Thank you Samuele for your instant reply!
I've changed the CFG_WEBSEARCH_DEF_RECORDS_IN_GROUPS value to 0 and now
the algorithm indeed works as expected (which means takes quite some
time to finish), but that's fine :)
I'm very eager to see all the new functionality that bibauthorid v1.2
brings and especially the WebAuthorProfile pages that we've seen during
your presentation!
Keep up the good (and hard) work,
Theodoros
On 15/5/2012 4:07 μμ, Carli Samuele wrote:
On 05/15/2012 02:39 PM, Theodoros Theodoropoulos wrote:
Hello everyone,
Quick question: why most of the authors never get extracted and disambiguated
from 100/700 tags for records that exist in the demo site?
(I'm running the latest master and the logs indicate that BibAuthorID is v1.1.1)
More specifically, Is there an easy answer to why I only get only a few lines
in the aid* tables for the demo records in Atlantis site, for
bibauthorid --update-personid --all-records (or)
bibauthorid --disambiguate --from-scratch
...no matter how low (or high) I set the
CFG_BIBAUTHORID_PERSONID_MIN_P_FROM_BCTKD_RA,
CFG_BIBAUTHORID_PERSONID_MIN_P_FROM_NEW_RA and
CFG_BIBAUTHORID_PERSONID_MAX_COMP_LIST_MIN_TRSH parameters?
FYI, if I specify a record id (for example bibauthorid --update-personid -i
100,102,103), records are being added as expected!
And now, the enhancement req: the -i parameter should also accept ranges, as it
happens with bibindex (something like -i 100-103,105). This is very useful for
mass creation of author 'clusters'.
Apologies if everything is handled in v1.2 of bibauthorid :)
Kind regards,
Theodoropoulos Theodoros
Hello Theodoros,
you are not doing anything wrong, and the answer is painfully simple!
Really recently the behavior of search engine have been changed through the
option 'CFG_WEBSEARCH_DEF_RECORDS_IN_GROUPS' in the main configuration file
which defaults to 10.
Unfortunately, this means that search engine is not reporting a comprehensive
list of records any more as it ever did before, and this broke bibauthorid for
it is using search engine to retrieve a comprehensive list of existing and not
deleted records.
A patch to solve this issue is on it's way and will come to master soon, along
with the merging algorithms and lots of other improvements.
For the moment forcing that configuration value to 0 should force search engine
to run 'as legacy' and everything should work as expected.
Should that not be the case please let me know and we can investigate together
what else could be wrong!
Thank you very much for your understanding,
Samuele
