Collecting a booking early (within the bookings lead period) with a
due date override before the booking start syncs end_date to a date
that precedes start_date. Koha::Booking->store has no interval
validation and the clash checks' date windows never see the inverted
period, so the broken row is stored silently - reproduced via AddIssue
on this patchset.
- pull start_date back to the checkout date whenever the synced due
date precedes it, in both AddIssue and AddRenewal, so the booking
window follows the loan instead of inverting
- cover both paths in Circulation.t; the AddRenewal case needs the
Biblio::check_booking exemption from the previous patch, as the
synced window no longer overlaps the stored one
Test plan:
1. prove t/db_dependent/Circulation.t
2. Set bookings_lead_period to e.g. 5 for an itemtype and enable
SpecifyDueDate
3. Book an item of that itemtype for a patron starting in 3 days
4. Check it out to them today with an explicit due date of tomorrow
5. Without this patch the booking is stored with end_date before
start_date; with it both dates follow the checkout period
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Koha::Biblio->check_booking counts current checkouts on bookable items
against the pool of available items. Like Item::check_booking before
this patchset, it made no exception for the checkout that belongs to
the very booking being validated, so updating an issued booking's
period could clash against its own loan whenever the new period no
longer overlaps the stored one (the item_id exemption for booked items
only catches bookings that overlap the window under test).
This is masked in the end_date syncs earlier in this patchset because
the synced window always contains the stored start_date; it surfaces
as soon as a sync moves the whole period, and would equally bite the
extension pathway on bug 42790.
- skip checkouts whose booking_id matches the booking being validated
when counting occupied bookable items, mirroring the
Item::check_booking exemption
- cover the blocking and non-blocking checkout cases in Biblio.t
Test plan:
1. prove t/db_dependent/Koha/Biblio.t
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
When a checkout satisfying a booking is renewed, the booking previously
kept its original end_date, blocking the booked period from other
patrons for the wrong window and leaving the two records out of sync.
This patch updates C4::Circulation::AddRenewal to pull the linked
booking's end_date along with the renewed due date, mirroring the
behaviour introduced for AddIssue earlier in this patchset. As at
issue time, if the new due date would clash with a subsequent booking
(possible when the renewal was explicitly allowed past a clash), the
booking keeps its original end_date.
Together with the previous patch this makes a renewal the natural way
to extend an issued booking: CanBookBeRenewed refuses renewals that
would run into the next booking, and AddRenewal keeps the booking in
step with the loan.
Test plan:
1) Run t/db_dependent/Circulation.t and confirm the new
'AddRenewal | booking end_date sync' subtest passes
2) Check out a booked item, renew it from the checkouts screen and
confirm the booking end date follows the new due date
Sponsored-by: OpenFifth <https://openfifth.co.uk/\>
Sponsored-by: Westminster City Council <https://www.westminster.gov.uk/\>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
CanBookBeRenewed checked bookings by asking Koha::Item->find_booking
for the first booking that clashes with the renewal period. When the
renewing patron's own booking satisfies this checkout, that booking is
returned first and renewal is allowed - subsequent bookings for the
item were never considered. A patron could therefore renew a booked
item straight through the start of the next patron's booking.
Additionally, when staff renew with an explicit due date (the renewal
due date input on the checkouts screen, or the hard due date on
circ/renew.pl), CanBookBeRenewed validated a *calculated* due date
whilst AddRenewal then applied the *requested* one, so the requested
date was never checked against bookings at all.
This patch:
* repeats the booking lookup with the patron's own booking excluded,
refusing the renewal ('booked') when it would run into the next
booking for the item
* adds an optional $date_due parameter to CanBookBeRenewed, used in
place of the calculated due date for the bookings check, and threads
the requested date through from svc/renew and circ/renew.pl
Test plan:
1) Run t/db_dependent/Circulation.t - the extended
'CanBookBeRenewed | bookings' subtest covers both the calculated
and requested due date cases
2) Check out a booked item to its patron, add a subsequent booking for
another patron, and attempt to renew past the second booking's
start date from the checkouts screen; confirm the renewal is
refused as 'booked'
Sponsored-by: OpenFifth <https://openfifth.co.uk/\>
Sponsored-by: Westminster City Council <https://www.westminster.gov.uk/\>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
When checking out a booked item, staff may override the suggested due
date. If that override runs into the next booking for the item, the
previous patch would silently keep the booking's original end date -
staff had no way of knowing their override crossed another booking.
This patch adds an exclude_booking_id option to Koha::Item->find_booking
so a lookup can skip a specific booking (here: the patron's own booking
being collected), and uses it in CanBookBeIssued to detect when an
explicitly requested due date would run into the next booking for the
item. When it would, a new BOOKED_DUE_DATE_CLASH confirmation is raised
and displayed on the checkout screen, telling staff the linked booking
will keep its original end date if they continue.
Only explicitly requested due dates trigger the confirmation -
calculated due dates are clamped to the booking end date later in the
staff checkout flow, so checking them would produce false positives.
SIP is unaffected as it never passes a requested due date.
Test plan:
1) Run t/db_dependent/Circulation.t and t/db_dependent/Koha/Item.t
2) Create two consecutive bookings on the same item for two patrons
3) Check the item out to the first patron specifying a due date beyond
the second booking's start date; confirm the new confirmation
message appears and checkout only proceeds once confirmed
Sponsored-by: OpenFifth <https://openfifth.co.uk/\>
Sponsored-by: Westminster City Council <https://www.westminster.gov.uk/\>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Koha::Item->check_booking refuses any booking period that starts before
the item's current checkout is due back. That is correct when assessing
a new booking against someone else's loan, but it also fired when the
checkout in question is the booking's own linked loan - making it
impossible to ever move an issued booking's end_date, as any update
would 'clash' with its own checkout.
This patch skips the checkout guard when the checkout's booking_id
matches the booking_id passed for exclusion, i.e. when the checkout
belongs to the very booking being updated. This is a prerequisite for
keeping booking end dates in sync with checkout due dates in this
patchset, and for the booking extension pathway on bug 42790.
Test plan:
1) Run t/db_dependent/Koha/Item.t and confirm the check_booking
subtest passes, including the two new checkout-related tests
Sponsored-by: OpenFifth <https://openfifth.co.uk/\>
Sponsored-by: Westminster City Council <https://www.westminster.gov.uk/\>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
When a booked item is checked out, the due date defaults to the booking
end_date. Staff may override the due date at checkout time, which
previously left the booking end_date out of sync with the checkout,
blocking the booked period for longer or shorter than the actual loan.
This patch updates C4::Circulation::AddIssue to sync any due date
override back to booking.end_date the moment the booking transitions to
'issued', keeping the booking and the checkout aligned from the point
of issue. The existing Koha::Booking->store clash detection is used to
guard the update: if the overridden due date would clash with a
subsequent booking for the same item, the checkout proceeds with the
booking's original end_date instead.
Test plan:
1) Run t/db_dependent/Circulation.t and confirm the new
'AddIssue | booking end_date sync with due date override' subtest
passes
2) Create a booking for a patron on a specific item
3) Check the item out to the patron, overriding the suggested due date
4) Confirm the booking's end_date now matches the overridden due date
Sponsored-by: OpenFifth <https://openfifth.co.uk/\>
Sponsored-by: Westminster City Council <https://www.westminster.gov.uk/\>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds a custom aj_field_template to the sip accounts
This allows libraries to specify a template for the title (AJ) field in sip
messages
The template allows using item and biblio fields (item provides access to biblio. but we pass
both in order to makes the templates simpler
To test:
1 - Install patches
2 - updatedatabase, yarn build, restart_all
3 - Setup a sip account, KTD already has a sip patron, term1, you can use that
don't add the aj_field_template yet
4 - Find or create a biblio with 245a and 245b like 245$aTitle : $bSubtitle and add item with TESTBARCODE
5 - From the command line:
perl misc/sip_cli_emulator.pl -su term1 -sp term1 -l CPL -t CR -a localhost -p 6001 -m item_information --item TESTBARCODE
6 - Confirm AJ field shows "Title :" as expected
7 - Edit sip account, set aj_field_template to:
[% biblio.title %] [% biblio.subtitle %] [% item.barcode %]
8 - Repeat the sip test
9 - Confirm you get the response like:
Title : Subtitle TESTBARCODE
10 - Sign off!
Signed-off-by: Bob Bennhoff <bbennhoff@clicweb.org>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
GetIssuingCharges overwrites its initial $charge = 0 with the value of
itemtypes.rentalcharge, which is NULL when no rental charge is configured.
This causes 'uninitialized value' warnings in every caller that compares
$charge > 0 (automatic renewals, SIP checkout/renew, OPAC).
The fix normalizes NULL to 0 at the source. All callers already treat
the return value as numeric, so this is safe. Also removes a now-redundant
$charge && guard in AddIssue that was working around this.
Test plan:
1. Apply patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/Circulation.t
=> SUCCESS: Tests pass!
3. Sign off :-D
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Chris Mathevet <chris.mathevet@inLibro.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
prove t/db_dependent/Koha/ILL/Request.t
prove t/db_dependent/Koha/Plugins/ILL/Backends.t
NOTE: These must be run without having any ILL backend plugin installed
Signed-off-by: Hannah Prince <helm.consortium.nhs.net>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Bug 33544 combined the ILL request detail fields into one column. The
combined column kept an empty data property without the related-object
search metadata. A global table search therefore generated a query against
"me.", which the REST endpoint rejected with a 500 error.
Restore the extended attribute search mapping for the fields displayed in
the combined Request details column. Add Cypress coverage for both the REST
query and the matching result.
Test plan:
Before applying the patch:
1. Make sure the ILLModule system preference is enabled.
2. Open a patron record.
3. Click "ILL requests history" in the menu on the left.
4. Type any text in the Search box above the table.
5. Confirm that Koha shows a 500 Internal Server Error message.
Apply the patch.
After applying the patch:
1. Repeat steps 2 to 4 above.
2. Confirm that the table refreshes without an error. It may show no
matching requests.
3. If the patron has an ILL request, search for part of a title or author
shown under Request details.
4. Confirm that the matching request remains in the table.
Co-Authored-By: Loontik <loontik@nugged.team>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The + character in RFC3339 timezone offsets (e.g. +00:00) is decoded
as a space when interpolated directly into a URL query string. Use
Test::Mojo form => { q => ... } which properly URL-encodes the
parameter value.
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds regression tests proving that RFC3339 datetime values
and boolean values nested inside DBIC operators (-and, -or, -between)
in q= queries are not converted to their DB-native format.
Unit tests (Query.t):
- Datetime: >=, <=, range, -or, -between, nested -and/-or, me. prefix
- Boolean: true/false inside -and, -or, nested -and within -or
- Non-datetime values pass through unchanged
Integration tests (Objects.t):
- Full API stack tests using action_logs with controlled data
- Same operator/nesting coverage as unit tests
Test plan:
1. Apply patch
2. Run:
$ prove t/Koha/REST/Plugin/Query.t
$ prove t/db_dependent/Koha/REST/Plugin/Objects.t
=> FAIL: Tests fail! Values inside -and are not converted
3. Sign off :-D
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Store scalar preference values when saving the acquisition preferences
and clean up the invoice explicitly.
Exercise the rendered receiving behavior for a basket that inherits
AcqCreateItem, one that explicitly overrides it with "ordering", and a
basket that creates items on receiving. Keep the explicit "ordering"
case as a control for behavior that already works without the production
fix.
This changes tests only.
Automated tests:
yarn cypress run --spec \
t/cypress/integration/Acquisitions/OrderReceivePreferences_spec.ts
Co-Authored-By: Loontik <loontik@nugged.team>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Bug 41563 converted AcqCreateItem and UniqueItemFields to booleans when
adding them to Koha.prefs. Both preferences are strings: AcqCreateItem
controls where items are created, while UniqueItemFields contains a
pipe-separated list.
When a basket inherits AcqCreateItem=ordering, the boolean value matches
none of the receiving page branches and existing items are hidden. When
items are created on receiving, the numeric UniqueItemFields value also
fails when the page calls split().
Preserve both string values in Koha.prefs. Add a Cypress regression test
which renders orderreceive.pl and verifies the values exposed to
JavaScript.
Test plan:
Before applying the patch:
1. Set AcqCreateItem to "placing an order" and UniqueItemFields to
"barcode".
