On November 4th 2013, bibliographic and authorities unimarc formats will change. Default unimarc frameworks should be adapted to these changes. Affected fields for Unimarc/ Biblio: - 205$d and 208$d generate "=" - 4XX$t is repeatable - 501: new subfields $m and $r added - 503: new subfield $o - 510, 512, 513, 514, 515, 516, 517, 518, 520, 540: new subfield $n - 545: new subfields $e $h $i $j $n $z - 930: new subfields $w and $z Affected fields for Unimarc/ Authorities - 033: new field (Thoses changes are also needed for versions 3.08 and 3.10)
Created attachment 20709 [details] [review] Bug 10726: UNIMARC charts corresponding to 6th french edition of the bibliographic manual and 3rd english edition of authorities manual
According to the norm, the changes involved by this patch are: for Unimarc/ Biblio: - 4XX$t is repeatable - 503: new subfield $o - 545: new subfields $e $h $i $j $n $z - 930: new subfields $w and $z for Unimarc/ Authorities - 033: new field New subfield $n in 510, 512, 513, 514, 515, 516, 517, 518, 520, 540 is not in the norm. Subfields $m and $r for 501 where already in the frameworks.
Hi Sonia, is there a complete online reference for UNIMARC formats? I'm looking at your patch, for example you add 033 for authority framework. Now, according to http://www.ifla.org/node/7950/ 122, 127, 128, 333, 501, 502, 511, 512, 521, 522 and perhaps others are still missing. Sorry but I'm new to UNIMARC :)
Hello I will take a look tomorrow M. Saby
It is much more complex than I thought... To answer Bernardo's question, I would say that the official source for UNIMARC authority is : for authorities : http://www.ifla.org/node/7950/ (3rd ed, update 2012) for biblios : http://www.ifla.org/node/7974 (3rd ed, update 2012) But the translated manuals we have in France does not seem to be up to date. We have on BNF (French national library) for authorities : http://www.bnf.fr/documents/UnimarcA.pdf (translation made in 2004, based on the 2d english edition of 2001, with some mistakes of english verion corrected) for biblios : http://www.bnf.fr/fr/professionnels/anx_formats/a.unimarc_manuel_format_bibliographique.html (based on the 3rd english edition, in its 2008 version) And Sudoc academic network makes its own documents! http://www.abes.fr/Sudoc/Produire-dans-le-Sudoc2/Formats-d-echange/%28profilId%29/168/%28rubId%29/1875 This can explain discrepancies! What I am certain of : french universities need to follow ABES format. For french public libraries, I don't know. Maybe the BNF document For non french Unimarc libraries, maybe the english document. Sonia, could you tell us what is the source you used for your patch? I think it is this pdf from ABES website, but I am not sure... http://www.abes.fr/Media/Fichiers/Sudoc-Fichiers/Produire-dans-le-Sudoc/Document-fournisseurs-Avril-2013 Are you sure it applies to french public libraries too? Mathieu
I think that bib and auth default framework must be complete, independent on language. I'm working on this (auth), building a txt file from UNIMARC Concise Authorities Format (2009) (http://www.ifla.org/publications/unimarc-formats-and-related-documentation), adding updates (http://www.ifla.org/node/7950) It's a boring and error prone task :( With this I plan to check which fields/subfields are absent, and eventually write a patch to update EN default auth framework. I could also check what is missing on FR default auth framework, but writing in French is out of my reach :)
http://marc-must-die.info/index.php/Main_Page ;-) Seriously, I agree that the DEFAULT framework should be the same in all language, but we must be carefull with specific frameworks used by some type of libraries and/or some countries. Ex of specific field : 359 field, used only in french universities, to store the Table of contents (instead of 327 field) And all the ITEMS management is specific to each country. (Can you believe that?!) As you say, boring+++ Mathieu
Patch applied cleanly, go forth and signoff
See also http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11022 and http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11021 for unimarc frameworks. Mathieu
