Thank you for your insight into the state of SQLite "Typing." I think this further exposes the need to establish our level of conformance to a pretty much pervasive database standard (SQL.) I'm not certain that IBM was the initiator of the term "Gray Area" but they certainly have greatly increased it's influence at least in the world of computers.
In every IBM manual I ever used there was always the Standard, white back grounded "Standard" text and a very large portion of "Gray Area," gray background text, commonly referred to as "Enhancements" or "Extensions." If one avoided the gray areas, theoretically the exercises undertaken were "Platform independent." Perhaps if we intend to venture beyond the "Standard" (SQL) it might be time to put a little prominent "Gray" on the Wiki Pages. And, as IBM did/does allow one to error in his/her own choice of death by "Enhancement." Personally I always have thought the int -> int = int can be an extreme pain in the ... But, I know about it and have adapted. The same can be said for "Dynamic Typing." I just say make that text very "Gray." And if possible "User Controlled" optional. >From my standpoint, I was merrily cruising along with the "old" SQLite 2.8.x implementation. But the switch to 3,x,x has been at best, "eventful." Seems with each new day I either find something old obscurely broken or yet another way to shoot myself in the foot with a 3.x.x "Enhancement." :-) Ah, life on the "bleeding edge"! BTW. On the Windose machine, footprint (.dll size): 2.8.x, 219K; 3.x.x, 245k (today) equals +36k, If you had yet another 36K what could you do? As modern PDA's now seem to have a minimum of 8M or so. And these kind of enhancement requests just keep coming. Fred > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Sent: Tuesday, November 01, 2005 10:53 AM > To: sqlite-users@sqlite.org > Subject: Re: [sqlite] Proposed 3.3.0 changes. Was: 5/2==2 > > > "Fred Williams" <[EMAIL PROTECTED]> wrote: > > Ah the sticky wicket that is "Type less" :-) We now have issues > > evolving as a direct result of that feature in our cute > little database. > > We now seem to have by backing into it: Really Restricted > Integer, Real, > > DateTime (sort of), and Text. BLOB and CLOB away at your own risk! > > > > Is it time to officially declare/fully support some Types > and clear the > > air? > > > > SQLite is not "type-less". It uses manifest or dynamic typing instead > of static-typing which is what most other SQL database engines use > (and the SQL standard specifies). Manifest or dynamic typing is a > superset of static typing. The use of static typing in the > SQL standard > is a bug in that standard, in my view, than unnecessarily complicates > and restricts what you can do with SQL. The original SQL standard > specifies static typing so that implementations can use fixed-size > records. Static typing is an artifact of the implementation showing > through into the interface. Static typing in SQL is designed not to > help the users of SQL databases, but rather to help the implementors > of SQL database engines. > > SQLite is the only SQL database engine that I am aware of that offers > dynamic typing. This is not going to change. > > The difficult faced by SQLite is not how to deal with a dynamically > typed language (that's easy) but how to deal with a dynamic typing > in a way that is backwards compatible with the (broken) static typing > behavior of SQL. That is much harder. But it is achievable, I think. > > > -- > D. Richard Hipp <[EMAIL PROTECTED]> >