2. Create a basket that inherits the system preference. Add an order with one
item, close the basket, and start receiving it.
3. Confirm that the Items table has no existing item row and no Receive
checkbox.
4. Set AcqCreateItem to "receiving an order" and open the same order for
receiving.
5. Confirm that the Items section is empty: no item-entry fields and no Add
item button are shown.
Apply the patch.
After applying the patch:
1. Set AcqCreateItem back to "placing an order" and open the order for
receiving again.
2. Confirm that its existing item is listed and has a Receive checkbox.
3. Set AcqCreateItem to "receiving an order" and open the same order again.
4. Confirm that the Items section contains item-entry fields and an Add item
button.
Co-Authored-By: Loontik <loontik@nugged.team>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Tests were failing on upgrade route because they tried to create
EDIFACT_SI authorised values that already existed from the atomicupdate
script, causing unique constraint violations.
This patch:
- Removes test code that created EDIFACT_SI values (tests should assume
database defaults exist)
- Adds EDIFACT_SI category to auth_val_cat.sql for fresh installs
- Adds all 32 EDItEUR List 3B servicing instruction codes to
auth_values.yml for fresh installs
Now the authorised values exist in both installation paths:
- Upgrade route: Created by atomicupdate script
- Fresh install: Created by installer SQL/YAML files
Test plan:
1. prove t/db_dependent/Koha/EDI.t
2. prove t/db_dependent/Koha/Edifact/Order.t
3. Verify both tests pass without warnings
Sponsored-by: Martin Renvoize <martin.renvoize@gmail.com>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This commit adds comprehensive tests for servicing instruction
processing in both incoming quotes and outgoing orders:
1. Integration tests for incoming quote processing (Koha/EDI.pm)
- Tests LVC (coded) and LVT (freetext) extraction
- Tests multiple GIR occurrences with different servicing instructions
- Tests all combinations: LVC only, LVT only, and LVC+LVT together
2. Test EDIFACT quote file (QUOTES_SERVICING.CEQ)
- Contains 3 line items with 4 total orders
- Demonstrates different servicing instruction patterns
3. Fix for multi-occurrence servicing instruction extraction
- Modified quote_item() to extract servicing instructions
per GIR occurrence instead of across all occurrences
- Ensures each order gets only its own servicing instructions
- Aligns with existing multi-item processing for funds and stock rotation
Test coverage:
- t/db_dependent/Koha/EDI.t: 29 tests for incoming quote processing
- t/db_dependent/Koha/Edifact/Order.t: 18 tests for outgoing order generation
(previously added)
All tests pass successfully.
Sponsored-by: Martin Renvoize <martin.renvoize@gmail.com>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds comprehensive unit tests for the servicing instruction
functionality in t/db_dependent/Koha/Edifact/Order.t.
Changes:
- Updated test count from 6 to 7
- Added servicing_instruction => undef to test orders to prevent
TestBuilder from generating random invalid JSON
- Added servicing instruction subtest with 18 tests covering:
* Single LVC instruction
* Single LVT instruction
* Multiple LVC in one group (tests GIR segment overflow)
* Mixed LVC/LVT in one group
* Multiple groups
* Empty servicing instruction
* Invalid JSON handling (logged to EDI log)
- Tests correctly handle GIR segment 5-element limit and overflow
to additional segments
Test plan:
1. Run: prove t/db_dependent/Koha/Edifact/Order.t
2. Verify all tests pass
Sponsored-by: Martin Renvoize <martin.renvoize@gmail.com>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Prevents additional orders from subscriptions from being swallowed into the invoice total calculation
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Coleen Lilly <clilly@aacpl.net>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Sponsored-by: Anne Arundel County Public Library <https://www.aacpl.net/>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
- Add 'library' to the allowed x-koha-embed values for the
GET /patrons/{patron_id} endpoint
- Add unit test verifying the library embed returns library_id
and name
Test plan:
1. Apply patch
2. yarn api:bundle
3. prove t/db_dependent/api/v1/patrons.t
4. Verify all tests pass, including the new 'library embed'
subtest under get() which verifies that requesting
x-koha-embed: library returns the patron's library object
Sponsored-by: Büchereizentrale Schleswig-Holstein <https://www.bz-sh.de/>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
When OpacSuppression is enabled, records marked as suppressed
(biblio.opac_suppressed = 1) still appear in public lists viewed
through the OPAC. This is confusing for patrons.
This patch adds a filter_by_not_suppressed method to
Koha::Virtualshelfcontents and uses it in opac-shelves.pl to
exclude suppressed records from list display when OpacSuppression
is enabled. OpacSuppressionByIPRange is respected: users within
the allowed IP range still see all records.
Test plan:
1. Apply patch
2. Enable OpacSuppression system preference
3. Create a public list and add some titles to it
4. Suppress one of the titles (set 942$n = 1)
5. View the list in the OPAC
=> SUCCESS: Suppressed record is no longer shown in the list
6. Set OpacSuppressionByIPRange to match your IP (e.g. 127.)
7. Repeat step 5
=> SUCCESS: All records (including suppressed) are shown
8. Run:
$ ktd --shell
k$ prove t/db_dependent/Virtualshelves.t
=> SUCCESS: Tests pass!
9. Sign off :-D
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Laura_Escamilla <laura.escamilla@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
If a plugin reports a date_authored or date_updated we can't parse,
remove the invalid property from the metadata hashref and emit a
warning.
To test this, introduce a separate test plugin directory and test file.
Having a plugin that produces warnings in the main test directory would
require accounting for this warning in every existing test that loads
plugins. Correspondingly, this patch also fixes an invalid date in an
existing test plugin.
Test plan:
1. Apply patch.
2. Run `prove -r t/db_dependent/Koha/Plugins/` and verify there are no
test failures.
Sponsored-by: Open Fifth
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Laura_Escamilla <laura.escamilla@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
If dt_from_string throws an exception whilst trying to parse the input,
catch it and return an empty result.
Test plan:
1. Apply patch.
2. Prove t/db_dependent/Template/Plugin/KohaDates.t
Sponsored-by: Open Fifth
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Test plan:
1) In the database, verify how many biblios exist (select count(*) from biblio;)
2) Run prove t/db_dependent/Koha/Item.t
3) Check the count again, there should now be 3 extra bib records left from the test run
4) Apply patch
5) Repeat the process, the biblio count should never grow after test runs
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Laura_Escamilla <laura.escamilla@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
It should not fail the unit test t/db_dependent/check_sysprefs.t.
Test plan :
1.0) Without patch
1.1) Run : prove t/db_dependent/check_sysprefs.t
=> 'All tests successful'
1.2) Go to Cataloging > Automatic item modifications by age
1.3) Create some entries and save
1.4) Run : prove t/db_dependent/check_sysprefs.t
=>
# Failed test at t/db_dependent/check_sysprefs.t line 96.
# got: '1'
# expected: '0'
# Too many sysprefs in DB: automatic_item_modification_by_age_configuration
2.0) Apply patch
2.1) Run : prove t/db_dependent/check_sysprefs.t
=> 'All tests successful'
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
GetNormalizedUPC duplicated logic already available via the
MetadataExtractor (Bug 34828) and exposed as $biblio->normalized_upc.
All callers migrated:
- Scripts with $biblio available: use $biblio->normalized_upc
- C4::Search (only MARC::Record): reuses the extractor instance
already created for OCLC
Assisted-by: Sonnet 4.6 (Anthropic)
Test plan:
1. Apply patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/Koha.t \
t/db_dependent/Koha/Biblio.t
=> SUCCESS: Tests pass
3. Verify UPC numbers still display in OPAC and staff detail pages
(Syndetics cover images)
4. Sign off :-D
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Replaces all calls to GetNormalizedOCLCNumber() and
Koha::Util::MARC::oclc_number() with the extractor pattern.
OPAC scripts pass their already ViewPolicy-filtered $record directly
to the extractor. This preserves the existing filtering behavior
without introducing new abstraction layers:
Koha::Biblio::Metadata::Extractor->new( { metadata => $record } )->get_normalized_oclc
Intranet scripts and the Koha::Biblio convenience method use the
raw cached extractor (035 visibility is not affected by ViewPolicy
in practice, and intranet doesn't filter):
$biblio->normalized_oclc
Test plan:
1. Verify OCLC numbers display on OPAC detail, ISBD, MARC detail
2. Verify staff catalogue detail and search results
3. prove t/db_dependent/Koha/Biblio.t
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Assisted-by: Sonnet 4.6 (Anthropic)
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Moves get_normalized_oclc from the MARC21-specific extractor to the
base ::MARC class (035$a is flavor-independent) with improved matching:
- Case-insensitive OCoLC detection (matches ocolc, OCOLC, etc.)
- Generic orgcode prefix stripping (any parenthesized prefix, not
just literal "(OCoLC)") for resilience against cataloging typos
Test plan:
$ prove t/db_dependent/Koha/Biblio/Metadata/Extractor/MARC/MARC21.t
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Assisted-by: Sonnet 4.6 (Anthropic)
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Koha::Biblio::Metadata->store() now repairs recoverable MARCXML faults
before saving, so this test's invalid data no longer reaches the DB
and detail.pl never hits a decoding error.
Write the metadata directly via the DBIC row instead, bypassing
store()'s repair, so the test still exercises detail.pl's own
defensive handling of corrupt metadata.
Test plan:
prove t/db_dependent/selenium/regressions.t
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Tests verify that EDI ORDRSP cancellations now use
Koha::Acquisition::Order->cancel() rather than a raw ModOrder call:
- Order status, cancellation reason and date are set correctly
- On-order items linked via aqorders_items are deleted
- Purchase suggestions linked to the biblio are reverted from
ORDERED back to ACCEPTED
- Non-cancellation actions update suppliers_report only
- Missing ordernumbers are handled gracefully without dying
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: Sharon Dugdale <sharon.dugdale@cumberland.gov.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Test plan:
1) Apply just the unit test patch
2) Run t/db_dependent/Koha/Patrons.t
3) Test should fail
4) Apply remaining patch
5) Repeat step 3
6) Test should pass
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
ISBNs beginning with 979 do not have an ISBN-10 version, and need to be normalized to an ISBN-13 format. All others, if valid, can safely be normalized to ISBN-10 format (current behaviour).
To set up
1. Download the attached sample records with 979 ISBNs
2. Log into the staff interface and go to Cataloguing
3. Go to Stage MARC records for import and upload the sample MARC file
4. Stage and import the biblio records
5. Once imported, open the detail pages for some of the records
6. In another tab, do a catalogue search and open the detail pages for ISBN-10 records, for example biblionumbers 6 and 54
To test
1. Apply the patch and restart services
2. Log into the staff interface and go to Koha Administration, Global system preferences, Enhanced Content tab
3. Cycle through each cover image service, enabled by the system preferences
4. With each cover image service is enabled, refresh the record detail pages you have open. The ISBN-10 records should CONTINUE to show their cover images. The newly imported ISBN-13 records should NOW show their cover images.
5. Confirm tests pass: t/db_dependent/Koha.t
testing with ISBN 9798885795692 should return this image from Amazon: https://images-na.ssl-images-amazon.com/images/P/9798885795692.01.LZZZZZZZ.jpg
testing with ISBN 9798876536914 should return this image from Amazon: https://images-na.ssl-images-amazon.com/images/P/9798876536914.01.LZZZZZZZ.jpg
Sponsored-by: Horowhenua Libraries Trust
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
1- Corrected tests according to the comments
2- Removed unused import
3- Added syspref check of 'patronimages'
4- Tweaked swagger path schema
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Api changes adding GET route for patronimage
Test plan:
1- Apply the patch
2- Run `npm run api:bundle`
2- Go to administration -> sys.pref. -> set patronimages to allow
3- Log in koha using a user with "list_borrowers" or "edit_borrowers" privileges
4- Have a patron with a profile image (add an image if necessary by
searching for a patron and then clicking on the default profile image on
the top left)
5- Open a an API testing tool (such as postman), write
{intranet_url}/api/v1/patrons/{borrowernumber}/default_image
where {intranet_url} is the base intranet url of koha and {borrowernumber}
is the borrowernumber of a borrower that has an image
6- Add OAuth 2.0 credentials coming from koha (patron -> more -> manage
api keys) and generate a token
7- Send the request and notice the image of the patron is returned
8- Do step 5-7 with a borrowernumber of a patron without an image and a
borrowernumber of a patron that doesn't exist (and other edge cases
where there should be an error message displayed such as "Access
forbidden", "Authentication required", etc.)
