Thread: Re: [HACKERS] char types gone.

The Hermit Hacker wrote:

>> > 1. Regression tests make very extensive use of char16.  Seems to
>> > me that these should be changed to a name.  Same for tutorial and
>> > any other example references in the code.  Seem ok?
>>
>> ** down sinks the mast of the old postquel, there, can you see it blink
>> in the deep green water ? ** ** nostalgia on my face ** I think I forgot
>> to mention my nicest hobby, sailing **
>
>    Okay, this one has me...what benefit is there to char16 over
>char(16)?  Why, in the first place, was it implemented as char16 anyway?

char16 was there before varlena's got implemented, the basis for char(16).
There is no significant advantage of char16 over char(16) any more.

    As for "down sinks the mast"...so far as I've been able to tell,
we have *never* removed anything that wasn't replaced by a better way of
doing it (time travel with triggers)...we didn't leave anyone floundering
for a solution to it, we provided it...but now, its not forced on
*everyone*, just on those databases/tables that require it...

    Same functionality, better performance...

** up raises the ulralight carbon ultra high speed mast, I like that streamline,
of course for this the old wooden one had to go ;-) **
This was never intended as a critique, I think it is good that char2-16 goes.
Of course I stick to my real message of this thread:
    replace the char16 stuff in the tutorial and the regression tests with char(16)
    or even varchar(16) but not with name.

Andreas

On Mon, 23 Mar 1998, Zeugswetter Andreas wrote:

>     As for "down sinks the mast"...so far as I've been able to tell,
> we have *never* removed anything that wasn't replaced by a better way of
> doing it (time travel with triggers)...we didn't leave anyone floundering
> for a solution to it, we provided it...but now, its not forced on
> *everyone*, just on those databases/tables that require it...
>
>     Same functionality, better performance...
>
> ** up raises the ulralight carbon ultra high speed mast, I like that streamline,
> of course for this the old wooden one had to go ;-) **


Okay...just checking.  I've heard one major complaint about PostgreSQL in
the past concerning the fact that we are just gutting it and "taking the
Postgres out of PostgreSQL", which I feel is an unfair assessment...

As your analogy puts it above, we've replaced key components of the old
"wooden, three mast'r" with a more sleek and streamlined "carbon fiber
mast"...but that new mast holds the same sails, just gets us to the finish
line faster...

If anyone feels that we've removed something that was really unique and
not replacedit with something better/faster, or provided a work around,
please let us know...in most cases, there is alot of discussion both here,
and in a private list, over whether we go forward with a modification or
not, especially something that is part of the *base* structure of the
system.

We have an *Open* development forum here...if you don't like the route
something is (or appears to be) going, let us know...give us strong
arguements for/against...but don't wait until the discussion is done and
everyone is in agreement before popping up...:(