compare xbase and sql

------=_NextPart_000_0084_01C1B6D3.D561DDC0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Once again hello...
=20
Is it true that finding and updating a record using an sql database is
faster that xbase?
=20
Paul

------=_NextPart_000_0084_01C1B6D3.D561DDC0--

compare xbase and sql



Hi Paul,

It depends on many factors, but mainly the size of your databases. In most
areas SQL is much faster, however if your SQL-database is transactional,
adding/updating a record in dbIII is probably faster. The main performance
differences become visible when you search a record not in the index but in
the text, or for all loops, for example when you need a complete resultset
out of a large database. When it comes to complex queries and aggregate
functions (group by) or sorting, dbIII is completely lost.

As an example I use SAPDB for my Miva ecommerce app (not Merchant) and
connect to a database with all Zipcodes in the USA (29000 records). The
simple query "select state, name, population from zipcodes where name LIKE
'_EW%' order by state" returns 222 results within 0.44 seconds, the second
search is there in 0.031sec (cached).

On the other hand, inserting these records (insert into zipcodes
values('....') took about half an hour (autocommit on). With dbIII this
would have been finished probably in less than 10minutes.

For small datasets (a couple of thousand records that don't need things like
grouping or order by), dbIII offers lots of advantages especially in terms
of ease of administration, and I think that very often SQL is an overkill
for normal webpublishing. For more sophisticated applications, analytical
functions, big databases, lots of connected users and/or higher security and
better data integrity SQL is a must. However, to the defense of dbIII one
can add that it depends greatly on your "application logic" and how you
structure the database. Using childdatabases and storing & reusing query
results can greatly improve dbIII's performance.

Markus



-----Original Message-----
From: [email protected] [mailto:[email protected]]On
Behalf Of Paul Berg
Sent: Friday, February 15, 2002 7:22 PM
To: Miva Users
Subject: [meu] compare xbase and sql


Once again hello...

Is it true that finding and updating a record using an sql database is
faster that xbase?

Paul

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (<A HREF =/"http://www.grisoft.com)/">http://www.grisoft.com)</A>.
Version: 6.0.323 / Virus Database: 180 - Release Date: 2/8/2002

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (<A HREF =/"http://www.grisoft.com)/">http://www.grisoft.com)</A>.
Version: 6.0.323 / Virus Database: 180 - Release Date: 2/8/2002

compare xbase and sql



Hey,

(dbase...sorry, not very onto it today. getting a little warm)
I've knoticed the MvQUERY aswell, what are the limitations? what can it do
and what can't it do?

Paul
----- Original Message -----
From: Markus Gieppner <[email protected]>
To: Paul Berg <[email protected]>; Miva Users <[email protected]>
Sent: Saturday, February 16, 2002 1:40 PM
Subject: RE: [meu] compare xbase and sql


> Hi Paul,
>
> It depends on many factors, but mainly the size of your databases. In most
> areas SQL is much faster, however if your SQL-database is transactional,
> adding/updating a record in dbIII is probably faster. The main performance
> differences become visible when you search a record not in the index but
in
> the text, or for all loops, for example when you need a complete resultset
> out of a large database. When it comes to complex queries and aggregate
> functions (group by) or sorting, dbIII is completely lost.
>
> As an example I use SAPDB for my Miva ecommerce app (not Merchant) and
> connect to a database with all Zipcodes in the USA (29000 records). The
> simple query "select state, name, population from zipcodes where name LIKE
> '_EW%' order by state" returns 222 results within 0.44 seconds, the second
> search is there in 0.031sec (cached).
>
> On the other hand, inserting these records (insert into zipcodes
> values('....') took about half an hour (autocommit on). With dbIII this
> would have been finished probably in less than 10minutes.
>
> For small datasets (a couple of thousand records that don't need things
like
> grouping or order by), dbIII offers lots of advantages especially in terms
> of ease of administration, and I think that very often SQL is an overkill
> for normal webpublishing. For more sophisticated applications, analytical
> functions, big databases, lots of connected users and/or higher security
and
> better data integrity SQL is a must. However, to the defense of dbIII one
> can add that it depends greatly on your "application logic" and how you
> structure the database. Using childdatabases and storing & reusing query
> results can greatly improve dbIII's performance.
>
> Markus
>
>
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]]On
> Behalf Of Paul Berg
> Sent: Friday, February 15, 2002 7:22 PM
> To: Miva Users
> Subject: [meu] compare xbase and sql
>
>
> Once again hello...
>
> Is it true that finding and updating a record using an sql database is
> faster that xbase?
>
> Paul
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (<A HREF =/"http://www.grisoft.com)/">http://www.grisoft.com)</A>.
> Version: 6.0.323 / Virus Database: 180 - Release Date: 2/8/2002
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (<A HREF =/"http://www.grisoft.com)/">http://www.grisoft.com)</A>.
> Version: 6.0.323 / Virus Database: 180 - Release Date: 2/8/2002
>
>
>

Downloading from Another Server



Hello!