9- Notice an appropriate error message is displayed
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: David Nind <david@davidnind.com>
Bug 35797: Rename test file from 'image.t' to 'patron_images.t'
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
To test:
1. prove t/db_dependent/api/v1/idp.t
Signed-off-by: David Cook <dcook@prosentient.com.au>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds a parameter to call the method intranet_catalog_biblio_enhancements_toolbar_button with a biblionumber in the page catalogue/detail.pl.
Modification of cat-toolbar.inc to display new buttons correctly.
To test :
1. Apply the patch
2. Take a plugin who implements the method intranet_catalog_biblio_enhancements_toolbar_button
3. Check if the parameter biblionumber is passed correctly at this method
4. Create the new button in HTML in the plugin method
(button must be like the ones that are in the toolbar of the page detail.pl)
5. Go to the page in browser and check if the new button is displayed
Signed-off-by: Roman Dolny <roman.dolny@jezuici.pl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Added a unit test to confirm that search_for_display orders manually positioned news first, followed by unnumbered news by publication date descending.
Test passes:
prove t/db_dependent/Koha/AdditionalContents.t
Signed-off-by: Gretchen Maxeiner <MaxeinerGL@gcc.edu>
Signed-off-by: David Cook <dcook@prosentient.com.au>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This test built a "bad" biblio by storing invalid MARCXML directly via
Koha::Biblio::Metadata->store(), which now rejects it. Simulate
legacy corrupt data already in the database by writing it straight to
the DBIC row, bypassing the new validation, instead.
Signed-off-by: Baptiste Bayche <baptiste.bayche@inlibro.com>
Signed-off-by: Baptiste Wojtkowski <baptiste.wojtkowski@biblibre.com>
Sponsored-by: OpenFifth
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
AddBiblio/ModBiblio/ModBiblioMarc gain an optional warnings arrayref
option: any repair messages recorded by Koha::Biblio::Metadata->store()
(see previous commit) are appended to it via object_messages, so a
caller can show the cataloguer what was automatically fixed instead of
the fix happening invisibly.
AddBiblio also now distinguishes Koha::Exceptions::Metadata::Invalid
from other errors in its catch block and rethrows it, so a caller can
tell "this record is unrecoverable" apart from an unexpected internal
error and react accordingly (see following bugs for callers doing so).
The inline StripNonXmlChars calls in TransformHtmlToMarc are removed,
since stripping is now handled centrally and consistently by
Koha::Biblio::Metadata->store() rather than ad-hoc at every call site.
Signed-off-by: Baptiste Bayche <baptiste.bayche@inlibro.com>
Signed-off-by: Baptiste Wojtkowski <baptiste.wojtkowski@biblibre.com>
Sponsored-by: OpenFifth
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Koha::Biblio::Metadata->store() now validates that MARCXML round-trips
through MARC::Record before saving it, rather than saving whatever was
given and letting corruption surface later as an opaque failure
somewhere downstream.
Two faults are recovered automatically, and either or both may apply
to the same record:
- non-XML characters (recorded as nonxml_stripped, recovered via
StripNonXmlChars)
- datafields with no subfields at all, which MARC::File::XML refuses
to parse (recorded as empty_datafield_stripped, recovered by
removing the empty datafield)
Each recovered fault is logged, reported via object_messages for the
caller to act on, and recorded as a biblio_metadata_errors row (tag
and subfield included where identifiable) for later review - except
where an identical row already exists for the same metadata_id, so a
source record that stays dirty across repeated saves (e.g. re-harvested
from an external repository) doesn't pile up duplicate rows for the
same fault. Existing rows are otherwise left alone on a clean save,
since they're review flags requiring explicit resolution, not a cache
to invalidate.
If the MARCXML cannot be recovered at all, Koha::Exceptions::Metadata::Invalid
is thrown. The recovery logic is exposed as a repair_marcxml class
method so other callers that parse MARCXML ahead of a store() (see
following bugs) can reuse it instead of re-implementing it.
Signed-off-by: Baptiste Bayche <baptiste.bayche@inlibro.com>
Signed-off-by: Baptiste Wojtkowski <baptiste.wojtkowski@biblibre.com>
Sponsored-by: OpenFifth
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Sanjar Tulkinov Anvar o'g'li <sanjarbiy11@gmail.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
login_core only cleared leftover session state on the failed-auth
branch. The earlier 'SIP access not authorized' return (unknown
terminal id) left any account/ils/institution/policy/credentials from
a prior connection on a reused prefork worker in place.
Move the reset to the top of login_core so every non-success exit path
leaves clean state, and only a successful api_auth re-establishes the
session. This keeps the no-authenticated-state-until-auth invariant in
a single place rather than duplicating the key list per branch.
Also extend the unit test to cover the unknown-terminal path on a
reused worker.
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Sanjar Tulkinov Anvar o'g'li <sanjarbiy11@gmail.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
- t/C4/ClassSplitRoutine/RegEx.t (new): exercises _apply_substitution via
the public split_callnumber interface — basic substitution, alternate
delimiters, \n/$n backrefs, /gi combined flag, /e rejection, and malformed
or uncompilable rules (including the new logger warn path).
- t/Koha/Regex/Replacement.t: extend the subst() helper to handle the ig/gi
combined modifier and add a subtest that confirms case-insensitive global
replacement and backref case preservation.
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Baptiste Wojtkowski <baptiste.wojtkowski@biblibre.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Note: While the methodology in Koha::Regex::Replacement may seem
bespoke, it's similar to how Python's re.sub(pattern, repl, string)
regex substitution works.
Unit tests:
prove -v t/Koha/Regex/Replacement.t \
t/SimpleMARC.t \
t/db_dependent/MarcModificationTemplates.t \
t/Labels_split_Regex.t
Signed-off-by: David Cook <dcook@prosentient.com.au>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Baptiste Wojtkowski <baptiste.wojtkowski@biblibre.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Add a small single-pass helper, Koha::Regex::Replacement::expand_template,
that expands a user-supplied regex replacement string as data: numbered
backreferences ($1..$n, ${n}, the legacy \1..\9 form), named captures
($+{name}, ${name}) and the \n \r \t escapes. Every other character is
copied through literally, so \@{...}/${...} interpolation blocks end up as
text rather than being run. This is used to replace the s///ee double-eval.
The routine is fully qualified at the call sites rather than exported, and
case folding is left out to keep it small.
Signed-off-by: Sanjar Tulkinov Anvar o'g'li <sanjarbiy11@gmail.com>
Signed-off-by: David Cook <dcook@prosentient.com.au>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Baptiste Wojtkowski <baptiste.wojtkowski@biblibre.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The tests for desensitize and alert flags were failing because
items were created with random itemtypes that could have
sip_magnetic=1. This caused unexpected behavior since magnetic
media items don't get desensitized and trigger alert flags.
Fix by explicitly creating itemtypes with sip_magnetic=0 in:
- test_checkout_desensitize
- test_renew_desensitize
- test_checkin_v2
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: Marcus Michel <marcus.michel@rbkc.gov.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Add unit tests for SIP magnetic media support.
Test plan
1) Run t/db_dependent/SIP/Transaction.t in ktd and confirm it passes
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: Marcus Michel <marcus.michel@rbkc.gov.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
GetAgeRestriction is no longer used by any caller. This patch removes
it from the exports list and deletes the old test file.
Test plan:
1. Run tests without this patch:
$ ktd --shell
k$ prove t/db_dependent/Circulation.t \
t/db_dependent/Reserves.t \
t/Circulation/AgeRestrictionMarkers.t
=> SUCCESS: Tests pass!
2. Apply this patch
3. Run:
k$ prove t/db_dependent/Circulation.t \
t/db_dependent/Reserves.t \
t/db_dependent/Koha/Policy/Biblio/AgeRestriction.t
=> SUCCESS: Tests pass!
4. Sign off :-D
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Amended-by: Jonathan Druart
Removed this from the commit message:
"The function body is kept for now as it is not harmful and could be
used by third-party code. It is no longer exported."
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds age restriction handling to the Koha namespace:
- Koha::Biblio->age_restriction: metadata extractor that parses the
agerestriction field using the AgeRestrictionMarker system preference.
Returns the numeric age or undef. Uses single return with last pattern.
- Koha::Policy::Biblio::AgeRestriction->check($biblio, $patron): policy
class that evaluates whether a patron meets the age restriction.
Returns a Koha::Result::Boolean (with payload on restriction).
Test plan:
1. Apply this patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/Koha/Policy/Biblio/AgeRestriction.t
=> SUCCESS: Tests pass!
3. Sign off :-D
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
_upload_file, _download_file, _list_files and _rename_file's 'no directory configured' checks still called add_message() directly, skipping the status persistence this bug added elsewhere in the file. Route them through _abort_operation() so all error paths in Local.pm stay consistent.
prove t/db_dependent/Koha/File/Transport/Local.t
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Two related inconsistencies remained across the three transport
backends' error-handling paths:
1) Only Koha::File::Transport::SFTP's _abort_operation() persisted an
"errors" status snapshot to the status column on every failed
operation. FTP never did, and Local had no _abort_operation() at
all - so an SFTP transport's displayed status could reflect a
single transient failure from real usage (e.g. an EDI cron job)
while the exact same failure on FTP/Local left the stored status
untouched (stale, possibly still showing the last explicit
test_connection() result). This is exactly the kind of unclear
status reporting bug 42656 is about.
2) FTP's _abort_operation() carried its extra detail under an
operation_params key, while SFTP (and Local's plain add_message
calls) used path. The change_directory() success payload had the
same split: FTP used directory, SFTP and Local used path. The
admin template's change_directory row display depended on FTP's
mismatched key names, which meant it silently failed to show the
tested directory for SFTP transports.
Adds a shared _record_error($operation, \%payload) to the
Koha::File::Transport base class - add_message() plus status
persistence - and routes all three backends' _abort_operation() (Local
gains one) through it. Renames FTP's operation_params/directory keys
to path throughout, matching SFTP and Local, and simplifies the
template to read operation.detail.path unconditionally instead of
ORing two differently-named keys.
_record_error() persists the full trace of operations recorded on the
transport object so far this session (every success leading up to the
failure), not just the failing operation in isolation - the same
"full trace" shape TestTransport::process() already builds for an
explicit test_connection() run. A long-lived transport used across a
multi-step operation (e.g. an EDI download loop touching several files
before one fails) will therefore show every successful step leading up
to the failure. It also logs the failure - including the built trace -
via Koha::Logger at warn level, so it's visible in the Koha logs
without needing to query the transport's status column.
Test plan:
1. prove -r t/db_dependent/Koha/File/ t/db_dependent/Koha/BackgroundJob/TestTransport.t t/db_dependent/Koha/Edifact/Transport.t t/db_dependent/api/v1/file_transports.t
=> All tests successful (48 tests total), including:
- new "_abort_operation persists status consistently" subtests
in FTP.t, SFTP.t and Local.t that reload the transport from
the DB after a simulated failure and confirm the status
column was updated with the correct operation code and path
detail;
- a new "_record_error() tests" subtest in Transport.t proving
the persisted trace includes a prior success message plus the
failure (not just the failure alone), and that a warn-level
log line mentioning the operation and error was emitted.
2. koha-qa.pl -v 2 --more-tests
=> [PASS], no critic/pod warnings.
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Two issues in the operation_description BLOCK's status text, both
flagged in review:
1) The IF/SET that computed the " passed."/" failed." status text sat
between the SWITCH statement and its first CASE. Template Toolkit
only executes CASE-labelled content inside a SWITCH block, so this
was dead code - status was never actually set, and every operation
row silently rendered with no passed/failed text at all. It also
checked operation.error, a field that doesn't exist on the
operation hash (the error message lives under operation.detail.error,
and whether the operation failed is operation.status eq 'error');
even if the code path were reachable it would always have taken the
"passed" branch.
2) Relying on a leading space baked into the " passed."/" failed."
string literals to separate them from the preceding </span> is
fragile. Made the space explicit in the markup instead, and dropped
the leading space from the string values.
Moves the status computation above the SWITCH (using operation.status),
and adds an explicit space before [% status | html %] on every CASE line.
