Postgres Pro
Downloads
Home   >   mailing lists

Thread: HISTORY addition

HISTORY addition

From
Tatsuo Ishii
Date:
Bruce,

Please add followings to HISTORY:

Formats with the correct number of columns for UNICODE in psql (Patrice)
Add ISO 8859-5,6,7,8 support (Tatsuo)
Optimize LIKE/ILIKE when using single-byte encodings (Tatsuo)

Also I suggest we note somewhere that now octet_length(text) returns
the length BEFORE compression for an incompatible change.
--
Tatsuo Ishii

Re: HISTORY addition

From
Bruce Momjian
Date:
> Bruce,
> 
> Please add followings to HISTORY:
> 
> Formats with the correct number of columns for UNICODE in psql (Patrice)
> Add ISO 8859-5,6,7,8 support (Tatsuo)
> Optimize LIKE/ILIKE when using single-byte encodings (Tatsuo)

Done.

> Also I suggest we note somewhere that now octet_length(text) returns
> the length BEFORE compression for an incompatible change.
  octet_length(text_col) now returns non-compressed length (Tatsuo, Bruce)

Done.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026

Re: HISTORY addition

From
Tatsuo Ishii
Date:
> > Thanks. But I did nothing for this. Please remove my name.
> 
> But I did very little code for it myself.  Just copied from another
> function.  Finding the problem was the majority of the work, and you did
> that.

Thanks.

> > BTW, are we going to make followings happen for 7.2?
> > 
> > * Add octet_length_server() and octet_length_client() (Thomas, Tatsuo)
> > * Make octet_length_client the same as octet_length()
>  
> No, we will leave octet_length() alone and revisit for 7.3.  It is a low
> priority and adding these would require initdb, and it is too late in
> the release cycle anyway.

Ok.
--
Tatsuo Ishii

Re: HISTORY addition

From
Bruce Momjian
Date:
> > > Also I suggest we note somewhere that now octet_length(text) returns
> > > the length BEFORE compression for an incompatible change.
> > 
> >    octet_length(text_col) now returns non-compressed length (Tatsuo, Bruce)
> > 
> > Done.
> 
> Thanks. But I did nothing for this. Please remove my name.

But I did very little code for it myself.  Just copied from another
function.  Finding the problem was the majority of the work, and you did
that.

> BTW, are we going to make followings happen for 7.2?
> 
> * Add octet_length_server() and octet_length_client() (Thomas, Tatsuo)
> * Make octet_length_client the same as octet_length()
No, we will leave octet_length() alone and revisit for 7.3.  It is a low
priority and adding these would require initdb, and it is too late in
the release cycle anyway.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026

Re: HISTORY addition

From
Tatsuo Ishii
Date:
> > Also I suggest we note somewhere that now octet_length(text) returns
> > the length BEFORE compression for an incompatible change.
> 
>    octet_length(text_col) now returns non-compressed length (Tatsuo, Bruce)
> 
> Done.

Thanks. But I did nothing for this. Please remove my name.

BTW, are we going to make followings happen for 7.2?

* Add octet_length_server() and octet_length_client() (Thomas, Tatsuo)
* Make octet_length_client the same as octet_length()
--
Tatsuo Ishii