|
More site info...
Common forum | Forum profile
|
|
Forum profile page for Common forum on http://sphinxsearch.com.
This report page is the aggregated overview from a single forum: Common forum, located on the Message Board at http://sphinxsearch.com.
This forum profile page summarizes the general forum statistics such as: Users Activity, Forum Activity, and Top Authors, which are reported in either a table or graph below for a given reporting time period.
Additional forum profile information for "Common forum" on the Message Board at http://sphinxsearch.com is also shown in the following ways:
1) Latest Active Threads
2) Hot Threads for Last Week
Warning: These statistics are generated using 'best efforts' and can experience delays and reporting errors at times. Please note that such statistics do not constitute a forum's popularity and/or exact posting volumes at any given reporting period.
|
|
|
|
|
Posting activity on Common forum:
|
|
Week
|
Month
|
3 Months
|
|
Threads:
|
73
|
284
|
907
|
|
Post:
|
132
|
620
|
1,994
|
|
|
Common forum Posting activity graph:
|
Top authors during last week:
user's latest post:
Showing columns in result
Published (2009-11-26 20:11:00)
You are aware sphinx is a 'index', not a datastore. http://sphinxsearch.com/faq.html#row-storage To get original data you need to look it up in your own database.
user's latest post:
Suggestions for string attributes
Published (2009-11-19 16:34:00)
> Any idea when this will be implemented in trunk? > I'm not sure. BTW I would *not* trust this patch, in a production environment. I quickly hacked it together to get it to work.
user's latest post:
English and Greek characters
Published (2009-11-19 09:26:00)
Under better investigation my db is latin1 swedish ci and i want to ask if i can use in the config the latin1 swedish ci to search greek characters?
user's latest post:
How do i make two indexes?
Published (2009-11-24 15:09:00)
Thanks, i figured it had to be something like that but didnt want to presume too much!! As for making Sphinx work with ASP, do you know if this is theoretically possible, and if so any resources that i could take a peek at to help me along the way??
user's latest post:
prefix indexing problem
Published (2009-11-25 17:01:00)
> I use xmlpipe with static xml file and I don't use mysql. If I change data source to mysql then prefix search works normally.
user's latest post:
Index fail: Lost connection to...
Published (2009-11-26 15:38:00)
Hi Zam, First, ensure that the directory for searchd.pid exists (and is writeable): #cd /home/adam/sphinx/var/log/ #ls -al searchd.pid If it is there, stop sphinx and see if it is still there (should be gone). If it is still there, delete it. If the directory doesn't exist, then mkdir var and possibly log. Then reindex catalogue using rotate: #/scripts/indexer catalogue --rotate Then restart sphinx: (wherever searchd is)...
user's latest post:
Passing multiple search parameters
Published (2009-11-26 07:09:00)
Barry, please see the sample code for this issue: Index for invitation model-- indexes :activity_id indexes urpose_id indexes escription...
user's latest post:
main index size in main + delta...
Published (2009-11-26 09:16:00)
> > --merge-dst-range only removes the old key words, but it deosn't physically remove it > from the index according to the documentation. I had also didn't notice any decrease of > the index size.. > > > > Ok. > Now the question is: is it expected behaviour (not compating the index size) or is it a > bug? > I would suspect it is a bug, because what is the difference to killlist then?...
user's latest post:
main index size in main + delta...
Published (2009-11-20 21:30:00)
--merge-dst-range only removes the old key words, but it deosn't physically remove it from the index according to the documentation. I had also didn't notice any decrease of the index size.. > Hi, > > I am wondering about one thing. > If I merge two indexes (~100GB in size) and delta (~1GB in size) and use > --merge-dst-range which have such values that all documents from main index are filtered, > will main index...
|
|
|
|
Latest active threads on Common forum::
Started 2 months ago (2009-09-24 14:13:00)
by adammac
> Here I am using the same filed(NAME) for sql_attr_str2ordinal. Is it correct or Should I
> write as shown below?
>
> sql_query = SELECT ID, NAME, NAME AS NAME_ORD, ADDRESS FROM EMPDETAILS
> sql_attr_str2ordinal = NAME_ORD
This is the proper way.
select id,name,name as name_ord,etc from table
sql_attr_str2ordinal = name_ord
Started 4 months ago (2009-07-28 19:31:00)
by Arantor
> Basically I want to know if there is a way to have a string attribute so I can filter by
> category. It is not possible for me to convert the database system to use integers for
> categories.
You can word search on it. Using the extended2 mode you can do column independent
searches.
"@category word" for example.
There is a sql_attr_string being developed, it's in the SVN ...
Started 1 day, 22 hours ago (2009-11-26 15:38:00)
by adammac
Hi Zam,
First, ensure that the directory for searchd.pid exists (and is writeable):
#cd /home/adam/ sphinx/var/log/
#ls -al searchd.pid
If it is there, stop sphinx and see if it is still there (should be gone). If it is still
there, delete it.