Also adds t/db_dependent/Koha/BackgroundJob/TestTransport.t, which was
entirely missing: it exercises process() end-to-end (enqueue, then
process, then reload both the job and the transport) and confirms the
persisted trace records every operation in sequence - successes before
the failure that ended the run - rather than collapsing to just the
final error.
Test plan:
1. prove t/db_dependent/Koha/BackgroundJob/TestTransport.t
=> All tests successful, confirming process() persists the full
operation sequence (order preserved, at least one success and
one error present) on both the job and the transport's own
status column.
2. Manually rendered the fixed operation_description BLOCK with
Template Toolkit against mock operations (mixed success/error) and
confirmed every row now shows "Connection passed.", "Change
directory (/incoming) passed.", ..., "Change directory (/outgoing)
failed.- error:No such directory" - status text present with
correct spacing, where before the fix no status text rendered at
all.
3. prove -r t/db_dependent/Koha/File/ t/db_dependent/Koha/Edifact/Transport.t t/db_dependent/api/v1/file_transports.t
=> All tests successful (47 tests total), no regressions.
4. koha-qa.pl -v 2 --more-tests
=> [PASS], no critic/pod warnings.
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Koha::File::Transport's three backends disagree on what
change_directory(undef) means: FTP navigates to the server root,
SFTP falls back to the SFTP session's own home directory, and Local
fell back to the configured download_directory or, if that's also
unset, silently to '.' - the Koha process's own working directory,
which has no relation to any configured transport location.
test_connection() calls change_directory(undef) unconditionally for
both the download and upload directory checks on every config save
(via the auto-enqueued TestTransport background job). For a Local
transport configured with only one of download_directory/
upload_directory, the untested direction's list_files() call would
silently list the Koha process's own cwd (e.g. the Koha install root)
instead of failing cleanly.
Add a private _working_directory($configured_field) helper that
returns the tracked change_directory() override or the given
configured field, but never falls back to '.'. _upload_file,
_download_file, _list_files and _rename_file now use it and record a
clear "not configured" error instead of silently acting on an
unrelated directory when neither is set.
FTP's root and SFTP's home-directory fallbacks are left as-is: unlike
Local's '.', they navigate to a real, meaningful location for that
specific connection rather than leaking unrelated filesystem content,
so they aren't part of this fix.
Test plan:
1. prove t/db_dependent/Koha/File/Transport/Local.t
=> All tests successful, including the new subtest proving that
list_files()/rename_file()/download_file() fail cleanly (with a
"not configured" error) rather than falling back to '.' when
their respective directory isn't configured, while the
configured direction (upload, in the test) still works.
2. prove -r t/db_dependent/Koha/File/ t/db_dependent/Koha/Edifact/Transport.t t/db_dependent/api/v1/file_transports.t
=> All tests successful (43 tests total), no regressions.
3. koha-qa.pl -v 2 --more-tests
=> [PASS], no critic/pod warnings.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
t/db_dependent/Log.t
- Add subtest verifying logscriptaction() logs to the SCRIPTS module,
respects the ScriptLog preference, prefixes loginfo with $0, appends
the info parameter, and honours custom action values.
t/Koha/Script_scrub_argv.t
- New test file exercising all branches of _scrub_argv():
--flag value form, --flag=value form, non-sensitive flags, case-
insensitive matching, multiple sensitive patterns (token, secret,
apikey, api_key), and a flag immediately followed by another flag.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
FTP, Local and SFTP list_files() were unified to return the same flat
shape, but consumers still had no portable way to ask a transport for
its current remote directory - the underlying libraries' own methods
of that name don't agree on what "cwd" means. Net::SFTP::Foreign's
cwd() is a pure read-only accessor, but Net::FTP's cwd() is a mutator:
called with no argument it defaults to "/", issues a real CWD command,
and returns a boolean rather than a path.
The koha-plugin-transport-browser plugin discovered this the hard way:
it read $transport->{connection}->cwd directly to capture the initial
working directory, which happened to work for SFTP but silently reset
the FTP connection to root and captured "1" (a boolean) as the "path"
on every request, breaking directory navigation for FTP transports.
Add a public current_directory() method to the Koha::File::Transport
base class, following the same connect-on-demand pattern as
list_files()/upload_file()/etc, with a _current_directory() hook each
subclass implements using its own real read-only accessor: pwd() for
FTP, cwd() for SFTP, and the tracked {current_directory} (falling back
to download_directory) for Local. Consumers should use this instead of
reaching into transport internals.
Test plan:
1. prove t/db_dependent/Koha/File/Transport.t
t/db_dependent/Koha/File/Transport/FTP.t
t/db_dependent/Koha/File/Transport/Local.t
t/db_dependent/Koha/File/Transport/SFTP.t
=> All tests successful, including the new current_directory()
tests subtest in each transport test file, which mock the
underlying libraries (Net::FTP / Net::SFTP::Foreign) to assert
each backend reads its own real read-only accessor.
2. koha-qa.pl -v 2 --more-tests
=> [PASS] for all four changed modules, no critic/pod warnings.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The FTP and Local transports both explicitly filter out '.' and '..'
pseudo-entries when listing a directory, but the SFTP transport never
did - it relied on the assumption that Net::SFTP::Foreign's ls() already
excludes them. It doesn't: ls() returns exactly whatever entries the
remote server includes in its SSH_FXP_READDIR response, with no default
filtering. Whether '.'/'..' show up therefore depends entirely on the
remote SFTP server implementation, producing inconsistent results between
otherwise identically configured SFTP accounts.
Test plan:
1. prove t/db_dependent/Koha/File/Transport/SFTP.t
=> fails before this patch (listing includes '.' and '..')
=> passes after this patch
2. prove -r t/db_dependent/Koha/File/Transport/ t/db_dependent/Koha/File/Transports.t
=> all still pass
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The MLSD path, SFTP, and Local always return all six documented
list_files() keys (filename, longname, size, perms, mtime, type) - undef
where the value is unknown, but the key always present. The NLST
fallback path (servers without MLSD support) only set filename, size,
mtime and type, omitting longname and perms entirely rather than
setting them to undef.
Functionally harmless for consumers reading $file->{perms} (Perl
returns undef for a missing hash key too), but it breaks the "same flat
shape" contract this bug established, and would trip up any consumer
using exists() instead of defined() to check for a key.
Test plan:
1. prove t/db_dependent/Koha/File/Transport/FTP.t
=> All tests successful, including the new assertion that longname
and perms keys are present (as undef) on the NLST fallback path.
2. koha-qa.pl -v 2 --more-tests
=> [PASS], no critic/pod warnings.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Net::SFTP::Foreign's ls() returned file metadata with size, permissions
and modification time nested inside a Net::SFTP::Foreign::Attributes
object, while FTP and Local returned flat hash keys and silently
dropped directories. Callers of list_files() needed backend-specific
parsing to get at file metadata or to know a directory even existed,
defeating the purpose of the shared transport API.
All three transports now return the same flat hashref shape per entry:
filename, longname, size, perms, mtime, type ('file'/'directory'/
'other'). Directories are included everywhere; '.' and '..' remain
excluded everywhere. FTP's MLSD path prefers the UNIX.mode fact for
exact Unix permissions, falling back to an approximation from the RFC
3659 perm capability fact when UNIX.mode isn't available, and parses
the modify fact into mtime; its NLST fallback (servers without MLSD)
probes each entry with SIZE/MDTM to recover size, mtime and a
best-effort type.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Koha::File::Transport::FTP::_list_files parsed the LIST / "ls -l" output
with a non-greedy date capture that only consumed the first of the three
date tokens (the month), leaving the day and time glued to the front of
the filename. Every filename returned for an FTP transport was therefore
prefixed with the listing date, e.g.
-rw-r--r-- 1 user group 1234 Jan 01 12:00 QUOTES_413514.CEQ
=> filename "01 12:00 QUOTES_413514.CEQ"
Callers (Koha::Edifact::Transport, marc_ordering_process.pl) then asked
the server for a file that does not exist, so no files were downloaded,
renamed or deleted over FTP. SFTP and Local transports were unaffected.
This regressed when the EDI transport was moved onto the file_transports
system (bug 38489). The previous Koha::Edifact::Transport::ftp_download
listed files with Net::FTP->ls (NLST, bare names); the refactor switched
to Net::FTP->dir (LIST) so it could return the richer hashref shape used
by the SFTP and Local transports, but the parse was faulty.
The date field in "ls -l" output is a fixed run of three whitespace
separated tokens (Mon DD HH:MM or Mon DD YYYY). Consume it explicitly so
the filename is captured cleanly, preserving spaces and still skipping
directories. The LIST-based approach and the filename/longname/size/perms
return shape are kept so the structure stays consistent across transport
classes (relied on by consumers, including out-of-tree ones).
Test plan:
1) prove t/db_dependent/Koha/File/Transport/FTP.t
2) git checkout Koha/File/Transport/FTP.pm and re-run: the list_files()
subtest fails on the filename/size/perms assertions
3) prove t/db_dependent/Koha/Edifact/Transport.t
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The default `_match=contains` matching builds a LIKE query even on
integer columns, so a plain `?object=42` also matches object values
like 420 or 142. The test's dynamically-assigned patron borrowernumber
(auto_increment, never reset by transaction rollback) could drift into
a value containing "42" as a substring, intermittently inflating the
match count in CI. Add `_match=exact` since the test's intent is an
exact object id match.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
To test:
- cypress run --spec t/cypress/integration/KohaTable/PatronSearch_spec.ts
- cypress run --spec t/cypress/integration/KohaTable/KohaTable_spec.ts
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
local_holdgroup_match was being hardcoded to 1 in the two LocalHoldsPriority
allocation paths (GiveLibrary and GiveLibraryGroup), regardless of whether the
chosen item actually satisfied the LocalHoldsExclusivity conditions. This
caused staff to be directed to pull an item that would not trigger the hold at
checkin when the two features used different patron/item controls.
The fix removes the premature local_holdgroup_match => 1 from both LHP paths and
lets _flag_local_holdgroup_matches evaluate all allocated items using the correct
LocalHoldsExclusivity* sysprefs.
Also fixes a pre-existing missing bind-variable placeholder in the
INSERT INTO tmp_holdsqueue statement (17 columns, only 16 ?).
Test updates:
- Existing local_holdgroup_match test gains the matching LocalHoldsExclusivity
sysprefs so the flag is set via the correct code path.
- New subtest reproduces the reported bug: LHP (HomeLibrary/homebranch) targets
a non-local item while exclusivity (PickupLibrary/holdingbranch) should not
flag it — confirms local_holdgroup_match is 0 after the fix.
Signed-off-by: Trevor Diamond <trevor.diamond@mainlib.org>
Sponsored-by: Main Library Alliance <https://www.mainlib.org/>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
LocalHoldsExclusivity and LocalHoldsPriority are adjacent but distinct
features: priority alters *which patron's hold* gets filled first, while
exclusivity restricts *which items* may fill a hold during a time window.
Each should be usable independently of the other.
Changes:
- Rename LocalHoldsPriorityExclusivityPeriod to LocalHoldsExclusivityPeriod
- Add LocalHoldsExclusivityPatronControl (PickupLibrary/HomeLibrary)
- Add LocalHoldsExclusivityItemControl (homebranch/holdingbranch)
- HoldsQueue: add _flag_local_holdgroup_matches() that flags
local_holdgroup_match on allocated items using hold-group matching
and the new exclusivity control prefs, regardless of LocalHoldsPriority
- Reserves: CheckReserves exclusivity check now computes its own
"is this item local" using the new prefs and hold-group siblings,
independent of the LocalHoldsPriority matching variables
- Atomicupdate handles both fresh install and rename-from-old-pref paths
- Pref UI note clarifies independence from LocalHoldsPriority
Test plan:
1. Apply patch, run updatedatabase.pl
2. Verify three new sysprefs exist in Administration > System preferences
> Circulation: LocalHoldsExclusivityPeriod, LocalHoldsExclusivityPatronControl,
LocalHoldsExclusivityItemControl
3. Set LocalHoldsPriority to "Don't give" (None)
4. Create a library hold group with at least two libraries (e.g. Lib-A, Lib-B)
5. Create a third library (Lib-C) NOT in the hold group
6. Create an item at Lib-B
7. Place a hold for a patron at Lib-A (pickup at Lib-A)
8. Set LocalHoldsExclusivityPeriod to 7, PatronControl to HomeLibrary,
ItemControl to homebranch
9. Run the holds queue builder (misc/cronjobs/holds/build_holds_queue.pl)
10. Verify hold_fill_targets.local_holdgroup_match = 1 for that hold
(item at Lib-B is in the same hold group as patron at Lib-A)
11. Check in the item at Lib-C — it should NOT fill the hold (exclusivity
blocks the non-local item within the 7-day window)
12. Check in the item at Lib-B — it SHOULD fill the hold (local match)
13. Reset: delete the hold, create a new one with a patron at Lib-C
(pickup at Lib-C). Run the queue again.
