[1/7] from: ammoncooke::yahoo::com at: 5-Sep-2001 17:40
Hi, I am creating a database program for a client. This client is currently using M$ SQL Server. I know REBOL will interface with M$SQl, but I want to know if any one here has had expeience with both M$SQL, & a Database written in REBOL. What is the speed comparison? Is REBOL viable database Server? Thanks!! Ammon
[2/7] from: tim:johnsons-web at: 5-Sep-2001 18:19
On Wed, Sep 05, 2001 at 05:40:12PM -0700, Ammon Cooke wrote:
> Hi, > > I am creating a database program for a client. >This client is currently using M$ SQL Server. >I know REBOL will interface with M$SQl, but I want to know if any >one here has had expeience with both M$SQL, & a Database written in REBOL.
I believe that /command is set up for that. With /core or /view, you might have to roll your own tcp protocol to query MS-SQL Server.
>What is the speed comparison? Is REBOL viable database Server?
I'm speaking from my own experience here.... but for what it is worth, I have found it a matter of scale. For small databases via cgi, I use rebol reading either delimited data exported from a commercial database manager or using rebol's load/save functions to maintain the data as native rebol block definitions. Rebol The Official Guide is heavily invested in that sort of strategy. For larger databases I use C/C++ with either compiled-in APIs or an external server. I would think that a well-written interface from rebol to ms-sql would be something to be proud of! tim
> Thanks!! > Ammon > > -- > To unsubscribe from this list, please send an email to > [rebol-request--rebol--com] with "unsubscribe" in the > subject, without the quotes.
-- Tim Johnson <[tim--johnsons-web--com]> http://www.johnsons-web.com
[3/7] from: ric:getorbital at: 5-Sep-2001 18:55
Hi Ammon, IMHO, the answer to your question depends a lot on the size and complexity of the database, and the risks to it's integrity. On one end, if the database only has a few tables, they are on the order of thousands of rows, and no one will access the database _except_ through your program, then REBOL is better than MSSQL. It is less overhead, easier to write and just as fast (faster if the database can reside in RAM). On the other end, if the database has, say, 100 tables, some of them contain hundreds of thousands of rows, people need ad hoc reporting, and the integrity of the database is critical to the business, then MSSQL is a better choice. A third alternative is MySQL, which is simpler and just as fast as MSSQL, but doesn't provide as good of protection of the integrity of the data as MSSQL. Lately, I've been developing with REBOL and MySQL. The combination is dynamite! DocKimbel's Rebsite offers an excellent library for working with the two. -- Ric
[4/7] from: tim:johnsons-web at: 5-Sep-2001 20:01
On Wed, Sep 05, 2001 at 06:55:20PM -0700, Ric Gagliardi wrote:
> Lately, I've been developing with REBOL and MySQL. The combination is > dynamite! DocKimbel's Rebsite offers an excellent library for working with > the two.
Hey Ric! Where is that Rebsite? thnx tj
[5/7] from: ric:getorbital at: 5-Sep-2001 20:28
DocKimbel's Rebsite is: http://rebol.dhs.org/index.r. The webpage for the MySQL driver is: http://rebol.dhs.org/mysql/main.html I don't have a whole lot of experience with it yet, but so far it's gone without a hitch. -- Ric
[6/7] from: john:thousand-hills at: 6-Sep-2001 0:01
It works, it is name-valued and so far is importable to the more serious databases. Bottom line, there is no "server" bogging down the services and it is simple and easy to program. Its a choice of need and cost. I use it professionally for small applications, like a "scripts library" that our programmers share for commonly used code. John At 08:28 PM 9/5/2001 -0700, you wrote:
[7/7] from: coussement:c:js:mil:be at: 6-Sep-2001 9:48
> I am creating a database program for a client. This client is > currently using M$ SQL Server. I know REBOL will interface with M$SQl, > but I want to know if any one here has had expeience with both M$SQL, & a > Database written in REBOL.
[ I am using the two for one year know, in our Computer Aided Psychotechnical Testing system. I found out the two are complementary: I use native REBOL blocks for storing work data on local station or distant servers, and when those have been handled, I transfer them to SQL Server for archive and data transmission to other projects. The addressing of native REBOL blocks is _a lot_ easier to manage that a distant access to a SQL Server - just think about the network problems that always exists ... The system works fine: we're talking about more than 2000000 - two millions - rows which have be treated here ! BTW, I've now design a new OODB system in REBOL, which we're currently testing. This will replace the native blocs in text files. ]
> What is the speed comparison? Is REBOL viable database Server? > > [
REBOL can be very fast in finding back some data into list or blocks. I didn't ever have a speed problem, even with the volume we are confronted to. I just think the difficulty should be implementing data integrity in REBOL, such as this offered in SQLServer. The language can do it, but it's a relative complex matter - just think about the lock policy ... Hopes this help ==christophe ]