DBD::CSV or DBD::DBM or something else?

[meneg]

Hello all and thanks in advance for any help you may give.

I'm currently in the process of making a cgi website for a family business
which would carry a few online orders and a few database entries. I soon
came to realize using postgresql - which I used last year when I was
learning php - would be an overkill because we talk about no more that a
thousand entries in a couple of small tables. Also since our hosting is
remote, I don't want to deal with remote databases. If I had it on my
linux machine, I wouldn't really care if it was an overkill or not.

Anyway, to the point. I read a a couple of decent perl books in the past
couple of months and I remember - not exactly where - that DBD::CSV is a
great alternative to a full featured SQL server or a hard to maintain
plain flat file. Also I recenty read somewhere DBD:BM to mentioned for
exactly the same motives.

I was reading the manpages of these two but I'm really not convinced what
I should use. I need any a advice you may give me. What made it worce, I
read that DBD:BM has several dbase types to cover (although one of them
is dominant with better support).

The database would only have a couple of tables with no more that a
thousand entries each but these should be of boolean, numeric and string
scalar values that (the string scalars) could do at least some thousand
words (keeping comments and stuff).

Any suggestion on the matter would be greatly appreciated,

- meneg.

 

[A. Sinan Unur]

Hello all and thanks in advance for any help you may give.

I'm currently in the process of making a cgi website for a family
business which would carry a few online orders and a few database
entries.

Strictly speaking, this is not a Perl question but ...

I soon came to realize using postgresql - which I used last
year when I was learning php - would be an overkill because we talk
about no more that a thousand entries in a couple of small tables.
Also since our hosting is remote, I don't want to deal with remote
databases.

Is this shared hosting? If so, and if the database is properly set up, it
would be much more secure to use Postgresql or MySQL.

Anyway, to the point. I read a a couple of decent perl books in the
past couple of months and I remember - not exactly where - that
DBD::CSV is a great alternative to a full featured SQL server or a
hard to maintain plain flat file.

The user id under which the web server is running will need to be able to
read the CSV files. That means, the whole world would need to be able
toread them. Putting financial information in those files is a sure
recipe for disaster.

Please note that your question is independent of the language you are
programming in.

Sinan.

 

[meneg]

The user id under which the web server is running will need to be able to
read the CSV files. That means, the whole world would need to be able
toread them. Putting financial information in those files is a sure
recipe for disaster.

Please note that your question is independent of the language you are
programming in.

Sinan.

Thanks for the response. Your feedback is very helpful and I see your
point - also your point that this is more of a general dbase question -.
However I still don't drop the question on file dbases because it more of
a "do you trust your webserver?" or a "do you trust your hosting
administrator?" question. And I mean someone would argue that properly
setting up those files to be treated by the system and the webserver is
not always worse than having a badly administered database. Of course in
"average" situations postgres or mysql would be more secure but it can be
argued that a file can as well be protected under good udministering.

-meneg