I was wondering if the handling of binary data has come up on this list. I have just started looking at sqlite and really like what I see, but I was a bit confused at how binary data is handled. The philosophical decision was made to basically allow people to store what they want where they want regardless of type in any column they choose.
"SQLite seeks to overcome this misfeature by allowing you to store any kind of data into any kind of column and by allowing flexibility in the specification of datatypes." - taken the datatypes section. Now why do you have to go to the trouble of escaping? I understand from a technical standpoint that the \000 byte is treated special as the end of string, but why can't the column store how much data is actually in the column? I see how this could be a performance hit, but I tend to think its important to strive for correctness and more importantly a clean interface than to impose these kind of restrictions on the caller. I guess what I am asking for is transparent handling of binary data. You shouldn't have to escape anything, other clients shouldn't have to know you escaped anything and you shouldn't have to encode the data. The author states that strong typing found in most SQL engines is "an example of the implementation showing through into the interface." Isn't the way binary data handled in SQLite an example of this same tendency? thanks, --eric --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
