Valentina Server

[Ruslan Zasukhin]

On 6/25/04 2:19 PM, "olivier vidal" <vidal_olivier at yahoo.fr> wrote:

>> Yes, this should be different files.
>> 
>> Then you cannot get them corrupted. You will always just append info to
>> them. This operation cannot corrupt file.
>> 
> what do you tyhink of :
> 
> - Command writing in the "journal" file" that the operation N°XX,
> MethodName +"update", date XXXXX, time XXXXX, on the database 1,
> baseObject 1, BEGINS.
> - rs=cursor1.update()
> - database1.Bo1.flush
> - Command writing in the "journal" file" that the operation N°XX,
> MethodName +"update", date XXXXX, time XXXXX, on the database 1,
> baseObject 1, ENDS.
> 
> 
> - Command writing in the "journal" file" that the operation N°XX,
> MethodName +"update", date XXXXX, time XXXXX, on the database 2,
> baseObject 1, BEGINS.
> - rs=cursor1.update()
> - database2.Bo1.flush
> - Command writing in the "journal" file" that the operation N°XX,
> MethodName +"update", date XXXXX, time XXXXX, on the database 2,
> baseObject 1, ENDS.
> 
> 
> After a crash, and a restart of the software, we could, as it, to
> verify if in the data base 1 the last operation correctly ended. right
> ?

Oliver,

Yes, right, 

All this is good, Just remember
    the more you do flushed the slower it works.

In fact you try now to build some kind of transactions.
Or more correctly to say -- recovery system.

Not easy task.
Keep this in mind.


Also note, journal can be readable for human eye.
This is good for human.

But it can be in some very compact coded format
To eat less disk space and be faster.


-- 
Best regards,
Ruslan Zasukhin      [ I feel the need...the need for speed ]
-------------------------------------------------------------
e-mail: ruslan at paradigmasoft.com
web: http://www.paradigmasoft.com

To subscribe to the Valentina mail list go to:
http://lists.macserve.net/mailman/listinfo/valentina
-------------------------------------------------------------