If you check the default framework shipped with Koha, you will notice several fields are linked to authority records (even though they might be hidden in the editor): MariaDB [koha_kohadev]> SELECT tagfield,tagsubfield,authtypecode FROM marc_subfield_structure WHERE authtypecode IS NOT NULL AND authtypecode<>'' AND frameworkcode='' GROUP BY tagfield,tagsubfield,authtypecode ; +----------+-------------+--------------+ | tagfield | tagsubfield | authtypecode | +----------+-------------+--------------+ | 100 | a | PERSO_NAME | | 110 | a | CORPO_NAME | | 111 | a | MEETI_NAME | | 130 | a | UNIF_TITLE | | 440 | a | UNIF_TITLE | | 600 | a | PERSO_NAME | | 610 | a | CORPO_NAME | | 611 | a | MEETI_NAME | | 630 | a | UNIF_TITLE | | 648 | a | CHRON_TERM | | 650 | a | TOPIC_TERM | | 651 | a | GEOGR_NAME | | 654 | a | TOPIC_TERM | | 655 | a | GENRE/FORM | | 656 | a | TOPIC_TERM | | 657 | a | TOPIC_TERM | | 658 | a | TOPIC_TERM | | 662 | a | GEOGR_NAME | | 690 | a | TOPIC_TERM | | 691 | a | GEOGR_NAME | | 696 | a | PERSO_NAME | | 697 | a | CORPO_NAME | | 698 | a | MEETI_NAME | | 699 | a | UNIF_TITLE | | 700 | a | PERSO_NAME | | 710 | a | CORPO_NAME | | 711 | a | MEETI_NAME | | 730 | a | UNIF_TITLE | | 796 | a | PERSO_NAME | | 797 | a | CORPO_NAME | | 798 | a | MEETI_NAME | | 799 | a | UNIF_TITLE | | 800 | a | PERSO_NAME | | 810 | a | CORPO_NAME | | 811 | a | MEETI_NAME | | 830 | a | UNIF_TITLE | | 896 | a | PERSO_NAME | | 897 | a | CORPO_NAME | | 898 | a | MEETI_NAME | | 899 | a | UNIF_TITLE | +----------+-------------+--------------+ But looking at etc/zebradb/marc_defs/marc21/biblio-koha-indexdefs.xml you will notice the $9 subfields is not correctly added to the Koha-Auth-Number:w index. This has the consequence of authority usage count not being precise if those fields are used and linked.
Created attachment 58267 [details] [review] Bug 17788: (MARC21) Add $9 fields to Koha-Auth-Number:w index Looking at the default framework's fields that are linked to authority records, there's a divergence with the Zebra index definitions. This yields to authority usage count be incorrect for users searching for authority records. MariaDB [koha_kohadev]> SELECT tagfield,tagsubfield,authtypecode FROM marc_subfield_structure WHERE authtypecode IS NOT NULL AND authtypecode<>'' AND frameworkcode='' GROUP BY tagfield,tagsubfield,authtypecode ; +----------+-------------+--------------+ | tagfield | tagsubfield | authtypecode | +----------+-------------+--------------+ | 100 | a | PERSO_NAME | | 110 | a | CORPO_NAME | | 111 | a | MEETI_NAME | | 130 | a | UNIF_TITLE | | 440 | a | UNIF_TITLE | | 600 | a | PERSO_NAME | | 610 | a | CORPO_NAME | | 611 | a | MEETI_NAME | | 630 | a | UNIF_TITLE | | 648 | a | CHRON_TERM | | 650 | a | TOPIC_TERM | | 651 | a | GEOGR_NAME | | 654 | a | TOPIC_TERM | | 655 | a | GENRE/FORM | | 656 | a | TOPIC_TERM | | 657 | a | TOPIC_TERM | | 658 | a | TOPIC_TERM | | 662 | a | GEOGR_NAME | | 690 | a | TOPIC_TERM | | 691 | a | GEOGR_NAME | | 696 | a | PERSO_NAME | | 697 | a | CORPO_NAME | | 698 | a | MEETI_NAME | | 699 | a | UNIF_TITLE | | 700 | a | PERSO_NAME | | 710 | a | CORPO_NAME | | 711 | a | MEETI_NAME | | 730 | a | UNIF_TITLE | | 796 | a | PERSO_NAME | | 797 | a | CORPO_NAME | | 798 | a | MEETI_NAME | | 799 | a | UNIF_TITLE | | 800 | a | PERSO_NAME | | 810 | a | CORPO_NAME | | 811 | a | MEETI_NAME | | 830 | a | UNIF_TITLE | | 896 | a | PERSO_NAME | | 897 | a | CORPO_NAME | | 898 | a | MEETI_NAME | | 899 | a | UNIF_TITLE | +----------+-------------+--------------+ This patch adds the missing ones to the authority number index as it is done for the rest of the fields. To test: - Verify that etc/zebradb/marc_defs/marc21/biblios/biblio-koha-indexdefs.xml contains intries pointing the $9 subfield of all the fields in the 'tagfield' column above, to the Koha-Auth-Number:w index. - Sign off :-D
I added known NORMARC and UNIMARC devs to CC, so they look at this issue and fill bugs for their MARC flavours if needed. This bug/patch covers MARC21.
