AW: AW: filling a db on CD, what happens?
luis at anachreon.co.uk
Wed Feb 6 19:37:17 EST 2008
That'd be Structured Query Language, don't be that kind to it...
Andre Garzia wrote:
> I've been using both RDBMS and flat files. A quick opinion is that
> with SQL you don't need to reinvent the wheel and this is specially
> good when you're going to do cross references and searching. SQL
> stands for Simple Query Language, it's really easy to make queries
> with SQL.
> With flat files, you need to roll your own search routines and this
> can be a huge bottleneck on your code. If you go brute force looking
> every single record in your thousand+ database, you're sure that will
> take a while. If you decide to optimize your searches and build your
> own indexing and searching, then you'll spend a lot of time in there.
> So, for my projects, I settle like this: if there's heavy searching go
> with SQL, if it's just data storage and retrieval, then stack files
> will do.
> I must say, I am in love with stack files, they are so flexible!
> On 2/6/08, Tiemo Hollmann TB <toolbook at kestner.de> wrote:
>> Hi Derek,
>>> Like I said though, it depends entire upon the type of database you are
>>> working with. I don't use SQL as I find it to be rather bulky. I use
>>> Flat Files, and I write all of my own database handlers, so my coding is
>>> not dependent on SQL or some other type of database and it's limitations.
>> Interesting to read, did you ever compared the performance on "big" datas?
>> Up to now I havn't found a performing way how to loop through 20.000 records
>> in a flat file for generic selection with a search term compared with a
>> select - where. Looping through 20.000 records and searching for a sting
>> with lineoffset() to select a list of datas wasn't fast enough for me.
>> Perhaps you found a better trick?
>> use-revolution mailing list
>> use-revolution at lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your subscription preferences:
More information about the Use-livecode