Subject RE: [Firebird-general] Article
Author Steve Summers
|-----Original Message-----
|From: Ann W. Harrison [mailto:aharrison@...]
|Sent: Tuesday, June 22, 2004 04:40pm
|To: Firebird-general@yahoogroups.com; Firebird-general@yahoogroups.com
|Subject: Re: [Firebird-general] Article
|
|
|Steve,
|
| Other system (Postgres for one) allow adding user functions,
|but the mechanism is completely different from Firebird. I
|think that would be very interesting to write up. Will you
|contact Jon or shall I?
|
|Thanks,
|
|
|Ann

Before I agree to this, I need consensus that I'm the best candidate (or at
least, the best of the willing candidates, if any?) for this. (And of
course, I need a better idea of how big a project this is.)

I don't do much programming anymore- I design the features and UI stuff that
other people actually code. I've never had any need to learn (or had much
desire to learn) the nuances of COM/DCOM, multithreading, web application
programming, component creation (beyond little things like multi-line
buttons), etc. I'm a lot more interested in making useful, easy to use
application programs than figuring out how to implement multithreaded COM
interfaces. I LOVE to program, but my role as designer is a lot like walking
on one of those airport slidewalks- I can get where I want to go a lot
faster and with a lot less (personal) effort with all those motors helping
me along.

Anyway, a couple years ago, the combination of:

* Needing to understand InterBase better to try to figure out our corruption
problems,

* My frustration with the "Software Requirement Specifications in MS Word
Documents" approach to conveying my design work, and

* Wanting to spend some time programming again, before my skills completely
atrophied,

led me to develop a requirements and "issues" tracking database.

I chose Delphi because that's what we use, and Firebird because Interbase
6.0.1.6 was a dead end, and I didn't want to pay license fees to deploy it
company-wide. I learned a LOT in that process (it took about 300 hours,
spread out over about a year), and had a lot of fun. And now I have a MUCH
better understanding of IB/FB, SQL, UDFs, client/server programming, etc. I
think I know what I'm doing for reasonably simple stuff like this.

The problem is, as Nando pointed out, DDJ is probably not all that
interested in what is MAYBE "intermediate" level programming in Delphi for
Win32.

Now, as far as UDFs, the main things I did with them were blob functions for
searching. My Requirements database uses an RTF editor (WPTools). I tend to
use a LOT of italics and bold text, multiple font sizes, tables, etc, which
makes using "Containing" to find text strings pretty unreliable because of
all the formatting codes. So I got permission from Julian Ziersch (the
WPTools guy) to use his RTF to Plain Text conversion routine to make a UDF
that converts RTF blobs to text, and then searches the plaintext for a
specified string. (That code is in the version of FreeUDFLib that I provided
to Claudio, that's posted on his site.) I also played with using the
"Searchtext" function concept to do "weightings" - returning the number of
occurances of the text string in the blob field, for example.

This was all interesting to ME, but it wasn't really that technical.
Starting with the FreeUDFLib source code in Delphi made it pretty easy. Is
this DDJ article material? Does it show off Firebird in interesting ways?
I don't know.

If the answer is yes, and the deadline isn't too soon, and the project not
too large, and nobody more qualified than me wants to do it, I'll do it. I
know that's a lot of "ifs", but my guess is it's the same if's that are
stopping any of the other candidates from discussing doing it themselves.