I have the need to download a file from another server
(I am creating a trascation file on the server that I
need to download to my Miva server). The name of the
file is always the same. I would like the download to
occur when a particular file/page (.mv) is open (the
file displays a report that uses the information in
the file that I need to download).

Any help would be appreciated.

Thanks!
CA

__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com

Downloading from Another Server



I tried simply reading it via MvIMPORT (I have all of
the required permissions), but that didn't work.

--- Cas <[email protected]> wrote:
> Hello!
>
> I have the need to download a file from another
> server
> (I am creating a trascation file on the server that
> I
> need to download to my Miva server). The name of the
> file is always the same. I would like the download
> to
> occur when a particular file/page (.mv) is open (the
> file displays a report that uses the information in
> the file that I need to download).
>
> Any help would be appreciated.
>
> Thanks!
> CA
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Sports - Coverage of the 2002 Olympic Games
> http://sports.yahoo.com
>

Downloading from Another Server



Hi Cas,

Did you try MvCALL?

Markus



-----Original Message-----
From: [email protected] [mailto:[email protected]]On
Behalf Of Cas
Sent: Saturday, February 16, 2002 11:27 AM
To: [email protected]
Subject: Re: [meu] Downloading from Another Server


I tried simply reading it via MvIMPORT (I have all of
the required permissions), but that didn't work.

--- Cas <[email protected]> wrote:
> Hello!
>
> I have the need to download a file from another
> server
> (I am creating a trascation file on the server that
> I
> need to download to my Miva server). The name of the
> file is always the same. I would like the download
> to
> occur when a particular file/page (.mv) is open (the
> file displays a report that uses the information in
> the file that I need to download).
>
> Any help would be appreciated.
>
> Thanks!
> CA
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Sports - Coverage of the 2002 Olympic Games
> http://sports.yahoo.com
>

Downloading from Another Server



Thanks Mark! Yes, there is a problem when using
MvCALL. It seems the <newlines> are disregarded and
the file is read in as one big long record. This isn't
good as I have to process each record separately.

I tried doing an MvCALL, then MvExPORT, then MVImport
and the problem is still that their are no <newlines>
therefore only ONE big record.

I validated this by simply MvIMPORTING the file I am
trying to MvCALL (I just copied it to my server).
Everything works fine then, so I know the <newlines>
are in the file originally.


Any ideas?

--- Markus Gieppner <[email protected]> wrote:
> Hi Cas,
>
> Did you try MvCALL?
>
> Markus
>
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On
> Behalf Of Cas
> Sent: Saturday, February 16, 2002 11:27 AM
> To: [email protected]
> Subject: Re: [meu] Downloading from Another Server
>
>
> I tried simply reading it via MvIMPORT (I have all
> of
> the required permissions), but that didn't work.
>
> --- Cas <[email protected]> wrote:
> > Hello!
> >
> > I have the need to download a file from another
> > server
> > (I am creating a trascation file on the server
> that
> > I
> > need to download to my Miva server). The name of
> the
> > file is always the same. I would like the download
> > to
> > occur when a particular file/page (.mv) is open
> (the
> > file displays a report that uses the information
> in
> > the file that I need to download).
> >
> > Any help would be appreciated.
> >
> > Thanks!
> > CA
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Yahoo! Sports - Coverage of the 2002 Olympic Games
> > http://sports.yahoo.com
> >

compare xbase and sql



------_=_NextPart_000_01C1B78E.6FD73DF0
Content-Type: text/plain;
charset="iso-8859-1"

Strictly Speaking, SQL is a database manipulation language and Xbase is a
database file format.
It is actually possible to use SQL to manipulate an Xbase database if you
happen to be running on Windows using ODBC, and have some sort of program
that can handle SQL queries and access ODBC databases.
Miva actually can do this (on Windows), on the other hand manipulating
Xbase files with Miva's own database handling capabilities is probably
faster.
SQL is very much faster to write and understand than a miva program to do
the same thing, however speed will really depend on the specific action
you want to perform, and the specific SQL software that you are using.
The real advantage of SQL over Miva is not so much speed of execution, but
more the ability to perform complex database queries, particularly when
more than one table is involved (Miva 'databases' are actually individual
tables)
In SQL you can write a few lines which opens multiple database tables,
reads a set of records, and outputs the results into a new table, sorted
based on fields from several of the originating tables.
To do this purely with Miva could take dozens or hundreds of lines.


Paul Berg wrote:

> Once again hello...
>
> Is it true that finding and updating a record using an sql database is
> faster that xbase?
>
> Paul



------_=_NextPart_000_01C1B78E.6FD73DF0
Content-Type: text/plain; charset=us-ascii