[InterMine Dev] lucene index creation patch lost in v1.4?

Julie Sullivan julie at flymine.org
Wed May 6 09:22:57 BST 2015


Yes, please! That would be very helpful.

On 06/05/15 08:09, Joe Carlson wrote:
> OK. I was wrong.  org.intermine.web.search.KeywordSearch (also forgot to
> add the class name to the package last email) has all the old lucene
> code, but apparently it all now in
> org.intermine.api.lucene.KeywordSearch. There is a bit of code that
> needs to be transferred from the old module to the new one.
>
> I’ve added some code to my version in
> https://github.com/JoeCarlson/intermine/blob/phytozome-v10.2/intermine/api/main/src/org/intermine/api/lucene/KeywordSearch.java
>
> Should I try to cherrypick the checkin and make a pull request?
>
> Joe
>
>
>> On May 5, 2015, at 1:27 PM, Joe Carlson <jwcarlson at lbl.gov
>> <mailto:jwcarlson at lbl.gov>> wrote:
>>
>> I think I see the issue here a little more clearly.
>>
>> The lucene index creation code was moved from
>> org.intermine.web.search.saveIndexToDatabase to
>> org.intermine.api.lucene.saveIndexToDatabase. There are 2 lines of
>> code that did not get migrated to the new method:
>>
>>         LOG.info("Saving signature to database...");
>> writeObjectToDB(os,MetadataManager.SEARCH_INDEX_SIGNATURE,TextUtil.generateRandomUniqueString());
>>
>> (well, really 1 line) I'm testing out restoring this line in the
>> migrated method at line 430. It looks to me that the code for checking
>> to see if the search index signature is still in the disk file still
>> exists in org.intermine.web.search.loadIndexFromDatabase
>>
>> oe
>>
>> On 05/05/2015 12:04 PM, David Goodstein wrote:
>>> Joe Carlson had committed a patch to lucene index creation to prevent
>>> InterMine, on  tomcat restart, from always trying to pull the
>>> compressed index from the database and expanding it onto disk, even
>>> when it already exists on disk and is unchanged.  That patch seemed
>>> to have been included as far back as 1.3.1, but now appears to be
>>> lost.  Was this accidental, or was it interacting poorly with some
>>> aspect of 1.4?
>>>
>>> thanks,
>>>
>>> -David
>>>
>>>
>>> --
>>> David M. Goodstein, Ph.D.
>>> Phytozome Group Lead
>>> Eukaryotic Data Systems
>>> Joint Genome Institute - U.S. Dept. of Energy
>>> http://phytozome.jgi.doe.gov
>>>
>>
>
>
>
> _______________________________________________
> dev mailing list
> dev at intermine.org
> http://mail.intermine.org/cgi-bin/mailman/listinfo/dev
>



More information about the dev mailing list