Re: [HACKERS] [GENERAL] workaround for lack of REPLACE()

Joe Conway wrote:
> I took Tom's advice and added wrapper functions around text_substr() and
> bytea_substr() to cover the 2 argument case.
>
> I also added tests to strings.sql to cover substr() on toasted columns
> of both text and bytea.
>

Please replace the original patch (substr.2002.08.14.1.patch) with the
attached. It includes everything from the previous one, plus newly
implemented builtin functions:

replace(string, from, to)
-- replaces all occurrences of "from" in "string" to "to"
split(string, fldsep, column)
-- splits "string" on "fldsep" and returns "column" number piece
to_hex(int32_num) & to_hex(int64_num)
-- takes integer number and returns as hex string

All previously discussed on the list; see thread at:
http://archives.postgresql.org/pgsql-hackers/2002-07/msg00511.php

Examples:

SELECT replace('yabadabadoo', 'ba', '123') AS "ya123da123doo";
ya123da123doo
---------------
ya123da123doo
(1 row)

select split('joeuser(at)mydatabase','@',1) AS "joeuser";
joeuser
---------
joeuser
(1 row)

select split('joeuser(at)mydatabase','@',2) AS "mydatabase";
mydatabase
------------
mydatabase
(1 row)

select to_hex(256::bigint*256::bigint*256::bigint*256::bigint - 1) AS
"ffffffff";
ffffffff
----------
ffffffff
(1 row)

Tests have been added to the regression suite.

Passes all regression tests. I've checked the strings.sql script in a
multibyte database and it works fine also. I'd appreciate a good look by
someone more familiar with multibyte related issues though.

If it is OK, I'd like to hold off on docs until this is committed and
after beta starts.

If there are no objections, please apply.

Thanks,

Joe