Hi!
On May 25, Hagen H?pfner wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> > No, it depends on the "pointer" size. (not C pointer, but a pointer
> > to a record - a record number for a fixed row_format). In MyISAM a
> > record cannot be shorter than a pointer size. Try to create, say, 8
> > decimal (1,0) fields, you'll see that every single one of then takes
> > only one byte.
>
> Ok, but now let's come back to my initial question ;-) How to map a
> decimal value mapped to a bit sequence ... is it similar to
> real/float/double (IEEE 754) or do you use another method? Where can I
> find a detailed documentation about at least: how many bytes are
> required for what kind of decimal? You mentioned (1,0) requires one
> byte ... I know (18,9) requires 8 bytes (one byte for 9 digits) and
> (20,10) requires 10 bytes. But what happens in between?
>
> I know this are a lot of questions, but I am really interested in
> understanding it (without analyzing all the source code in detail ;-)
> ) ...
Check the strings/decimal.c file in the source distribution - it
contains a comment with a description of the format.
Regards,
Sergei
--
__ ___ ___ ____ __
/ |/ /_ __/ __/ __ \/ / Sergei Golubchik <serg@stripped>
/ /|_/ / // /\ \/ /_/ / /__ MySQL AB, Senior Software Developer
/_/ /_/\_, /___/\___\_\___/ Osnabrueck, Germany
<___/ www.mysql.com
