On 24 Jul 2006, at 09:10, Richard Holland wrote:
From my understanding, if table bbb is a subclass of table main, then
the basic structure should look like this:
id_key references main(id_key)
bbb_id_key
gene
col1
primary key (id_key, bbb_id_key)
ie. it should have an FK that references main, include all the columns
from main, plus additional column(s) of its own which when combined
with
the FK create a unique primary key for bbb. Once all this is in place,
you can add extra columns to bbb such as col2.
in addition MEditor requires that at least one row is populated so
something
as follows works fine:
mysql> select * from Sc__main__main;
+--------+------+------+-------------------+
| id_key | gene | col1 | some_other_id_key |
+--------+------+------+-------------------+
| 2 | bla | 0 | 5 |
+--------+------+------+-------------------+
1 row in set (0.00 sec)
mysql> select * from Sc__bbb__main;
+--------+------+------+
| id_key | gene | col2 |
+--------+------+------+
| 2 | bla | 0 |
+--------+------+------+
1 row in set (0.00 sec)
cheers,
Richard
On Mon, 2006-07-24 at 09:11 +0200, Alexandre Gattiker wrote:
Hi Damian,
Here is an example schema that is sufficient to hang marteditor.
CREATE TABLE Sc__main__main (
id_key int auto_increment primary key,
gene varchar(10),
col1 int
);
CREATE TABLE Sc__bbb__main (
id_key int auto_increment primary key,
gene varchar(10),
col2 int
);
Making all column names unique didn't seem to solve it.
CREATE TABLE Sc__main__main (
id_key int auto_increment primary key,
gene varchar(10),
col1 int
);
CREATE TABLE Sc__bbb__main (
id_key2 int auto_increment primary key,
gene2 varchar(10),
col2 int
);
Best
Alexandre
Damian Smedley wrote:
On Thu, 20 Jul 2006, Alexandre Gattiker wrote:
Damian Smedley wrote:
On Thu, 20 Jul 2006, Alexandre Gattiker wrote:
Hello,
If a mart database accidentally contains two __main tables for
the same
dataset, marteditor in "Naive" mode hangs in the following loop at
DatabaseDatasetConfigUtils.java line 3342.
do you mean two tables with exactly the same name? All our ensembl
datasets have two __main tables per dataset so that shouldn't be a
problem. What are the names of your main tables out of interest -
we will
investigate
my tables were named something like:
Scerevisiae__aaa__main
Scerevisiae__bbb__main
According to the doc the part in the middle is irrelevant, and there
should only be one __main table with a given prefix part, if I
understand correctly.
guess our docs are a bit misleading on this - we will improve it for
the
0_5 release :) Your table names should be fine (eg) in
hsapiens_gene_ensembl we have:
hsapiens_gene_ensembl__gene__main with gene_id_key
hsapiens_gene_ensembl__transcript__main with gene_id_key,
transcript_id_key
suspect it is the combination of your main tables and keys that is
causing
the hang rather than the table names - what are your keys?
cheers
Damian
Cheers
Alexandre
--
Alexandre Gattiker
Swiss Institute of Bioinformatics, Genome Bioinformatics Group
Biozentrum Tel. +41 61 267 1579
Klingelbergstrasse 50 Fax +41 61 267 1585
4056 Basel [EMAIL PROTECTED]
Switzerland http://www.biozentrum.unibas.ch/primig
--
Richard Holland (BioMart Team)
EMBL-EBI
Wellcome Trust Genome Campus
Hinxton
Cambridge CB10 1SD
UNITED KINGDOM
Tel: +44-(0)1223-494416
------------------------------------------------------------------------
-------
Arek Kasprzyk
EMBL-European Bioinformatics Institute.
Wellcome Trust Genome Campus, Hinxton,
Cambridge CB10 1SD, UK.
Tel: +44-(0)1223-494606
Fax: +44-(0)1223-494468
------------------------------------------------------------------------
-------