14. Verify hold_fill_targets.local_holdgroup_match = 0 for this hold
15. The item at Lib-B should now fill this hold at Lib-C checkin
(non-local patron, no exclusivity protection)
16. Set LocalHoldsExclusivityPeriod to 0 — verify non-local items can
fill holds normally regardless of hold_fill_targets flag
17. prove t/db_dependent/Holds/LocalHoldsPriority.t — passes
18. prove t/db_dependent/HoldsQueue.t — passes
19. prove t/db_dependent/Reserves.t — passes
Signed-off-by: Trevor Diamond <trevor.diamond@mainlib.org>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Covers the new LocalHoldsPriorityExclusivityPeriod behaviour and keeps
existing tests working with the schema change:
- t/db_dependent/Holds/LocalHoldsPriority.t: new subtest exercising
all four cases -- pref disabled, pref active on a fresh hold, pref
active past the window, and pref active with no hold_fill_targets
row (queue hasn't run).
- t/db_dependent/HoldsQueue.t: assert that after CreateQueue() with a
local-group match, both tmp_holdsqueue and hold_fill_targets rows
carry local_holdgroup_match = 1.
- t/db_dependent/Reserves.t, t/db_dependent/Koha/Hold.t: qualify the
two direct "INSERT INTO hold_fill_targets VALUES (...)" statements
with explicit column names so they continue to work now that
hold_fill_targets has one extra column.
Test plan:
prove t/db_dependent/Holds/LocalHoldsPriority.t \
t/db_dependent/HoldsQueue.t \
t/db_dependent/Reserves.t \
t/db_dependent/Koha/Hold.t
All tests should pass.
Sponsored-by: Main Library Alliance <https://www.mainlib.org/>
Signed-off-by: Trevor Diamond <trevor.diamond@mainlib.org>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
t/db_dependent/Log.t
- Add subtest verifying cronlogaction() prefixes loginfo with \$0
(the running script path), appends the info parameter, defaults the
action to 'Run', and respects custom action values.
t/Koha/Script.t
- Assert that \$Koha::Script::_cron is false when the module is loaded
without the -cron flag, confirming INIT/END blocks will not log.
t/Koha/Script_cron.t
- Assert that \$Koha::Script::_cron is true when the module is loaded
with the -cron flag, confirming INIT/END blocks will log.
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch implements the library_limit concept for Report/Reports. Staff with the correct permission and superlibrarians can set libaries to which the report should be visible, if the system preference is enabled.
Test plan:
Confirm Initial Settings
1. Log in to Koha staff with username: koha and password: koha and set your library as Centerville if it's not already.
2. Visit the Saved reports page, and create and save three reports: Report 1, Report 2, and Report 3. The SQL query for each report doesn't matter, it should just be something valid.
3. Visit the Patrons page and create two staff users: StaffReportLimitUser and StaffNoLimitPermission. Set their home libraries to Centerville and Fairfield, respectively.
4. Under Allow access to the reports module, ensure both users have report management permissions for Create, Delete, Execute.
Verify Patch Settings
5. Apply the patch and restart services. If using Koha Testing Docker (KTD): from the Koha repo directory, apply the patch (e.g. `git bz apply 16631`, or check out the patched branch with `git fetch && git checkout <branch>`), then run `ktd --shell` to enter the KTD shell and run `restart_all`. If the patched branch is already checked out, just enter the KTD shell and run `restart_all`.
6. Click Patrons in the top navbar to go to the Patrons page.
7. Search for "StaffReportLimitUser" in the patron search field.
8. Click StaffReportLimitUser to edit the patron record for this user.
9. Click More -> Set permissions to access the permissions screens for the StaffReportLimitUser.
10. In the row for Allow access to the reports module (reports), click Show details to view relevant permissions.
11. Confirm Manage Report Limits (manage_report_limits) is an option under Allow access to the reports module (reports).
12. Click the checkbox for Manage Report Limits to give this user this permission, and click Save.
13. Go to the System Preferences page (Click the More dropdown from the top nav bar, click Administration, click System Preferences).
14. Verify Reports is an option in the left side navigation bar, and that when clicked, you see the LimitReportsByBranch option.
15. Enable LimitReportsByBranch as described in step 14.
16. Go back to Saved reports as in Step 2.
17. Click Edit in the row menu for Report 1.
18. At the bottom of the Report 1 edit page, you should see a Library limitation select box.
Verify the Library Limitation Settings are Persisted
19. Choose Centerville from the select box and click Update SQL.
20. Reload the page, or navigate away from the page and go back to the edit screen for Report 1.
21. Verify that Centerville is still in the Library limitation select box.
22. Add Fairfield in the Library limitation select box and click Update and run SQL.
23. Go back to editing Report 1.
24. Confirm the Library limitation select box has both Centerville and Fairfield selected.
25. Duplicate Report 1.
26. Confirm the Library limitation select box has both Centerville and Fairfield selected.
27. Click Save Report.
28. Confirm the Library limitation select box has both Centerville and Fairfield selected.
Verify Reports are Limited Based on Library Limitations
29. Edit Report 1 to add Centerville in the Library limitation select box as the only choice and click Update SQL.
30. Edit Report 2 to add Fairfield in the Library limitation select box as the only choice and click Update SQL.
31. Report 3 should have no library limitations, so it should be visible to all libraries.
32. Go to Saved reports as in Step 2.
33. Confirm you do not see Report 2, and only Report 1 and Report 3.
34. In the upper right corner, click the drop down that says koha Centerville and click Set library to change your library.
35. Set your library to Fairfield.
36. Go to Saved reports as in Step 2.
37. Confirm you do not see Report 1, and only Report 2 and Report 3.
Verify Disabling LimitReportsByBranch Preference Shows All Reports
38. Disable LimitReportsByBranch preference as in steps 13 to 15.
39. Go to Saved reports as in Step 2.
40. Confirm Report 1, Report 2, Report 3 are visible.
41. Click to edit Report 1 and verify there is no Library limitation select box.
Verify Library limitations are Persisted When EnableFilteringReports is Toggled On or Off
42. Enable LimitReportsByBranch preference as in steps 13 to 15.
43. Ensure your library is set to Fairfield as in step 35.
44. Go to Saved reports as in Step 2.
45. Confirm Report 1 is not visible, and Report 2 and Report 3 are visible.
46. Click to edit Report 2 and verify the Library limitation select box is at the bottom of the page and that it has Fairfield selected.
Verify the Show all reports option shows all reports for users with the Manage Report Limits permission and superlibrarians
47. Go to Saved reports as in Step 2 and in the Show reports dropdown, select Show all reports, and click Apply filter.
48. Confirm the page reloads with all reports visible: Report 1, Report 2, and Report 3 should be visible.
49. Select Show my branch reports from the dropdown and click Apply filter.
50. Confirm Report 1 is not visible, and Report 2 and Report 3 are visible (you should only see reports for your branch, Fairfield).
Verify Report filters return the correct results
51. Go to Saved reports as in Step 2.
52. Verify you're viewing only your branch's reports in the filter.
53. Under Filter in the left sidebar, type "2" in the Keyword field and click Apply filter.
54. Confirm only Report 2 is visible.
55. Under Filter in the left sidebar, type "1" in the Keyword field and click Apply filter.
56. Confirm there are no reports visible, and the message "No saved reports match your criteria" is displayed.
Verify Library Limits column is available to superlibrarians and users with the Manage Report Limits permission
57. Go to Saved reports as in Step 2.
58. Click the Columns dropdown and ensure that the Library limits column is available to select for display. Select it.
59. Confirm that the Library limits column is now visible in the list of reports and that it shows the correct library limitations for each report (e.g., Report 1 should show 1 library limitation and Centerville (CPL) on hover, Report 2 should show 1 library limitation and Fairfield (FFL) on hover, and Report 3 should show no limitation).
Verify the Show all reports and Limit reports options are available to users that are not superlibrarian but have the Manage Report Limits permission
60. For the StaffReportLimitUser, add the permission for Manage Report Limits as described in steps 9 to 12. This will allow this user to see the Show all reports and Limit reports options.
61. Log out of the koha admin account. In the upper right corner, click the drop down that says koha Fairfield and click Log out.
62. Log in as StaffReportLimitUser, the previously created staff user with report permissions.
63. Confirm the staff user's library is Centerville in the upper right corner of the screen.
64. Go to Saved reports as in Step 2.
65. Confirm the filter option for Show reports is visible.
66. Click to edit Report 1.
67. Verify there is a Library limitation select box and that it can be interacted with.
Verify the Show all reports and Limit reports options are not available to users that are not superlibrarian and do not have the Manage Report Limits permission
68. Log out of the StaffReportLimitUser account.
69. Log in as StaffNoLimitPermission, the previously created staff user without the Manage Report Limits permission.
70. Confirm the staff user's library is Fairfield in the upper right corner of the screen.
71. Go to Saved reports as in Step 2.
72. Click the Columns dropdown and ensure that the Library limits column is not in the list of available columns to select for display.
73. Confirm the filter option for Show reports is not available.
74. Click to edit Report 2.
75. Verify there is no Library limitation select box at the bottom of the page.
[NEW] Verify library limits are auto-assigned and preserved correctly for non-permissioned users
76. Still logged in as StaffNoLimitPermission (Fairfield) from the previous section. Go to Saved reports and click New report. Enter "Limit Auto-Assign Test" as the report name and SELECT 1 as the SQL. Confirm there is no Library limitations select box visible on the form. Click Save.
77. Go back to Saved reports. Click Edit for Report 2 (limited to Fairfield). Add `-- test` as a trailing SQL comment and click Update SQL. This simulates a non-permissioned user saving an existing report that already has limits.
78. Go back to Saved reports. Click Edit for Report 3 (no library limits). Add `-- test` as a trailing SQL comment and click Update SQL. This simulates a non-permissioned user saving an existing report that has no limits.
79. Log out and log back in as StaffReportLimitUser (Centerville, has manage_report_limits).
80. Go to Saved reports. Use Show all reports if needed. Find "Limit Auto-Assign Test" and click Edit. Confirm the Library limitations select box shows Fairfield as the only selected library — the report was automatically limited to the creator's home library on create.
81. Find Report 2 and click Edit. Confirm the Library limitations select box still shows only Fairfield — the existing limit was preserved when the non-permissioned user saved the report.
82. Find Report 3 and click Edit. Confirm the Library limitations select box is empty — no limits were introduced when the non-permissioned user saved a report that had no existing limits.
Verify limited reports are not visible to users without permissions
83. While still logged in as StaffReportLimitUser from step 79, copy the URL for the edit page of Report 1.
84. Log out of the StaffReportLimitUser account.
85. Log in as StaffNoLimitPermission, the user without the Manage Report Limits permission.
86. Paste the copied URL for the edit page of Report 1 into the browser and navigate to it.
87. Confirm that you see an error message that says "This report is restricted and cannot be accessed by the current user." and that the page does not load the report details.
Run Unit Tests
88. Using Koha Testing Docker, in the Koha home directory run the command: ktd --shell to start the ktd shell.
89. Run command: reset_all to drop and recreate the koha database.
90. Run command prove t/db_dependent/Koha/Reports.t to run the test.
91. This file contains the unit tests for the reports_branches table and the new subtest `library limits enforcement on save and update`, which covers auto-assignment of branch limits on create and preservation of limits on update for non-permissioned users.
