http://www.RedOffice.com



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

[Regina Henschel]

Hi Lihua,

Zhu Lihua schrieb:

> Hi all,
>
> There is a time format: [HH]:mm:ss
>
> I found it has a value scope from -49710 to 49710, which means from
> Nov/23/1763 00:00:00 to Feb/6/2036 00:00:00. But this format show hour
> value as hours passed since Dec/30/1899, So it's value is 1193040:00:00.
>
> Input 49711 with this time format will display "###" in the cell. I
> guess it must be out side of the permitted scope. But why it has this
> limitation? I guess it must have something to do with the storage of the
> time in the memory. I wonder how it stored? and why it has this
> limitation? Could anyone give me some hint?

I'm not sure, because I'm no developer. As far as I know, Eike maintains
the number formatter. Perhaps ask him?

I found this:

The input number is converted to seconds and this seconds are limited to
  0xffffffff (=4294967295). But I do not know why they are limited.

You find it in /svtools/source/numbers/zformat.cxx.
The constant is in line 71
        const double _D_MAX_U_LONG_ = (double) 0xffffffff;
Line 2486 generates the ###
        if( floor( fTime ) > _D_MAX_U_LONG_ )
         {
           OutString = rScan.GetErrorString();
           return FALSE;
         }

kind regards
Regina

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

[Niklas Nebel]

On 10/16/09 11:56, Regina Henschel wrote:

> I found this:
>
> The input number is converted to seconds and this seconds are limited to
>  0xffffffff (=4294967295). But I do not know why they are limited.
>
> You find it in /svtools/source/numbers/zformat.cxx.
> The constant is in line 71
>     const double _D_MAX_U_LONG_ = (double) 0xffffffff;
> Line 2486 generates the ###
>     if( floor( fTime ) > _D_MAX_U_LONG_ )
>         {
>           OutString = rScan.GetErrorString();
>           return FALSE;
>         }

Just look one line below that. :-)
The time calculation (hours, minutes) is done in 32-bit unsigned
integer. It's a limitation, but there isn't really an immediate need to
change it.

Niklas

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

[Regina Henschel]

Hi,

Niklas Nebel schrieb:

> On 10/16/09 11:56, Regina Henschel wrote:
>> I found this:
>>
>> The input number is converted to seconds and this seconds are limited
>> to  0xffffffff (=4294967295). But I do not know why they are limited.
>>
>> You find it in /svtools/source/numbers/zformat.cxx.
>> The constant is in line 71
>>     const double _D_MAX_U_LONG_ = (double) 0xffffffff;
>> Line 2486 generates the ###
>>     if( floor( fTime ) > _D_MAX_U_LONG_ )
>>         {
>>           OutString = rScan.GetErrorString();
>>           return FALSE;
>>         }
>
> Just look one line below that. :-)
> The time calculation (hours, minutes) is done in 32-bit unsigned
> integer. It's a limitation, but there isn't really an immediate need to
> change it.

http://www.openoffice.org/issues/show_bug.cgi?id=63225

kind regards
Regina

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

[Zhu Lihua-3]

Hi Regina,

Thank you very much. So the number in the cell stands for the days after
certain date. And 49711*86400 = 4295030400, it bigger than
_D_MAX_U_LONG_, So it shows "###". And the other time format needn't
consider the integer portion of the number, so they have no such
limitation.

Thanks again.

Best regards,
Lihua

在 2009-10-16五的 11:56 +0200，Regina Henschel写道：

> Hi Lihua,
>
> Zhu Lihua schrieb:
> > Hi all,
> >
> > There is a time format: [HH]:mm:ss
> >
> > I found it has a value scope from -49710 to 49710, which means from
> > Nov/23/1763 00:00:00 to Feb/6/2036 00:00:00. But this format show hour
> > value as hours passed since Dec/30/1899, So it's value is 1193040:00:00.
> >
> > Input 49711 with this time format will display "###" in the cell. I
> > guess it must be out side of the permitted scope. But why it has this
> > limitation? I guess it must have something to do with the storage of the
> > time in the memory. I wonder how it stored? and why it has this
> > limitation? Could anyone give me some hint?
>
> I'm not sure, because I'm no developer. As far as I know, Eike maintains
> the number formatter. Perhaps ask him?
>
> I found this:
>
> The input number is converted to seconds and this seconds are limited to
>   0xffffffff (=4294967295). But I do not know why they are limited.
>
> You find it in /svtools/source/numbers/zformat.cxx.
> The constant is in line 71
> const double _D_MAX_U_LONG_ = (double) 0xffffffff;
> Line 2486 generates the ###
> if( floor( fTime ) > _D_MAX_U_LONG_ )
>          {
>            OutString = rScan.GetErrorString();
>            return FALSE;
>          }
>
> kind regards
> Regina
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@...
> For additional commands, e-mail: dev-help@...
>

--
Lihua


北京红旗贰仟软件技术有限公司
Beijing Redflag 2000 Software Co., Ltd.
Building No.2, Block A, Huilongsen, 18 Xihuan Nanlu
Beijing Economic-Technological Development Area
Beijing - P.R.China

Tel：[phone] ext.6141
http://www.RedOffice.com



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