Re: [PHP4BETA] php and MS SQL Server....
| From: | Manuel Lemos | Date: | Thu, 02 Mar 2000 22:08:36 +0000 |
| Subject: | Re: [PHP4BETA] php and MS SQL Server.... | ||
| References: | 1 | Groups: | php.version4 |
| Request: | Send a blank email to php-version4+get-11406@lists.php.net to get a copy of this message | ||
Hello Stig,
On 01-Mar-00 20:10:17, you wrote:
>> Last November Stig approached to me when I announce that a database
>> abstraction layer package named Metabase needed beta testers. He told me
>> about PEAR and a database abstraction layer that was going to be
>> distributed with it.
>>
>> He also mentioned that his intention was to avoid to have yet another
>> database abstraction layer and asked me if he could see the package. Of
>> course I granted him and many other people access to the Metabase beta
>> archives, but I never heard from him since then.
>I was hoping Metabase would be what I was looking for, but it went way
>beyond that I had in mind for PHP's standard distribution.
Why? You never told me nothing like that. Would it be more appropriate to
talk to me to see what if we could solve your concerns instead of pursuing
your own way?
>> Later I was invited to contribute with my classes but that was it. Nothing
>> happened.
>You were invited, but it's not like I go around nagging people into
>contributing. If I don't hear anything, I assume that the person doesn't
>have the time, the interest or something else.
Hey, I publically accepted and I was told to wait. If you didn't got that
message, that is a different story.
>> I don't mind whether it is going to be or not be included in any code
>> repository or if people just approach to me just to "borrow" some ideas. I
>> just developed Metabase for use in my PHP work. If more people can use it,
>> fine, hopefully they will get back to me with improvement suggestions bug
>> reports or new drivers that I didn't have the opportunity to develop.
>Manuel, DB and Metabase are different beasts. DB doesn't attempt to
>solve everything that Metabase does. The reason? Since this thing is
>after all distributed as part of PHP, it should be as flexible as
>possible, and restrict users as little as possible. People should have
I see your point. Where do you see any lack of flexibility or restrictions
for the users? You never mentioned anything like that before to me.
>the option to not add all this complexity if they don't want it, so for
>DB I want layers of features that people can opt using. That said, there
>is nothing that prevents the cools things in Metabase from being
>implemented on top of DB. I just don't consider that as important as the
>basic thing.
But think about this, the features that Metabase provides are there because
they are highly desirable. If you intend to make a simpler thing, soon
you realize you need to improve it to solve common database programming
problems that have to be addressed at the DBMS interface level.
Trust me Metabase used to be much simpler but then I realized that it was
not sufficient to develop truely DBMS independent applications, those that
you don't have to touch a single line of code to have them work in
different DBMSs.
Anyway, I plan to split the database access code from the database
management code inside DBMS driver classes, but without affecting backward
compatibility. The idea is to reduce the size of the driver classes at
application run time.
If you have other objections, it won't help me if you don't explain them to
me so I could evolve Metabase in right direction.
>> I just hoped that if PEAR is not going to be a open repository like CPAN
>> where anybody can upload their code to avoid others unnecessarily
>> duplicating efforts, they stop announcing that it will be that way.
>I don't understand what makes you say this. PEAR is, and will continue
>being, an open repository, and I have never implied otherwise.
I was commenting on the idea that Rasmus passed to me about requirements of
code approval and only accepting contributions to be turned Open Source.
>> >> > Anyway, why weren't the authors of the existing PHP code repositories
>> >> > contacted first to check if they are willing to implement the things
>> >> > you
>>
>> >Maybe they should have been, but aren't these people on the list anyway?
>> >PEAR didn't just pop up over night, there has been talk about it for a
>> >while.
>>
>> This is the talk about it.
>There has been talk about this for a long time, from before it got its
>first name. Search the mailing list for "pear" and "pecl", and you'll
>see that there has been talk about it before. There has been lots of
I know. I meant the talk is not over yet.
>commits to pear in the CVS logs, and I've even posted the API draft for
>DB to the list several months ago.
I didn't get this. Where was it posted?
Regards,
Manuel Lemos
Web Programming Components using PHP Classes.
Look at: http://phpclasses.UpperDesign.com/?user=mlemos@acm.org
--
E-mail: mlemos@acm.org
URL: http://www.mlemos.e-na.net/
PGP key: http://www.mlemos.e-na.net/ManuelLemos.pgp
--