->get_patrons has been removed but not replaced!
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Dealt with that previously in the module during the rebase.
It conflicted with bug 23281.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Either we display everything, or only the info from the attribute itself
(i.e. not its type)
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Tests are still passing that way, we can continue
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
We do no longer need this package, we can use
Koha::Patron::Attribute::Types directly instead.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Replace C4::Members::AttributeTypes->num_patron with
Koha::Patrons->filter_by_attribute_type
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
We can then now start to move methods from C4::Members::AttributeTypes
as well.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
There is already a method in Koha::Patron::Attribute to check the
uniqueness constraint, let us it to replace CheckUniqueness
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
This patch replace Koha::Patron->get_extended_attributes with
->extended_attributes
It's now a getter a setter method.
It permits to replace UpdateBorrowerAttribute and use
create_related from DBIx::Class
Notes:
* We face the same variable names difference than in a previous patch
(value vs attribute)
Bug 20443: Remove SetBorrowerAttributes
squash + RM get_extended_attributes
RM get_extended_attributes
SQUASH Bug 20443: Remove UpdateBorrowerAttribute and SetBorrowerAttribute
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
This subroutine was only used once, easy to replace.
SetBorrowerAttributes must replace the attributes for the ones logged-in
user is allowed to edit, that's why we filter by the library limits
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Koha::Patron::Attributes->search mimicks what is done in
Koha::AuthorisedValues->search.
But actually it should be more explicit when the caller use it.
For instance filter_by_branch_limitation (see discussion on bug 11983).
This will be useful for the following patches as we will need a way to
replace the $no_branch_limit flag.
When the $no_branch_limit flag is called, a simple ->search call should
be done.
When we want to limit on a specific library we can pass the branchcode
in paramter of filter_by_branch_limitation (this is not used yet).
If not passed the logged-in user library will be used by default.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
The GetBorrowerAttributes subroutine return the attributes for a given
patron.
Using get_extended_attributes we can acchieve it easily. The problematic
here is to restore the method's name (value vs attribute,
value_description vs description of the authorised value, as well as
display_checkout that should not be a method of Attribute, but
Attribute::Type instead)
value_description was used when the attribute types were attached to an
authorised value category. To avoid the necessary test in template and
controller there is now a $attribute->description method that will
display either the attribute's value OR the value of the authorised
value when needed. We should certainly use this one from few other
places.
Notes:
* This patch rename Koha::Patron->attributes with Koha::Patron->get_extended_attributes.
It will be renamed with Koha::Patron->extended_attributes in ones of the next
patches when it will become a setter as well.
* GetBorrowerAttributes did not care about the library limits, we still
do not
* The opac_only flag was not used outside of test, we drop it off.
* To maintain the existing behavior we add a default order-by clause to
the search method [code, attribute]
* From C4::Letters::_parseletter we always display the staff description
of the AV, There is now a FIXME to warn about it
* FIXMEs are not regressions, existing behaviors must be kept
* TODO add a new check to bug 21010 to search for inconsistencies in
patron's attributes attached to non-existent authorised values
* One test has been updated in Modifications.t, order_by is now
by default set to ['code', 'attribute']
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
We want to retrieve a specific patron's attribute for a given patron.
We then add a new method to Koha::Patron.
This patch add a getter method ->get_extended_attribute_value
to use the DBIx::Class relation
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Same as previously for methods that have been added by bug 17792.
It's better to be explicite and tell we are fetch the related attribute's type
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
These methods have been added to Koha::Patron::Attribute but are wrong, see bug 18339
We should use ->type->$method instead
Moreover the tests exist in another subtest, we do not need them.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
There is something wrong, and a regression has been caught by those
tests:
If an invalid date is passed from the add item form, the app now
crashes.
Before:
* if the date was completely invalid, the field was blanked
silently
* DateTime::Format::MySQL was used to convert dates, and it's not
strict at all. For instance, what happened in the selenium tests for
dateaccessionned: %Y-%m-%d was prefilled by the framework plugin, then
the biblionumber was added, we ended with something like (eg for today)
2020-03-234242 (with biblionumber=4242). DateTime::Format::MySQL
converts that to 2020-03-23
We must deal with invalid dates, but I do not think it is good to add it
back to Koha::Item->store, we will prefer to raise the error to the end
user, saying that something went wrong (and more specifically the
dates).