I have checked that all those fields are now indexed and this bugs it is signed off BUT under my undestading this opens a new bug regarding the authorities For example 650 SUBJECT ADDED ENTRY--TOPICAL TERM and 690 - LOCAL SUBJECT ADDED ENTRY--TOPICAL TERM (OCLC, RLIN) POINTS TO the same authority base --> TOPIC_TERM Under my understanding Koha should have new authority for cocver 690, 691, 692...If I am not wrong, If we don't add new authotiries, the official and local authorities will be the at the TOPIC_TERM and it is little weird What do you think?
Linking the same authority type to multiple bibliographic fields should not cause any problem. As those are custom use fields, I think it's ok as it is. If someone needs a different framework for them, it can be added via the configuration.
(In reply to Hugo Agud from comment #3) > I have checked that all those fields are now indexed and this bugs it is > signed off BUT under my undestading this opens a new bug regarding the > authorities > > For example 650 SUBJECT ADDED ENTRY--TOPICAL TERM and 690 - LOCAL SUBJECT > ADDED ENTRY--TOPICAL TERM (OCLC, RLIN) POINTS TO the same authority base > --> TOPIC_TERM > > Under my understanding Koha should have new authority for cocver 690, 691, > 692...If I am not wrong, If we don't add new authotiries, the official and > local authorities will be the at the TOPIC_TERM and it is little weird > > What do you think? I do not have a strong opinion on that; you could open a new report and invite other people to give feedback too.
Tomas: Why do we have 691 but not 791 (there is 751)
Adding a signoff line too for Hugo Agud (check bug history)
Created attachment 59604 [details] [review] Bug 17788: (MARC21) Add $9 fields to Koha-Auth-Number:w index Looking at the default framework's fields that are linked to authority records, there's a divergence with the Zebra index definitions. This yields to authority usage count be incorrect for users searching for authority records. MariaDB [koha_kohadev]> SELECT tagfield,tagsubfield,authtypecode FROM marc_subfield_structure WHERE authtypecode IS NOT NULL AND authtypecode<>'' AND frameworkcode='' GROUP BY tagfield,tagsubfield,authtypecode ; +----------+-------------+--------------+ | tagfield | tagsubfield | authtypecode | +----------+-------------+--------------+ | 100 | a | PERSO_NAME | | 110 | a | CORPO_NAME | | 111 | a | MEETI_NAME | | 130 | a | UNIF_TITLE | | 440 | a | UNIF_TITLE | | 600 | a | PERSO_NAME | | 610 | a | CORPO_NAME | | 611 | a | MEETI_NAME | | 630 | a | UNIF_TITLE | | 648 | a | CHRON_TERM | | 650 | a | TOPIC_TERM | | 651 | a | GEOGR_NAME | | 654 | a | TOPIC_TERM | | 655 | a | GENRE/FORM | | 656 | a | TOPIC_TERM | | 657 | a | TOPIC_TERM | | 658 | a | TOPIC_TERM | | 662 | a | GEOGR_NAME | | 690 | a | TOPIC_TERM | | 691 | a | GEOGR_NAME | | 696 | a | PERSO_NAME | | 697 | a | CORPO_NAME | | 698 | a | MEETI_NAME | | 699 | a | UNIF_TITLE | | 700 | a | PERSO_NAME | | 710 | a | CORPO_NAME | | 711 | a | MEETI_NAME | | 730 | a | UNIF_TITLE | | 796 | a | PERSO_NAME | | 797 | a | CORPO_NAME | | 798 | a | MEETI_NAME | | 799 | a | UNIF_TITLE | | 800 | a | PERSO_NAME | | 810 | a | CORPO_NAME | | 811 | a | MEETI_NAME | | 830 | a | UNIF_TITLE | | 896 | a | PERSO_NAME | | 897 | a | CORPO_NAME | | 898 | a | MEETI_NAME | | 899 | a | UNIF_TITLE | +----------+-------------+--------------+ This patch adds the missing ones to the authority number index as it is done for the rest of the fields. To test: - Verify that etc/zebradb/marc_defs/marc21/biblios/biblio-koha-indexdefs.xml contains intries pointing the $9 subfield of all the fields in the 'tagfield' column above, to the Koha-Auth-Number:w index. - Sign off :-D Signed-off-by: Hugo Agud <hagud@orex.es> Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Pushed to master for 17.05, thanks Tomas!
This patch has been pushed to 16.11.x and will be in 16.11.04.
Pushed to 3.22.x for 3.22.17
Pushed to 16.05.x, for 16.05.10 release