Hi Tibor and all,

As the developers were given info during the meeting, I just wrote a
basic description and install instructions - please see the srouce
code tree. Please also find answers below

> On Fri, 13 Aug 2010, Roman Chyla wrote:
>> Some of you had understandable worries it might not be so easy, but
>> actually it is not so bad -- I gave it a try today and I have a
>> working system where lucene reindexes everything as soon as it is
>> updated.
>
> The worries were among others with the sync kind of problems -- if
> objects fly between the two systems, would they remain the same, would
> they grow apart, is the setup vulnerable to update propagating issues
> from one system to another, etc.

because index is created by a writer which has instant access to the
invenio, and because values are only indexed (not stored - it is the
invenio, that takes care of that from database), those issues are
eliminated. The lucene is there as a search provider, returning docids
only

>
>> It took about 12 hours to finish
>
> For the INSPIRE content?  If not, for how many records?  And what is the
> index size, the number of terms, etc?  (usual interesting index stats to
> compare)

that was development time (the EVO argument went like: if it can be
shown it is easy and quick) - it was agreed we will do some tests when
we find the box with data (all inspire records), some time this week.
I have no real numbers with the 104 recs in the demo (all of them are
indexed at once in one batch in under 1s)

Best,

roman

>
> Best regards
> --
> Tibor Simko
>

Reply via email to