The (ugly) trick was in C4::Items::_mod_item_dates
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
The recently introduced Ill item creation for circulation feature was
not properly handling the move from AddItem to Koha::Item->new()->store.
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Removing previous change, again.
But this time in Object.pm
It fixes t/db_dependent/Koha/Object.t and t/db_dependent/Koha/Item.t is
still passing.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
This patch fixes an issue when editing items.
* The issue
Cannot blank a subfield when editing an item.
If you have an item with itemcallnumber=42, then edit it, blank it and
save. The itemcallnumber is still 42
* Why? (line numbers from https://gitlab.com/joubu/Koha/-/tree/bug_23463)
additem (and other item's edition forms) receives a list of tags,
subfields and values, generates a MARC::Record::XML then calls
ModItemFromMarc:
717 my $itemtosave=MARC::Record::new_from_xml($xml, 'UTF-8');
727 my $newitem = ModItemFromMarc($itemtosave, $biblionumber,
$itemnumber);
And ModItemFromMarc:
282 my $item = TransformMarcToKoha( $localitemmarc,
$frameworkcode, 'items' );
283 $item->{cn_source} = delete $item->{'items.cn_source'}; #
Because of C4::Biblio::_disambiguate
284 $item_object->set($item);
ModItemFromMarc never knows that the field has been blank.
Prior to bug 23463 we had a map of default values, and ModItemFromMarc
was doing:
426 my $item = TransformMarcToKoha( $localitemmarc,
$frameworkcode, 'items' );
427 my $default_values = _build_default_values_for_mod_marc();
428 foreach my $item_field ( keys %$default_values ) {
429 $item->{$item_field} = $default_values->{$item_field}
430 unless exists $item->{$item_field};
431 }
I do not want to reinsert that list of default values.
Here I wrote a generic method in Koha::Object to set the value passed in parameter,
or "blank" if not passed.
It's nulled if can be set to null in DB, or the default value is retrieved from
the schema info.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
From commit bf49eecdd87e2b29760226281ab1afc0a185c7f0
Bug 23463: Replace AddItem calls with Koha::Item->store
in build_sample_item:
- my $itype = delete $args->{itype}
- || $self->build_object( { class => 'Koha::ItemTypes' } )->itemtype;
+ # If itype is not passed it will be picked from the biblio (see Koha::Item->store)
So before we generated a new itemtype if not passed, now we pick the one from biblioitem->itemtype.
For this specific test we need to make sure they are different.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
get_dirty_columns only work for existing items.
This fixes t/db_dependent/ShelfBrowser.t
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
It seems that some of those tests could be removed as safe_delete is
widely tested in Koha/Item.t
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
For discussion, this patch revert the changes made previously.
This line exists in Koha::Item->store as it is the translation of:
if (exists $item->{'timestamp'}) {
delete $item->{'timestamp'};
}
that was coming from _do_column_fixes_for_mod (called from ModItem)
To preserve existing behavior I would be in favor of keeping it like that to
avoid regression, and deal with it separately if we want to improve/remove it.
So basically here we are setting it to undef in Koha::Item->store to make it
handle correctly by the parent Koha::Object->store. I agree that's kind of weird
and must be improved.
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
This sounds wrong as we should let the DBMS do that, but it was failing.
Here we are doing the same as Koha::Patron->store for dateenrolled
To recreate the failure, prove t/db_dependent/Koha/Item.t without this
patch:
DBD::mysql::st execute failed: Column 'timestamp' cannot be null [for
Statement "UPDATE `items` SET `more_subfields_xml` = ?, `timestamp` = ?
WHERE ( `itemnumber` = ? )" with ParamValues: 0='<?xml version="1.0"
encoding="UTF-8"?>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Wrong replacement in additem.pl
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Few occurrences have been added since this patchset has been originaly
written
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
No need to calculate cn_sort if not dirty when we store
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
This change to Koha::Item->store seems correct here, but tests from /db_dependent/Items.t is failing now.
Adjusting them to make them pass.
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
We do not want to log if nothing changed
How could we do that better?
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
So, we get less warnings. I do not really understand why...
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
And adding a FIXME about discard_changes
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
