From: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
---|---|
To: | Vivek Khera <khera(at)kcilink(dot)com> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Mysql to postgres tools -reviews? |
Date: | 2001-10-10 01:17:05 |
Message-ID: | 200110100117.f9A1H5K23983@candle.pha.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
I checked this list with /contrib/mysql and it seems most of these items
are covered in the conversion script. If you find any that need adding,
please contact the author. Thanks.
> >>>>> "GW" == Graham White <gawhite7(at)home(dot)com> writes:
>
> GW> I am looking for feedback from anyone that has moved their data from
> GW> Mysql to Postgres using the tools in contrib/mysql or any other data
> GW> porting tools on the net. Did any data get lost in the transfer?
> GW> Recommendations? Amount of time for transfers? Any feedback would be
> GW> greatly appreciated. Thanks in advance.
>
> I didn't migrate any data, just schemas. The tools were all
> inadequate, including the ones I found on the web, let alone the ones
> in the contrib section.
>
> Here are my notes for converting the schemas I had; obviously, you
> need to apply application specific knowledge to select from the richer
> set of PG datatypes that you don't have in MySQL, and vice versa. For
> example, in MySQL, I used a hand-packed unsigned integer to hold IP
> addresses (my app would pack/unpack the dotted quads) but in PG, I use
> the inet data type.
>
> The short of it is that you don't want to use a totally automated tool
> for this; you need to use your noodle and make intelligent choices of
> your available data types.
>
> --cut here--
> change # comments to -- comments
>
> change "int(10)" to "integer"
>
> change "integer unsigned NOT NULL auto_increment" fields to type
> "SERIAL PRIMARY KEY"
>
> change "integer unsigned" to "integer check(colname >= 0)"
>
> change "mediumtext" => "text"
>
> remove "Type = XXX" table option at end of create statements.
>
> change "timestamp(14)" => "timestamp default 'NOW()'" .. probably need
> trigger to update this kind of timestamp on every update as well.
>
> change "datetime" => "timestamp"
>
> move "KEY" indexes after column create with CREATE INDEX
>
> change "enum" fields to "varchar(X) check (fn in ('...','...'))"
>
> change "set" fields to some type of text field, and make app convert
> to/from comma separated string of values.
>
> change "tinyint(N)" to "smallint"
>
> change "UNIQUE KEY name" to "UNIQUE"
>
> delete "binary" attributes to char fields.
>
> delete any "default NULL" as they are redundant.
>
> "text NOT NULL" needs to have "default ''" for same behavior as in mysql.
>
>
> --
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Vivek Khera, Ph.D. Khera Communications, Inc.
> Internet: khera(at)kciLink(dot)com Rockville, MD +1-240-453-8497
> AIM: vivekkhera Y!: vivek_khera http://www.khera.org/~vivek/
>
> ---------------------------(end of broadcast)---------------------------
> TIP 5: Have you checked our extensive FAQ?
>
> http://www.postgresql.org/users-lounge/docs/faq.html
>
--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026
From | Date | Subject | |
---|---|---|---|
Next Message | Dr. Evil | 2001-10-10 02:49:00 | Data Design question: Storing formated documents |
Previous Message | Seung-won Hwang | 2001-10-10 01:06:37 | [Q] Index creation hangs |