Val 2.0?

[Ruslan Zasukhin]

on 2/16/04 2:22 PM, Andrew Sinning at andrew at learningware.com wrote:

> Have you produced a feature list for 2.0 yet?  What's your production
> timeline?

Hi Andrew,

We work now on V4RB and V4MD porting to 2.0 API.
Also we must resolve for such plugin products problem of
Mixing of LOCAL and Client SDKS.

V4MD almost done by functions.
V4RB yet require RB DB API support, and I think about adding into V4RB new
classes as BitSet, ArraySet, may be usage of interfaces, must be exceptions.
V4RB many work yet in the plugin itself.

Having these 2 products we will get one more direction of testing of new
kernel. All old examples must work.


-------------
New features in 2.0?  :-)

I will try from head make some list:

* we have re-write engine TOTALLY FROM ZERO.
    This process is going already 1.5 years actually.
    we have use all known to me modern C++ techniques and conceptions.

* [NEW] Cache. Works in 200-500 times faster of Valentina 1.x cache.
        Especially if big size.

* [NEW] Internal File system.
            works in 200-300 times faster of Valentina 1.x for BIG files.

* [NEW] SQL parser built with industry strength parser generator ANTLR.

* [NEW] SQL92 grammar + Valentina extensions to standard.
       huge part of SQL. Grammar is > 1000 lines

       -- ALTER TABLE much improved.
       -- CONSTRAINTS on fields and Tables.
                PRIMARY KEY, FOREIGN KEY, CHECK, DEFAULT, .
       -- UNION, INTERSECT, EXCEPT
       -- standard syntax of SQL functions CASE, TRIM, ...
       -- ...  

* [NEW] improved and added new functions,
        COOL: functions to work on related table

* [NEW] Unicode support, based on IBM ICU library.
        Valentina engine works on UTF16 now.

* [NEW] RAM based Database (!!!)
        and RAM based Table inside of Disk Based database.

* [NEW] Non indexed search on fields.

* [NEW] Binary link. New secret feature of Valentina.
        not implemented yet.

* [NEW] XML dump/load now can work on single Table.
        it can load data in Table/Database that already have data.

* [NEW] Valentina now will have own plugin system.
        we have made it and test it, but not yet put parts
        of Valentina on it. Probably not in 2.0

* [NEW] Indexes must be faster in few times on the search of single key.
        also there is hope to get faster sorting.
        I am not sure if I will be in time with one more
        interesting algorithm of work on indexes. In worse case
        will be added into 2.x

* [NEW] will be introduced new .tmp disk file for TMP files.
        this add more safety, and much faster db open after crash.
   
* [NEW] Now we have here huge set of tests directly in our C++ sources.
        Size of tests probably 50% of all our sources.
        This give us way do good testing of engine on each platform
        and on each compiler we support.
        Tests are part of engine itself.

    
-- 
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
-------------------------------------------------------------