Common forum | Forum profile

Breaking the output into two sections might help explain things.... ~~~~~~~~~~~~~~ > indexing index 'wiki_main'... > collected 20082 docs, 22.4 MB > sorted 3.7 Mhits, 100.0% done > total 20082 docs, 22423950 bytes > total 6.920 sec, 3240454.72 bytes/sec, 2902.02 docs/sec ~~~~~~~~~~~~~~ > indexing index 'wiki_incremental'... > collected 0 docs, 0.0 MB > total 0 docs, 0 bytes > total 0.079 sec,...

Hi joepilon, To be honest, no, I haven't really considered nor researched any other product (other than the KickFire appliance). About 2 years ago, I made the switch from Sybase to mySQL. The move was more b/c I wanted to abandon the remaining windows servers (actually, they were towers at the time) and move to all linux rack servers (debian). I have been following the MariaDB, but, I have been a little slow to test this db solution. I...

thanks a million for your response The bitwise OR seems to work for us - prior to this, we used $cl->SetMatchMode(SPH_MATCH_BOOLEAN) - searching for "tom" would also return "tommy" $cl->SetMatchMode( SPH_MATCH_PHRASE | SPH_MATCH_BOOLEAN ); - searching for "tom" would ONLY return "tom" as a complete word search - that's the way we want it to be We also...

> Sounds like you should aim to have it all in one query - then SetLimits applies to the > one query, and then sphinx handles merging of results - in such a way that paging will > work. This won't work with different fieldweights per index. > Maybe SetIndexWeights, can be used. Or you can somehow fudge the fields, so that > setFieldWeights can be used, for example my having different feilds in each index. Not > used it...

> Seems to be this would be an ideal use for the mysql Partition engine > > http://dev.mysql.com/doc/refman/5.1/en/partitioning-overview.html > Hmm, very interesting! I'm reading about it now and like, very much, what I see :-) Thanks for the pointer!

hi , on analysis i found out the error.. when i indexed 50,000 ducuments , on searching the authorname 'Isabel Wolff' , the database gives the results.. and the productid's are 1.680 2.1000 3.1001 4.1002 5.1184 6.1185 7.24875 on sphinx search with the same authorname its giving the result.. 1.680 2.1000 3.1001 4.1002 5.1184 6.1185 7.2487 and the difference is the last 7 th document it is giving only the four digit id.. i cant...

How do I update ? I've glanced the documentation but it only teach you how to recompile the new code, isn't there a way to update it correctly ? - I mean, without installing the RC-2 and removing the other one ?

What is the best solution to update the modified old data? This situation is not quite suitable for #live-update. Ie have a table with some data. Suppose that is updated on the principle described in #live-update (receive only new data, not indexing old). But what if the old ones have changed? Once you can do re-index only changed data?

As explained in document Phrase rank is based on a length of longest common subsequence (LCS) of search words between document body and query phrase. So if there's a perfect phrase match in some document then its phrase rank would be the highest possible, and equal to query words count. . . . In SPH_MATCH_EXTENDED mode, final weight is a sum of weighted phrase ranks and BM25 weight, multiplied by 1000 and rounded to integer Can any one...

I wish I could delete the thread :) I had an error in the query inside sql_query_killlist but "indexer" did not raise any warning or error (is this expexted?) but I decided to verify it anyway and fixed the error and the counts are OK, now. I first learned about the solution from here http://www.sphinxsearch.com/forum/view.html?id=4673, so thanks Barry! > Ah, I was talking too early. upon reading the docs I thought that...