On Aug 8, 2007, at 6:17 PM, Carl Troein wrote:

> Kjell Petersen wrote:
>
>> Charles Girardot wrote:
>>>> If so, which of the files in the sql directory
>>>> should be applied, and are there other database changes necessary?
>>>>
>>> it is normally base_mysql_new_tables_for_v13.sql for the db
>>> mofifications and base_mysql_new_data_for_v13.sql to populate my
>>> tables with initial and mandatory stuff.
>>>
>> I've started to clean up this, most of the content of the two files
>> above are now in place in the "regular" sql files and in a parallel
>> upgrade version set of sql files. With the input from Vegard I will
>> hopefully be able to finish this task shortly.
>
> I've done a bit of work on 1.2.18 today, and there's one more
> major bug that I'd like to fix (namely memory usage for 'normal'
> BASEfiles with many reporters - right now this is causing real
> problems for our installation). Other than that, it's only the
> 'new' stuff that remains.
>
> I'm getting somewhat confused by the many different versions that
> are present in sql/, so it seems best that I wait for you to sort
> it out.
> Since base_pgsql.sql is autogenerated from base_mysql.sql and
> base_mysql_data.sql, I'd like to make sure that everything that's
> needed is present in those files, in the same format as the old
> stuff (right now there are backticks around the new column names,
> and the mgedont stuff isn't in there yet (and I don't think there
> should have to be any ALTER TABLE statements there)).
> So that would take care of creating the database from scratch. The
> updating of existing databases is probably more important as this
> point, and since there are so many new tables (and much data) it
> seems reasonable to keep all that in external files (as you've
> done) and have dbmigrate load them.

I'll take a look.
Kjell: is there any more to be modified ?

>
>> So the upgrade of the DB schema and additional php/java code is
>> "straight forward", but some work remains. But how can we get the new
>> mandatory information on reporters (4 new columns in reporter table)
>> into the reporter.inc.php file ?
>
> What are these columns? So far I haven't found anything about
> them in the code that I've looked through.

These columns are needed for proper ADF (Array Description File) and  
MageML generation.
Those are (from my local reporter_columns.php ; function getColumns()):

....
array("bioSeqType",     XCOL_STRING,            "Sequence Type",                
1),
array("bioSeqPolymerType",      XCOL_STRING,            "Sequence Polymer 
Type",                1),
array("role",   XCOL_STRING,            "Reporter Role",                1),
array("controlType",    XCOL_STRING,            "Control Type",         1)

These columns take controlled vocabulary as input (see http:// 
www.ebi.ac.uk/miamexpress/help/adf/):
* Sequence Type:
  cDNA_clone
ds_oligo
genomic_DNA
PCR_amplicon
ss_oligo

Sequence Polymer Type:
DNA
RNA
Protein

Reporter Role:
Experimental
Control


Control Type:
control_biosequence - for example a spike
control_buffer - buffer spotted on the array
control_empty - nothing spotted on the array
control_genomic_DNA - e.g. salmon sperm DNA
control_label - landing lights
control_reporter_size - size standard
control_spike_calibration - spike at varying concentrations
control_unknown_type - an unknown control


I guess the question was how could we deal with these new mandatory  
columns automatically  ie with the migration? Since code has to be  
added to the
reporter_columns.php, I guess we can t really make this without  
requesting manual intervention of the BASE admins, right?

Charles

>
> //Carl
>
> -- 
>  Carl Troein - [EMAIL PROTECTED]
>  http://www.thep.lu.se/~carl/
>
> ---------------------------------------------------------------------- 
> ---
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a  
> browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> basedb-devel mailing list
> basedb-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/basedb-devel

=====================================
Charles Girardot
European Molecular Biology Laboratory
C. Boulin and E. Furlong Groups
Tel: +49-6221-387-8426
Email: [EMAIL PROTECTED]
Meyerhofstrasse 1 - Postfach 10.2209
69012 Heidelberg, Germany
=====================================



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
basedb-devel mailing list
basedb-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/basedb-devel

Reply via email to