92. All tests should pass.
93. Run command prove t/Koha/Auth/Permissions.t
94. Permissions unit tests in this file were updated.
95. All tests should pass.
96. Run command prove t/db_dependent/Reports/Guided.t
97. The Guided report tests were updated to account for the new library limitations and permissions.
98. All tests should pass.
Signoff
99. Sign off if the steps above check out.
Sponsored-by: Ramapo Catskill Library System
Co-authored-by: Casey Conlin <casey.conlin@gmail.com>
Signed-off-by: Emmi Takkinen <emmi.takkinen@koha-suomi.fi>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The ILL Vue app currently reads ILLModule and ILLPartnerCode through the system preferences service. That service requires parameters/manage_sysprefs, so a staff user with ILL permissions but without system preference administration permissions cannot load the ILL module.
Add an ILL-specific /ill/config REST endpoint protected by the ill permission, and switch the ILL Vue app to load its configuration from that endpoint. This follows the module-specific configuration endpoint pattern introduced by bug 33606 for the same manage_sysprefs problem in the ERM Vue app.
Security/compatibility considerations: the endpoint returns only allow-listed, non-secret ILL configuration values needed by the ILL Vue app. It does not expose the generic system preferences API, and /api/v1/sysprefs remains restricted to parameters/manage_sysprefs.
Regression test coverage: the new API test verifies that a staff user with the ill permission can read /ill/config, that a staff user without the ill permission cannot read it, and that the same ILL-only user still cannot access /api/v1/sysprefs.
Test plan:
1. prove -I. -It/lib t/db_dependent/api/v1/ill_config.t
2. prove -I. -It/lib t/db_dependent/api/v1/ill_users.t
Signed-off-by: David Nind <david@davidnind.com>
Amended-by: Jonathan Druart
chmod +x t/db_dependent/api/v1/ill_config.t
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
To test:
1. prove t/db_dependent/Circulation.t
2. Observe the following failures
t/db_dependent/Circulation.t .. 17/87
# Failed test 'Bug 42298: No autorenew before rule is respected'
# at t/db_dependent/Circulation.t line 1134.
# got: '0'
# expected: '1'
# Failed test 'Bug 42298: No autorenew before rule is respected'
# at t/db_dependent/Circulation.t line 1135.
# got: 'too_soon'
# expected: 'auto_renew'
# Failed test 'soonest_renew_date is not returned because this issue can be renewed'
# at t/db_dependent/Circulation.t line 1136.
# got: '2026-04-21T23:59:00'
# expected: undef
# Failed test 'No renewal before is 99, patron opted out of auto_renewal so can renew'
# at t/db_dependent/Circulation.t line 1141.
# got: '0'
# expected: '1'
# Looks like you failed 4 tests of 124.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Laura_Escamilla <laura.escamilla@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Transitioning a booking to 'issued' no longer skipped clash detection,
and Item::check_booking treats any live checkout on the item as a
clash. Switching an on-site checkout of a booked item to a regular
checkout (SwitchOnSiteCheckouts) therefore crashed AddIssue with an
unhandled Koha::Exceptions::Booking::Clash; collection could fail the
same way whenever the biblio's bookable item pool had shrunk since the
booking was placed.
- short-circuit no-op stores before validation; circulation issues
redundant status writes that must not re-trigger the checks
- rename _is_final_status_transition to _skip_clash_detection and also
skip when transitioning to 'issued' from an active status, as long
as the booked period and item are unchanged
- keep validating period/item changes and reactivations of terminal
bookings
- cover the on-site switch in Circulation.t and the store() semantics
in Booking.t
Test plan:
1. Enable OnSiteCheckouts and SwitchOnSiteCheckouts
2. Place a booking starting today for a patron
3. Check the booked item out to that patron from Circulation > Check
out with 'On-site checkout' ticked. Do NOT use the bookings page's
'Transform to checkout' button: it submits a fixed field set that
does not carry the on-site flag (Bug 36789), so the checkout would
silently be a regular one and step 4 could not reproduce the switch
4. The booking shows 'Issued'. Scan the same barcode for the same
patron again to switch the on-site checkout to a regular one; this
crashed with Koha::Exceptions::Booking::Clash before the patch and
now succeeds with the booking still 'Issued'
5. prove t/db_dependent/Koha/Booking.t t/db_dependent/Circulation.t
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Always include the plural version, not the singular one (the plural
imports the singular)
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Laura_Escamilla <laura.escamilla@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch replaces all calls to C4::Stats::UpdateStats with the wrapped
function from the Koha namespace. As this is the last function in C4::Stats, the
module is also deleted.
Several tests depended on transitive includes to Koha::Statistics through
C4::Stats. I've identified uses of that module by searching, and added the
necessary use in each case. I've run `prove -r t xt` with this patch applied and
didn't hit any relevant errors.
Test plan:
1. Inspect the patch, and verify each changed call maintains the same parameters.
2. Apply patch.
3. Prove the following files:
- t/db_dependent/Accounts.t
- t/db_dependent/Circulation.t
- t/db_dependent/Koha/Account.t
- t/db_dependent/Koha/ILL/Request.t
- t/db_dependent/Koha/Item.t
- t/db_dependent/Koha/Pseudonymization.t
- t/db_dependent/Koha/Recalls.t
Sponsored-by: Open Fifth
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Laura_Escamilla <laura.escamilla@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
To test:
1. Set both system preferences NotifyBorrowerDeparture and
MembershipExpiryDaysNotice to 10 days
2. Create or edit a patron account with the following settings:
- expiry date is exactly 10 days in the future
- add an email address on the patron account
- set messaging preferences to receive email notices for Patron
account expiry
3. Run the membership expiry cron job to generate a notice:
ktd --shell
perl misc/cronjobs/membership_expiry.pl -c
4. Check notices tab on patron account
--> A membership expiry notice has been queued for the patron
5. Click back to the patron details
--> Note that there is no alert that the patron account will be expiring
soon
6. Apply patch and restart_all
7. Refresh the patron details page
--> You should now see an alert that the account will be expiring soon
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
t/db_dependent/Holds.t, t/db_dependent/ILSDI_Services.t, and
t/db_dependent/api/v1/holds.t all fail if a site has a global
'Holds Allowed (Daily)' circulation rule configured, because
C4::Reserves::CanItemBeReserved checks holds_per_day before
holds_per_record/reservesallowed, so an ambient daily-cap rule
returns tooManyReservesToday and masks the status these tests
actually exercise (tooManyReserves, noReservesAllowed, OK, or the
API's hold_limit override). The affected subtests never set
holds_per_day themselves, so they unintentionally depend on the
database having no such rule configured.
Test plan:
1. On a fresh db, set a global 'Holds Allowed (Daily)' circulation
rule (Administration > Circulation and fine rules) to 1.
2. prove t/db_dependent/Holds.t t/db_dependent/ILSDI_Services.t
t/db_dependent/api/v1/holds.t
=> fails before this patch, passes after.
3. Remove the circulation rule and re-run the same tests.
=> still passes (no regression for the default/unconfigured case).
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
INSERT INTO action_logs (timestamp, user, module, action, object, info, interface) VALUES
(NOW(), 5, 'CIRCULATION', 'ISSUE', 5,
'{"issue":1,"branchcode":"CPL","itemnumber":1,"confirmations":["TOO_MANY"],"forced":["DEBT"]}', 'intranet'),
(NOW(), 5, 'CIRCULATION', 'RETURN', 5,
'1', 'intranet'),
(NOW(), 5, 'HOLDS', 'CREATE', 7,
'{"hold":7,"biblionumber":1,"itemnumber":1,"confirmations":["HIGHHOLDS"],"forced":["HOLD_POLICY_OVERRIDE"]}', 'intranet');
Test plan:
1) Without the patch applied, run the populate SQL above against the kohadev database.
2) Go to Tools > Log viewer
3) Under Modules, select only Circulation - leave the date range empty and click Submit.
4) Observe the ISSUE row's Info column shows the raw JSON payload as plain text, and the RETURN row shows the bare itemnumber "1" as plain text - neither is a clickable link to the item.
5) Go back, select only the Holds module, click Submit, and observe the CREATE row's Info column likewise shows raw JSON with no hold/item links.
6) Apply the patch.
7) Run yarn build && restart_all, then hard refresh the browser.
8) Repeat step 3.
9) Confirm the ISSUE row's Info column now shows a link reading "Item <barcode> CPL" (e.g. Item 3999900000001 CPL) - the branchcode comes from the JSON payload - and clicking it opens the item page. Override information is also displayed correctly
10) Confirm the RETURN row's Info column now shows a link reading "Item <barcode>" (no branchcode) opening the same item's detail page - verifying the bare-itemnumber payload path.
11) Repeat step 5. Confirm the CREATE row now shows "Hold 7 on biblio 1" linking to reserve/request.pl?biblionumber=1, followed by " - Item 1" linking to the item detail page. Override information is also displayed correctly
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds Koha::REST::Plugin::CSRF which enforces CSRF token
validation on state-changing requests (POST/PUT/DELETE/PATCH) when
a CGISESSID cookie is present.
Requests using OAuth2 or HTTP Basic Auth are not affected since they
don't use session cookies and are inherently immune to CSRF.
The client-side already sends the CSRF-TOKEN header via http-client.js
(added in Bug 34478). This patch adds the server-side enforcement.
Test plan:
1. Apply patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/Koha/REST/Plugin/CSRF.t
=> SUCCESS: Tests pass!
3. Run:
k$ prove t/db_dependent/api/v1/csrf.t
=> SUCCESS: Tests pass!
4. Tests cover:
- GET/HEAD/OPTIONS requests bypass CSRF check
- POST/PUT/DELETE without CGISESSID bypass CSRF check
- POST with CGISESSID but no CSRF-TOKEN returns 403
- POST with CGISESSID and invalid CSRF-TOKEN returns 403
- POST with CGISESSID and valid CSRF-TOKEN passes through
- PUT and DELETE enforcement
5. Sign off :-D
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch checks for waiting or in_transit items allocated to recalls, and excludes them from the Holds to Pull report.
To test:
1. Enable UseRecalls and set the recalls-related circulation rules
2. Check out an item to Patron A
3. Place an item-level hold on the item for Patron B
4. Log into the OPAC as Patron C and search for the item. Place an item-level recall on the item.
5. Go back to the staff interface and check in the item. This should prompt a pop-up to confirm the recall as waiting. Confirm the recall as waiting.
6. Go to Circulation -> Holds to pull.
7. Notice that even though the item is allocated for a recall, the hold is showing. It should not be showing in this page.
8. Apply the patch, restart services and refresh the page
9. The item allocated to a recall should no longer be showing in the Holds to pull report
10. Go to the record detail page and go to the Recalls tab
11. Revert the recall from its waiting status
12. Go back to the Holds to pull page. The record will now show for the hold.
13. Go to the Circulation -> Check in page
14. Go to the menu in the top-right of the page and click Set library
15. Change your library to a different one
16. Check in the item to confirm the recall as waiting and trigger a transfer
17. Go back to the Holds to pull page
19. The item should no longer show on the Holds to pull report
20. Confirm tests pass t/db_dependent/Koha/Holds.t
Sponsored-by: Catalyst IT
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
When RealTimeHoldsQueueUnallocated is enabled, the background job
passes unallocated => 1 and delete => 0 to update_queue_for_biblio.
This skips deleting existing queue entries and only processes holds
not already allocated, matching the --unallocated behavior of
build_holds_queue.pl.
Test plan:
1. Apply patches
2. Run:
$ ktd --shell
k$ prove t/db_dependent/Koha/BackgroundJob/BatchUpdateBiblioHoldsQueue.t
=> SUCCESS: Tests pass!
3. Tests verify:
- delete => 1 and no unallocated when pref is off
- delete => 0 and unallocated => 1 when pref is on
4. Sign off :-D
Signed-off-by: Alex Carver [Acerock7] <alex@rcls.org>
Sponsored-by: The Main Library Alliance <https://www.mainlib.org/>
Signed-off-by: Alex Carver [Acerock7] <alex@rcls.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch resolves three QA findings on the cashup reconciliation
work:
* Add a CSP nonce to the inline <style> block in cashup_summary.inc
so it is not flagged by the forbidden_patterns check.
* Document the private _get_session_end method on
Koha::Cash::Register::Cashup so pod_coverage passes.
* Mock CashupReconciliationNoteRequired to 0 in
t/db_dependent/Koha/Cash/Register/Cashup.t so the test does not
depend on the site preference value (it previously failed when
the syspref was set to "Require" because surpluses and deficits
are created without a reconciliation note in this test).
