db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kathey Marsden <kmarsdende...@sbcglobal.net>
Subject Re: GSoC 2011: Derby Test and Fix
Date Sat, 26 Mar 2011 16:08:51 GMT
On 3/25/2011 12:14 PM, siddharth srivastava wrote:
> Hi
> I am applying for Derby Test and Fix project during GSoC this year.
> I have shortlisted the following issues for GSoC 2011:
> I would like to have the view of the community on these issues and the 
> tentative time line, before I post a detailed proposal.
> 1) Derby 5014: Tests should restore the timeout values to default 
> after they are done running.
>          Timeline: May 24- 7th June 2011 (15 days)
> 2) Derby-4795: Starting network server with -ssl turns SSL off
>         Timeline: 8th April - 15th June 2011 ( 8 days)
> 3) Derby- 4779: NPE while inserting into a table which has a generated 
> column and an insert trigger
>        Timeline: : 16th April-30 June 2011 (15 days)
> 4) Derby-4249:Create a simple store recovery test in JUnit
>        Timeline: 1 July-23 July (23 days)
> 5) DERBY-3676 : Make the toString() method of Derby PreparedStatements 
> print out SQL text with ? parameters replaced by the values that have 
> been set so far
>       Timeline: 24th July- 7th August (14 days)
> 6) Store test conversion: updatelocks.sql (doesn't not have dedicated 
> Jira issue yet)
>      Timeline: 8th August - 14th August ( 7 days)
Siddharth I think this looks good except instead of updatelocks.sql I 
would prefer to see you convert one of the recovery tests to show that 
your simple recovery test is a good model.  Also if you look at 
updatelocks.sql you
will see it runs quite  a few subsql files and also I think is run in 
various modes so might be just very large and time consuming and not 
suitable for a seven day project.  I would suggest converting 
oc_rec[1-4].java tests which I think are set up to run consecutively to 
get the desired recovery behavior.   It would be good to look at these 
in your planning for DERBY-4249 and shave 4 days off the DERBY-4249 and 
add it to the conversion.

I think you propose a very achievable schedule and I admire that and 
think you should call out in your final proposal that you have made an 
effort to plan realistically.   Make sure you include technical aspects 
and challenges for each issue, for example for DERBY-3676, the technical 
changes might not be that complex but I am guessing there will be a fair 
amount of community discussion around the issue as there has been already.



View raw message