Created attachment 22519 [details] [review] Bug 10726 - Followup : missing subfields in previous specifications edition
Hi, some conflicts :( Auto-merging installer/data/mysql/fr-FR/marcflavour/unimarc_complet/Obligatoire/framework_DEFAULT.sql CONFLICT (content): Merge conflict in installer/data/mysql/fr-FR/marcflavour/unimarc_complet/Obligatoire/framework_DEFAULT.sql Auto-merging installer/data/mysql/en/marcflavour/unimarc/mandatory/unimarc_framework_DEFAULT.sql CONFLICT (content): Merge conflict in installer/data/mysql/en/marcflavour/unimarc/mandatory/unimarc_framework_DEFAULT.sql
Comment on attachment 20709 [details] [review] Bug 10726: UNIMARC charts corresponding to 6th french edition of the bibliographic manual and 3rd english edition of authorities manual Review of attachment 20709 [details] [review]: ----------------------------------------------------------------- I was going to attempt to rescue and look at this, but there are likely other issues that need to be looked at first. ::: installer/data/mysql/fr-FR/marcflavour/unimarc_complet/Obligatoire/framework_DEFAULT.sql @@ +1468,4 @@ > ('503', 'k', 'numérotation (chiffres arabes)', '', 0, 0, '', 5, '', '', '', 0, 0, '', NULL, '', ''), > ('503', 'l', 'numérotation (chiffres romains)', '', 0, 0, '', 5, '', '', '', 0, 0, '', NULL, '', ''), > ('503', 'm', 'localisation', '', 0, 0, '', 5, '', '', '', 0, 0, '', NULL, '', ''), > +('503', 'n', 'établissement présidant la localisation', , 0, 0, '', 5, '', '', '', 0, 0, '', NULL, '', ''), Was there an accidental deletion of '' after the text? @@ +1629,4 @@ > ('616', 'f', 'dates', '', 1, 0, '', 6, '', '', '', 0, 0, '', NULL, '', ''), > ('616', 'j', 'subdivision de forme', '', 0, 0, '', 6, '', '', '', 0, 0, '', NULL, '', ''), > ('616', 'x', 'subdivision de la catégorie sujet sous forme textuelle', '', 0, 0, '', 6, '', '', '', 0, 0, '', NULL, '', ''), > +('616', 'y', 'subdivision géographique', ', 0, 0, '', 6, '', '', '', 0, 0, '', NULL, '', ''), A single '?! Typo, right? ::: installer/data/mysql/fr-FR/marcflavour/unimarc_lecture_pub/Obligatoire/framework_DEFAULT.sql @@ +3224,4 @@ > INSERT INTO `marc_subfield_structure` (`tagfield`, `tagsubfield`, `liblibrarian`, `libopac`, `repeatable`, `mandatory`, `kohafield`, `tab`, `authorised_value`, `authtypecode`, `value_builder`, `isurl`, `hidden`, `frameworkcode`, `seealso`, `link`, `defaultvalue`) VALUES ('600', '2', 'code du système d''indexation', '', 0, 0, '', -1, '', '', '', NULL, 0, 'ARTI', '', NULL, NULL); > INSERT INTO `marc_subfield_structure` (`tagfield`, `tagsubfield`, `liblibrarian`, `libopac`, `repeatable`, `mandatory`, `kohafield`, `tab`, `authorised_value`, `authtypecode`, `value_builder`, `isurl`, `hidden`, `frameworkcode`, `seealso`, `link`, `defaultvalue`) VALUES ('600', '3', 'numéro de la notice d''autorité', '', 0, 0, '', 2, '', '', '', NULL, -1, 'ARTI', '', NULL, NULL); > INSERT INTO `marc_subfield_structure` (`tagfield`, `tagsubfield`, `liblibrarian`, `libopac`, `repeatable`, `mandatory`, `kohafield`, `tab`, `authorised_value`, `authtypecode`, `value_builder`, `isurl`, `hidden`, `frameworkcode`, `seealso`, `link`, `defaultvalue`) VALUES ('600', '9', 'koha internal code', '', 0, 0, '', 2, '', '', '', NULL, -1, 'ARTI', '', NULL, NULL); > +INSERT INTO `marc_subfield_structure` (`tagfield`, `tagsubfield`, `liblibrarian`, `libopac`, `repeatable`, `mandatory`, `kohafield`, `tab`, `authorised_value`, `authtypecode`, `value_builder`, `isurl`, `hidden`, `frameworkcode`, `seealso`, `link`, `defaultvalue`) VALUES ('600', 'a', 'nom', 'sujet', 0, 0, '', 2, '', 'SAUT', '', NULL, 0,'ARTI','''600b'',''600c'',''600d'',''600f'',''600x'',''600y'',''600z''', NULL, NULL); I think there are some quote issues.
Sonia doesn't work anymore at BibLibre. Unassigning to avoid signaling that this bug is still being supervized. Which could wrongly discourage someone to continue the work.
It looks like the necessary updates were included in a later update bug: Bug 18309 - UNIMARC update from IFLA for new Koha installations *** This bug has been marked as a duplicate of bug 18309 ***