* many bugfixes
* adding value_builder : you can map a subfield in the marc_subfield_structure to a sub stored in "value_builder" directory. In this directory you can create screen used to build values with any method. In this commit is a 1st draft of the builder for 100$a unimarc french subfield, which is composed of 35 digits, with 12 differents values (only the 4th first are provided for instance)
* bugfixes and improvements
* manage mandatory MARC subfields
* new table : authorised_values. this table contains categories and authorised values for the category. On MARC management, you can map a subfield to a authorised_values category. If you do this, the subfield can only be filled with a authorised_value of the selected category.
this submit contains everything needed :
* updatedatabase
* admin screens
* "links" management
* creation of a html-list if a subfield is mapped to an authorised value.
Note this is different from authorities support, which will come soon.
The authorised_values is supposed to contains a "small" number of authorised values for a category (less than 50-100). If you enter more authorised values than this, it should be hard to find what you want in a BIG list...
seems to work a few.
still to do :
* manage html checks (mandatory subfields...)
* add list of acceptable values (authorities)
* manage ## in MARC format
* manage correctly repeatable fields
and probably a LOT of bugfixes
The acquisition system in MARC version will work like this :
* marcimport will put marc records into a "breeding farm" table.
* when the user want to add a biblio, he enters first the ISBN/ISSN of the biblio. koha searches into breeding farm and if the record exists, it is shown to the user to help him adding the biblio. When the biblio is added, it's deleted from the breeding farm.
This commit :
* modify acqui.simple home page (addbooks.pl)
* adds import into breeding farm
Please note that :
* z3950 functionnality is dropped from "marcimport" will be added somewhere else.
* templates are in a new acqui.simple sub directory, and the marcimport template directory will become obsolete soon.I think this is more logic
Fixed muffed quotes in &gettemplate calls.
Added a temporary print statement in &z3950menu, so it'll print
something instead of giving a browser error.
marcimport.pl has been almost completly rewritten, so LOT OF BUGS TO COME !!! You've been warned. It seems to work, but...
As with my former messages, nothing seems to have been changed... but ...
* marcimport now uses HTML::Template.
* marcimport now uses MARC::Record. that means that when you import a record, the old-DB is populated with the data as in version 1.2, but the MARC-DB part is filled with full MARC::Record.
<IMPORTANT NOTE>
to get correct response times, you MUST add an index on isbn, issn and lccn rows in biblioitem table. Note this should be done in 1.2 too...
</IMPORTANT NOTE>
<IMPORTANT NOTE2>
acqui.simple manage biblio, biblioitems and items tables quite properly. Normal acquisition system manages biblio, biblioitems BUT NOT items. That will be done in the near future...
</IMPORTANT NOTE2>
what's next now ?
* bug tracking, of course... Surely a dozen of dozens...
* LOT of developpments, i'll surely write a mail to koha-devel tomorrow (as it's time for dinner in France, and i plan to play NeverwinterNights after dinner ;-) ...
parameters in to the templates
Didn't make any changes to bulkmarcimport.pl recently, so I'm probably
squashing somebody else's changes. Will have to fix that...
NOT use it on a production system yet! It needs to be more adaptable to
different MARC implementations.
Note that the script will spit out a plain text copy of the records being
imported to STDOUT, so you will probably want to redirect to a file, or pipe to
less.
to do Z39.50 searches. Daemon will also wait to see if the user looks
beyond the first couple of pages of results, and will download more results
if necessary.
