Hi Jan,
The regex exclude is system-wide. I might have to do the regex
filtering when loading the data.
The problem is that there will be a process querying solr for matching
records which will be "alerts".
I have to prevent that the matching entries are loaded but not indexed.
Thanks for the tips.
J
On Sun, 26 Jun 2011 13:21:51 +0200, Jan Høydahl <jan....@cominvent.com>
wrote:
When marking a line in your app, you could re-index that line with a
field set to hide=true so that subsequent queries will not show that
line.
Regarding regex exclude: Will that be same for all users of system or
is it per user? If it is system-wide I'd consider using a back-end
job
which re-indexes matching lines with a field exclude=true. This is
more efficient than running all the regexes query-time.
--
Jan Høydahl, search solution architect
Cominvent AS - www.cominvent.com
Solr Training - www.solrtraining.com
On 24. juni 2011, at 14.54, Js wrote:
Hi List,
I'm looking into some options on what technology to adopt
building a specific logfile search solution.
At first glance it looks
like Solr is the tool I'm looking for. I intend to write a web-based
front end for end users
What would be a possible approach to tackle
following requirements? In other words how could these requirements
be
translated into Solr on a high level.
I'm not asking for solutions, just
pointers, approaches, tips, Solr features to look at, possible
pitfalls,
...
- A query results into a set of results.
- Individual records
from this query should have the ability to be marked so (although
they
match the query) those specific records don't show anymore when the
same
query is rerun.
- I don't want to delete data from the db/index
- I
want to avoid that my application has to take care of excluding
parts of
the returned data by keeping track which record id's to exclude.
- A
query should exclude the records which have a match in a possibly
large
growing list of regexes.
Thanks!
Jelle