Test plan:
1. Apply the patch.
2. Set CashupReconciliationNoteRequired to "Require".
3. Run prove t/db_dependent/Koha/Cash/Register/Cashup.t -- it should
pass.
4. Set CashupReconciliationNoteRequired to "Don't require" and
re-run -- it should still pass.
5. Run koha-qa.pl on the bug branch -- pod_coverage on
Koha/Cash/Register/Cashup.pm and forbidden_patterns on
cashup_summary.inc should now pass.
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: Jackie Usher <jackie.usher@westsussex.gov.uk>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds configuration options for cashup reconciliation and
enhances the user interface with preview functionality and improved
interaction patterns.
Configuration Options:
1. CashupReconciliationNoteRequired (System Preference):
- Optional validation requiring reconciliation notes when discrepancies exist
- Ensures documentation of surplus/deficit causes when needed
- Can be disabled for less stringent workflows
- Applied at cashup completion time
2. CashupReconciliationNoteAuthorisedValue (System Preference):
- Allows use of authorized value dropdown for reconciliation notes
- Provides standardized note options (e.g., "Till count error", "Cash removed")
- Falls back to free text textarea when not configured
- Improves consistency and reporting capabilities
UI Enhancements:
1. Cashup Summary Preview:
- Preview cashup summary for in-progress cashups
- Shows current transaction totals before completion
- Clearly indicates preview vs. final summary status
- Helps staff verify amounts before committing
2. Improved User Experience:
- Conditional note fields based on configuration
- Authorized value dropdowns when configured
- Clear validation messages
- Responsive UI updates
Backend changes:
- Enhanced Koha::REST::V1::CashRegisters::Cashups with preview support
- Database atomicupdate script for system preferences
- Validation logic for required notes
Frontend changes:
- Preview modal functionality in cashup_modal.js
- Conditional rendering of note input fields
- Authorized value dropdown integration
- Enhanced summary display logic
Test plan:
1. Apply patches and run database update
2. Configure CashupReconciliationNoteRequired:
- Enable preference
- Perform cashup with discrepancy without note - should fail
- Add note and retry - should succeed
- Disable preference - verify note optional
3. Configure CashupReconciliationNoteAuthorisedValue:
- Create AV category (e.g., CASHUP_NOTE) with values
- Set preference to category name
- Verify dropdown appears in cashup modal
- Clear preference - verify textarea shown
4. Test cashup preview:
- Start cashup on register
- Click "Preview cashup summary"
- Verify current totals displayed
- Add transaction and re-preview - verify updated
- Complete cashup - verify final summary
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: Jackie Usher <jackie.usher@westsussex.gov.uk>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch implements an optional two-phase cashup workflow for point
of sale operations, along with improved error handling and validation.
Two-Phase Cashup Workflow:
- Staff can initiate cashup (CASHUP_START) and complete later
- Prevents new transactions on registers with cashups in progress
- Tracks cashup sessions with start/end timestamps
- Supports both quick cashup (immediate) and staged cashup workflows
- Updates batch cashup workflow operations across multiple registers
Error Handling & Validation:
- Centralized exception handling for cashup operations
- Clear validation messages for missing parameters
- Prevents cashup when no cash transactions exist
- Improved handling of zero/negative amounts
- Informative error messages for user guidance
Negative Amount Support:
- Supports negative cashup amounts for cashup deficits
- Automatic detection when actual cash is less than expected amount
- Creates appropriate CASHUP_DEFICIT records
- UI handles both positive and negative reconciliation amounts
- Updated calculations and display logic throughout
Modal Refactoring:
- Eliminates code duplication between register.tt and registers.tt
- Centralized cashup modal functionality in cashup_modals.js
- Shared confirm_cashup.inc and trigger_cashup.inc templates
- Consistent user experience across all cashup workflows
- Improved maintainability and code organization
Backend changes (Koha::Cash::Register):
- start_cashup(): Creates CASHUP_START action to begin session
- cashup_in_progress(): Detects active cashup sessions
- add_cashup(): Enhanced with two-phase completion support
- outstanding_accountlines(): Respects cashup session boundaries
Frontend changes:
- Interactive modals for starting and completing cashups
- Real-time calculation displays
- Support for reconciliation with actual amounts
- Batch operations UI for multiple registers
- Responsive error messaging and validation feedback
Test plan:
1. Apply patches and restart services
2. Run prove t/db_dependent/Koha/Cash/Register.t
3. Run prove t/db_dependent/Koha/Cash/Register/Cashup.t
4. Test single register workflows:
- Start cashup, add transactions, verify blocked
- Complete cashup with reconciliation amounts
- Test with positive, negative, and zero amounts
5. Test multi-register workflows:
- Select multiple registers on registers page
- Perform batch cashup operations
- Verify each register processes correctly
6. Test error conditions:
- Attempt cashup with no transactions
- Attempt cashup with invalid amounts
- Verify clear error messages displayed
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: Jackie Usher <jackie.usher@westsussex.gov.uk>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The previous follow-up restored warn in Object::store but missed
several test files that trigger DB exceptions via ->store(). These
tests now fail Test::NoWarnings because the warning is emitted but
not expected.
This patch adds warning_like wrappers to 15 additional test files,
asserting that the expected DB exception warning is emitted.
Test plan:
1. Apply patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/api/v1/biblios.t \
t/db_dependent/Koha/Holds.t \
t/db_dependent/api/v1/return_claims.t \
t/db_dependent/api/v1/libraries.t \
t/db_dependent/Koha/Booking.t \
t/db_dependent/api/v1/patrons.t \
t/db_dependent/SIP/Message.t \
t/db_dependent/Koha/Patron.t \
t/db_dependent/Koha/Acquisition/Fund.t \
t/db_dependent/api/v1/acquisitions_orders.t \
t/db_dependent/Koha/ILL/Requests.t \
t/db_dependent/Koha/Patron/Relationship.t \
t/db_dependent/Koha/Hold.t \
t/db_dependent/Circulation/Returns.t \
t/db_dependent/api/v1/bookings.t
=> SUCCESS: Tests pass!
3. Sign off :-D
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Bug 42391 removed the warn from Koha::Object::store when it moved
exception translation to the exception_action hook. This lost useful
debugging information from the Plack error logs.
This patch restores the warning by adding a try/catch in store() that
stringifies and warns the exception before rethrowing. The warning now
contains the translated Koha exception message (e.g. 'Broken FK
constraint') rather than the raw DBI error, since translation happens
before the exception reaches store().
Note: delete() does not warn. Ideally the logging would live in the
ExceptionMapper itself (via an injected logger), covering all code paths
uniformly. That is a larger refactor tracked separately.
Tests that assert on these warnings are updated to match the new
message format.
Test plan:
1. Apply patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/Koha/Object.t \
t/db_dependent/Koha/Objects.t \
t/db_dependent/Koha/Schema/ExceptionMapper.t \
t/db_dependent/Koha/Item.t \
t/db_dependent/Koha/Suggestions.t \
t/db_dependent/Koha/Account.t \
t/db_dependent/Koha/ApiKey.t \
t/db_dependent/Koha/Checkouts/ReturnClaim.t \
t/db_dependent/Circulation/MarkIssueReturned.t \
t/db_dependent/Suggestions.t \
t/db_dependent/Biblio.t
=> SUCCESS: Tests pass!
3. Sign off :-D
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Bug 41604 made every page do a bit more work when it loads. This
Selenium test didn't wait for that work to finish before typing into
the search box, so sometimes it searched too early and got the wrong
results.
This adds a short wait after each page load so the test only searches
once the page is actually ready.
Introduced by: Bug 41604: Save user decision per column when visibility changes
Test plan:
prove t/db_dependent/selenium/patrons_search.t
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Warning: The allowCypressEnv configuration option is enabled. This allows any browser code to read values from Cypress.env(). This is insecure and will be removed in a future major version.
1. Replace Cypress.env() calls with cy.env() (for sensitive values) or Cypress.expose() (for public configuration)
2. Set allowCypressEnv: false in your Cypress configuration to disable Cypress.env()
Learn more: https://on.cypress.io/cypress-env-migration
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The Koha::Schema::Util:: namespace adds unnecessary nesting for a
single-purpose class. This follow-up moves it to
Koha::Schema::ExceptionMapper, which sits alongside Koha::Schema
itself — a better fit given it's tightly coupled to the Schema's
exception_action hook.
Changes:
- Koha::Schema::Util::ExceptionTranslator → Koha::Schema::ExceptionMapper
- Update Koha::Schema to use the new module
- Move test file accordingly
- Remove now-empty Koha/Schema/Util/ directory
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Extend the ExceptionTranslator to handle two additional MySQL error
patterns:
- NOT NULL violations ('Column X cannot be null' and 'X doesn't have
a default value') → Koha::Exceptions::Object::NotNull
- Not in database errors → Koha::Exceptions::Object::NotInStorage
Add Koha::Exceptions::Object::NotNull exception class with a property
field identifying the offending column.
Add Koha::Exceptions::Object::UnhandledDBError as a catch-all for DBI
errors that don't match any known pattern. This closes the gap where
unmatched MySQL errors could leak as DBIx::Class::Exception to callers
(e.g. templates that no longer catch that type).
The exception_action handler now distinguishes three cases:
- Koha exception objects (from txn_do re-throws): passed through via
'die $msg' to preserve the original exception type. Without this,
txn_do's rollback-and-rethrow would stringify the Koha exception
and wrap it in UnhandledDBError, losing the specific type.
- DBI errors (prefixed 'DBI Exception:'): translated if possible,
otherwise wrapped in UnhandledDBError.
- DBIC internal errors (e.g. 'Not in database'): translated if
possible, otherwise thrown as DBIx::Class::Exception.
Note: NotInStorage translation works for update operations but not for
delete-on-detached-Row where DBIC bypasses the Schema's exception_action.
Those cases remain DBIx::Class::Exception.
Update tests:
- OAIServers.t, Z3950Servers.t: assert NotNull with property check
- Hold.t: assert NotNull, remove warn suppression hack
- Objects.t: update-not-in-storage cases assert NotInStorage
- Upload.t: delete-not-in-storage stays DBIx::Class::Exception
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The 'Delete a patron that cannot be deleted' test (case 3) was using a
stale $r_us object from case 2 (not in storage), so it was not actually
testing FK constraint deletion. The assertion on DBIx::Class::Exception
passed by accident.
This follow-up:
- Fetches a fresh Koha::Patron object for case 3
- Clears $e_us/$e_them before the new test
- Asserts Koha::Exceptions::Object::FKConstraintDeletion, which is what
exception_action now produces for FK deletion errors
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
DBIx::Class provides an exception_action hook that intercepts every
throw_exception call before it becomes a DBIx::Class::Exception. By
installing this hook on Koha::Schema via a connection() override, all
database exceptions are automatically translated into Koha exceptions
at the source.
This eliminates the need for translate_exception catch blocks in every
caller:
- Remove try/catch from Koha::Object::store
- Remove try/catch from Koha::Object::delete
- Remove try/catch from Koha::Item::add_to_bundle
- Remove translate_exception convenience method from Koha::Schema
- Remove safe_do (already removed in previous follow-up)
ExceptionTranslator is refactored to work on plain error message strings
instead of DBIx::Class::Exception objects, since exception_action
receives the message before wrapping.
Unmatched errors fall through to DBIx::Class::Exception->throw() as
the default, preserving backward compatibility.
The warn $_->{msg} that was in Object::store is no longer needed since
exceptions are translated before they propagate. Tests that asserted on
that warning (via warning_like) are updated to use throws_ok directly.
Test plan:
1. Apply patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/Koha/Schema/Util/ExceptionTranslator.t
=> SUCCESS: Tests pass!
k$ prove t/db_dependent/Koha/Object.t
=> SUCCESS: Tests pass!
k$ prove t/db_dependent/Koha/Item.t
=> SUCCESS: Tests pass!
k$ prove t/db_dependent/Koha/Patron/Modifications.t
=> SUCCESS: Tests pass!
3. Sign off :-D
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Bug 21820 made the Zebra queue update conditional on the new
ElasticsearchEnableZebraQueue system preference, which defaults to off.