If the directory doesn't exist, then mkdir var and possibly log.
Then reindex catalogue using rotate:
#/scripts/indexer catalogue --...
Started 2 years, 1 month ago (2007-10-16 22:09:00)
by shodan
> I searched PHP API and it seems to me that this method (UpdateAttributes) does not
> support update for MVA attributes (?).
It does not support MVAs yet, indeed. There are plans to implement that.
Started 5 months, 3 weeks ago (2009-06-06 12:07:00)
by red
Sorry, I should have clarified that after restarting the high loads are permanent. I
understand after a restart the first few minutes server load goes up, but I'm talking
about the loads staying much higher in definitely.
Started 1 day, 17 hours ago (2009-11-26 20:11:00)
by barryhunter
You are aware sphinx is a 'index', not a datastore.
http://sphinxsearch.com/faq.html#row-storage
To get original data you need to look it up in your own database.
|
|
Hot threads for last week on Common forum::
Started 4 days, 21 hours ago (2009-11-23 16:35:00)
by barryhunter
> I currently have sphinx being used to index a mediawiki installation. However i now need
> to expand its functionality so that it can index our corporate intranet. Im under the
> impression that this is possible but cant find anything on how to actually do it. What id
> like to know is whether its possible to have multiple indexes so each can be searched
> independantly of ...
Started 2 days, 22 hours ago (2009-11-25 14:46:00)
by adammac
Hi Kovacs,
Try changing:
min_prefix_len = 1
to:
min_prefix_len = 0
in sphinx.conf, then reindex with --rotate
#/usr/local/bin/indexer <indexname> --rotate
Started 1 day, 22 hours ago (2009-11-26 15:38:00)
by adammac
Hi Zam,
First, ensure that the directory for searchd.pid exists (and is writeable):
#cd /home/adam/sphinx/var/log/
#ls -al searchd.pid
If it is there, stop sphinx and see if it is still there (should be gone). If it is still
there, delete it.
If the directory doesn't exist, then mkdir var and possibly log.
Then reindex catalogue using rotate:
#/scripts/indexer catalogue --...
Started 1 week, 1 day ago (2009-11-19 13:50:00)
by barryhunter
>
> - Is there an option to fire one single query and get all the results at one shot instead
> of separate queries for displaying faceting.
You already using RunQueries, which is the best idea.
See:
http://sphinxsearch.com/news/39.html
the latest versions will have optimizations to help with this. Perhaps you could try the
SVN version?
>
> - We found ...
Started 6 days, 2 hours ago (2009-11-22 10:54:00)
by Arantor
> I wanted to ask those Sphinx users who have implemented multiple instances of searchd
> about the load on the server and whether it is worth it.
> I have one instance running that searches a database of 80000 records based on user
> input. I need the second instance in order to set up Sphinx to search another table that
> will hold past user queries.
Why not one instance ...
Started 2 days, 19 hours ago (2009-11-25 18:23:00)
by barryhunter
Please supply an example query. And maybe your 'source's, just the sql_query and
attributes would probably be enough.
At the moment I dont beleive its clear what you are trying to do.
Started 3 days ago (2009-11-25 13:07:00)
by barryhunter
> tried to kill 12745,
All you doing is trying to kill the grep command itself, - which will have already
terminated ;)
What actully happens when try to start the daemon? Did you check the searchd log?
Sounds like an unclean shutdown, perhaps just try removing the .spl files, doing a
reindex and starting searchd.
Failing that just delete all the index files (/var/sphinx/*.sp?)...
Started 1 week ago (2009-11-20 21:30:00)
by ramana
--merge-dst-range only removes the old key words, but it deosn't physically remove it
from the index according to the documentation. I had also didn't notice any decrease of
the index size..
> Hi,
>
> I am wondering about one thing.
> If I merge two indexes (~100GB in size) and delta (~1GB in size) and use
> --merge-dst-range which have such values that all documents from...
Started 1 week, 1 day ago (2009-11-20 10:59:00)
by barryhunter
> Is it best practice to remove punctuation marks (.,'" etc.) or not from the text before
> indexing.
No need usually. Any chars not in 'charset_table' will not be considered parts of words -
ie as seperators.
Pretty sure most people (and I know I do) just throw the raw text as sphinx - and it
handles 'cleaning' and tokenising - in fact even an option top strip html.
> ...
Started 5 days, 2 hours ago (2009-11-23 10:44:00)
by barryhunter
The API has a UpdateAttributes function
http://sphinxsearch.com/docs/current.html#api-func -updateatttributes
You can have your application update Sphinx when a new rating comes in.
|
|
Size: 620 bytes
