ok, i will start on some of this at some point. in the short term,
however, rolling back to 2.4.3 has fixed the problem and i am going to
push new indexes to my production servers. My question is now is, when i
roll the swish-e back from 2.4.4 to 2.4.3 on the production machines,
should i replace the perl stuff? I know i can't make use of Peter's SAS
stuff, but i am going to stick with SWISH::API in the short term anyway.
yeah, right, i guess i will see for myself if i need to rewind the perl
stuff.
bill, i will follow your suggestions below when i have some time to chase
this further, but it seems to be some version wonkiness.
Brad
---------------------
Brad Miele
VP Technology
IPNStock.com
866 476 7862 x902
bmiele@ipnstock.com
On Wed, 8 Nov 2006, Bill Moseley wrote:
> On Wed, Nov 08, 2006 at 07:50:37AM -0800, brad miele wrote:
>> Essentially, the issue is that certain seraches seem to fail against our
>> full index, yet they work on smaller indexes using the exact same config
>> options.
>>
>> here is my meandering attempt at showing the involved stuff.
>>
>> first, the searches,
>> - search against full index:
>> bwayipn02# swish-e -w "Corey Rich" -f indexes/master_ipn.index
>> # SWISH format: 2.4.4
>> # Search words: Corey Rich
>> # Removed stopwords:
>> err: no results
>> .
>>
>> - search against smaller test index:
>> bwayipn02# swish-e -w "Corey Rich" -f test.index
>> # SWISH format: 2.4.4
>> # Search words: Corey Rich
>> # Removed stopwords:
>> # Number of hits: 54
>> # Search time: 0.001 seconds
>> # Run time: 0.009 seconds
>
> I guess I'd dump both indexes and look at the word positions and
> metanames for that phrase.
>
> Or, I might also run indexing with -T indexed_words saved to a file
> and again look at how that phrase is added to the index.
>
> Then I'd start wondering about the searches not working.
>
> --
> Bill Moseley
> moseley@hank.org
>
> Unsubscribe from or help with the swish-e list:
> http://swish-e.org/Discussion/
>
> Help with Swish-e:
> http://swish-e.org/current/docs
> swish-e@sunsite.berkeley.edu
>
>
>
Received on Wed Nov 8 10:04:43 2006