The 'basic_quote_processing' subtest in t/db_dependent/Koha/EDI.t counts
the log lines process_quote emits, and two of those are the
"specialUpdate biblioserver" traces that ModZebra logs only when the
Zebra queue path runs. When the SearchEngine preference is set to
Elasticsearch, that path no longer runs by default, so the subtest now
records 6 log lines instead of 8 and fails:
# Failed test '8 log lines recorded for passed quote file'
# got: '6'
# expected: '8'
# Failed test 'Trace recorded biblioserver index'
Test Plan:
1) Set the SearchEngine system preference to 'Elasticsearch'
2) prove t/db_dependent/Koha/EDI.t
3) Note the basic_quote_processing subtest fails with "got: '6', expected: '8'"
4) Apply this patch
5) prove t/db_dependent/Koha/EDI.t
6) Note the tests pass!
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
cy.get("#report_builder").contains("Submit").click();
cy.intercept("GET", "/api/v1/erm/eUsage/monthly_report/*").as(
"reportRequest"
);
cy.wait("@reportRequest");
We should intercept before the click.
Test plan:
yarn cypress run --spec t/cypress/integration/ERM/UsageReports_spec.ts
Run it several times and confirm that it always passes
The failure was:
<testcase name="Custom reports Should show the Display by year filter when filter is required" time="0.000" classname="Should show the Display by year filter when filter is required">
<failure message="Timed out retrying after 10000ms: `cy.wait()` timed out waiting `10000ms` for the 1st request to the route: `reportRequest`. No request ever occurred.
https://on.cypress.io/wait" type="CypressError"><![CDATA[CypressError: Timed out retrying after 10000ms: `cy.wait()` timed out waiting `10000ms` for the 1st request to the route: `reportRequest`. No request ever occurred.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
1) Set system preference 'PatronDuplicateMatchingAddFields' to empty (no fields ticked)
2) Go to <staff_url>/cgi-bin/koha/members/memberentry.pl?op=add and fill in the mandatory fields for a brand new patron (one that definitely doesn't exist yet)
3) Click 'Save'
4) Notice you get the 'Duplicate patron record?' dialog, even though no patron like this exists
5) Apply the patch. Restart plack
6) Repeat steps 2-3 with another new patron
7) Confirm the patron is created fine this time, no false duplicate error
8) Also try setting 'PatronDuplicateMatchingAddFields' to a field (e.g. 'city') and submit a new patron with that field left blank - confirm no false duplicate there either
9) Run:
prove t/db_dependent/Koha/Patrons.t
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The bug this fixed only manifests when the library's configured
timezone (C4::Context->tz) is not UTC: Koha::Object::_recursive_fixup
converts an incoming RFC3339 instant to that timezone before storing
it in the timezone-naive 'bookings.start_date'/'end_date' columns, so
a UTC-anchored boundary shifts by the timezone offset on write.
None of the existing Cypress specs catch this, because they never run
with the browser/test-runner timezone diverging from the configured
library timezone - both always come from the same container clock.
Add a Perl-level API test that mocks C4::Context->tz to a non-UTC zone
(America/New_York) and asserts the boundaries sent by place_booking.js
round-trip to the correct library-local date, pinning this regression
independently of any particular environment's timezone.
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
When there is a filter with the "status" param, we add a validation step
to secure the use of such a filter.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Note that t/Makefile.t and xt/check_makefile.t were doing the same test.
Merging them.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
$Test::Strict::TEST_STRICT = 0 is only set for all_perl_files_ok
It's no longer used since bug 42126
Test plan:
prove t/db_dependent/00-strict.t
should still return green
Remove a use Modern::Perl for mainpage.pl
=> Should still return green
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Bug 40866 introduced JSON logging for HOLDS CREATE but stored null for
'confirmations' and 'forced' when no overrides were present. This patch
mirrors the earlier change to CIRCULATION ISSUE by defaulting them to
empty arrays, keeping the two log streams structurally identical and
simplifying reporting queries.
Changes:
C4/Reserves.pm:
- Default confirmations/forced to [] in the AddReserve log payload.
t/db_dependent/Reserves.t:
- Update the Bug 40866 subtest to assert the empty-array defaults on
normal holds, and fix a 'my $log_data' redeclaration.
tools/viewlog.pl:
- Extend the CSV export JSON-decode block to also cover HOLDS logs,
guarding the item lookup on itemnumber being present (HOLDS payloads
may lack one) and filling the CSV item columns from the decoded
payload. The legacy bare-itemnumber fallback is restricted to
CIRCULATION so a legacy HOLDS hold_id is not mistaken for an
itemnumber.
koha-tmpl/intranet-tmpl/prog/js/viewlog.js:
- Extend the info column override rendering (renderCircInfo) to HOLDS
entries so forced/confirmations arrays display translated, exactly as
they do for CIRCULATION. The log viewer moved to an ajax datatable in
bug 29016, so this replaces the earlier json_found change to
viewlog.pl/viewlog.tt.
misc/maintenance/migrate_action_logs_to_json.pl:
- Extended to handle legacy pre-Bug 40866 HOLDS CREATE logs where info
is a bare hold_id, recovering branchcode / biblionumber / itemnumber
from old_reserves / reserves.
- Uses the same keyset-pagination pattern as the circulation loop so
rows are not skipped under --commit.
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Previously, CIRCULATION ISSUE logs would store just the itemnumber when
there were no confirmations or forced overrides, but would store a full
JSON structure when overrides were present. This inconsistency makes
reporting difficult as queries need to handle both formats.
This patch changes AddIssue to always store logs in JSON format,
matching the pattern already used in AddRenewal. The JSON structure
now consistently includes:
- issue: issue_id
- branchcode: branchcode
- itemnumber: itemnumber
- confirmations: array of confirmation codes (empty array if none)
- forced: array of forced override codes (empty array if none)
Benefits:
- Consistent data format simplifies reporting and analysis
- Matches existing RENEWAL logging pattern
- Display code (tools/viewlog.pl) already handles both formats,
so backward compatibility is maintained
Test plan:
1. prove t/db_dependent/Circulation.t
2. Perform a normal checkout - verify action log contains JSON
3. Perform a checkout with overrides - verify confirmations/forced
are populated in the JSON
4. Check Tools > Log viewer - both types should display correctly
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Example: supplier_report on the basket page
Signed-off-by: Jason Robb <jrobb@sekls.org>
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
- Modify column-visibility.dt event handler to store the visibility
preference of the user
- Improve _dt_force_visibility() to correctly pick the correctly visibility value depending
on visibility_condition, is_hidden, and the user preference
- Add _dt_get_expected_column_visibility() to guess what's the expected visibility
- Force boolean true/false (vs 1/0) using !!
- Rely on a global variable kohaTableUserChoices to store the user
choice. It will be part of the state and store in localStorage.
Signed-off-by: Jason Robb <jrobb@sekls.org>
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Jason Robb <jrobb@sekls.org>
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Jason Robb <jrobb@sekls.org>
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
So we protect other calls
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The previous two commits on this bug fixed two ordering assumptions in
the RETRY loop of _allocateWithTransportCostMatrix(): resetting $inf at
the start of each attempt, and splicing unallocated task indices out of
@requests in descending order (removing a lower index first would shift
the position of any higher index still waiting to be removed).
This is a third instance of the same underlying problem, a few lines
earlier in the loop that compacts the working cost matrix $m before
that splice runs. @unallocated is built by iterating over agents (item
rows), pushing the column index Munkres assigned to each row that
turned out infeasible - and Munkres' row-to-column assignment is a
permutation, not a monotonic function of the row index, so
@unallocated is no more reliably sorted here than it was for the
splice.
That compaction loop scans matrix columns in ascending order and
advances a single pointer into @unallocated, expecting to meet each
entry in that same ascending order. When @unallocated arrives out of
order, the pointer never catches up: columns that should have been
dropped are left in place, and the columns after them are never
shifted left to fill the gap. From that point on, $m no longer lines
up with @requests, which *is* correctly reduced by the splice - so on
the next retry pass a hold's transport cost gets evaluated using a
different hold's matrix row. Depending on the exact misalignment, this
can silently drop a fillable hold from the queue while a genuinely
available item sits unused, or fill a hold via the wrong-cost item.
It didn't show up in the "many holds" regression test added earlier on
this bug, because that test's transport cost matrix is uniform (every
disabled pair costs the same); this defect only surfaces once the
columns competing for a scarce item have distinct costs.
Sort @unallocated once, ascending, and use that copy for the
compaction pass; reverse it for the splice, which needs descending
order.
Test plan:
1. Run the new t/db_dependent/HoldsQueue/RetryAllocationOrdering.t -
it calls _allocateWithTransportCostMatrix() directly with a small
hand-built scarce transport cost matrix and a pinned srand() (the
specific item-to-branch permutation needed to trigger the bug
depends on the shuffle() inside the routine). Reverting just this
patch makes 3 of its 6 subtests fail: the L2 pickup hold is dropped
even though an available item could have filled it.
2. Run t/db_dependent/HoldsQueue/TransportCostOptimizations.t to
confirm no regression in the existing retry-loop coverage.
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
InstallPlugins forces every discovered plugin class to be deleted and
re-required from disk on every call, so plugin_methods reflects the
.pm file as it currently exists rather than whatever was already
loaded into a long-lived Plack worker. That deletion used
Symbol::delete_package, which clears the target package's whole
symbol table -- including any already-loaded packages nested under it
(e.g. a REST API controller a plugin bundles alongside itself, such as
Koha/Plugin/Foo/Controller.pm). Those nested classes keep their own
%INC entry, so they are never re-required afterwards and are left
permanently blanked out for the rest of the worker's life.
This was caught when running
`prove -r t/db_dependent/Koha/Plugins/ t/db_dependent/Koha/REST/Plugin/`
on top of the original patch. Tests fail several files that have nothing
to do with the bug being fixed (Biblio_and_Items_plugin_hooks.t,
Holds_hooks.t, TemplateIncludePathHook.t, REST/Plugin/PluginRoutes.t),
because Koha::Plugin::Test::Controller (a bundled test-fixture
controller living in t/lib/plugins/Koha/Plugin/Test/Controller.pm)
gets its symbol table wiped as collateral damage every time
Koha::Plugin::Test is reloaded.
_forget_plugin_class() replaces the delete_package call with the same
technique used by the CPAN module Class::Unload (not added as a
dependency here, just the technique): only clear the target class's
own symbols and reset its @ISA, explicitly leaving any nested package
entries in its symbol table alone. Module::Pluggable discovers bundled
nested classes as their own entries in the plugin list regardless, so
they still get reloaded correctly on their own turn.
I also annotate the no strict 'refs' block added by the previous commit
with a '## no critic' justification, as flagged by koha-qa.pl.
Test plan:
1. Confirm the bug exists without this patch:
a) git checkout <previous commit> (drop this commit, keeping the
first 3 commits of the bug applied)
b) Inside KTD:
prove -r t/db_dependent/Koha/Plugins/ t/db_dependent/Koha/REST/Plugin/
c) Observe failing files unrelated to this bug's own tests:
Biblio_and_Items_plugin_hooks.t, Holds_hooks.t,
TemplateIncludePathHook.t and REST/Plugin/PluginRoutes.t.
2. Confirm the fix:
a) Re-apply this commit
b) Re-run the same command from 1(b) -- all tests now pass
3. Confirm the original bug this branch set out to fix is still fixed,
both via the manual UI steps from the first commit's test plan
(install Batch Permission Modifier Plugin 1.3.1, upload 1.4.1 over
it via Tools > Plugins, confirm the renamed method takes effect
without restarting Plack) and via:
prove t/db_dependent/Koha/Plugins/Plugins.t
(see the 'InstallPlugins() reloads stale symbol tables tests'
subtest)
4. Confirm QA tooling is clean:
koha-qa.pl -c 4
=> PASS for Koha/Plugins.pm and all touched test files, no critic
or tidiness failures.
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The FK-chain resolution added in the previous patch only handles ONE
level of nesting: it requires $parent_alias to be a direct relationship
of $main_source. This holds for every +count embed currently declared
in api/v1/swagger/ (all exactly one level deep, e.g.
biblio.items+count), but if that assumption is ever violated - a
deeper nested +count, a parent relation with a coderef (custom join)
condition, or no matching FK column name - the previous code silently
fell back to the pre-fix expression ($parent_alias.$fk_self), which can
reference a join alias that isn't available inside DBIC's limiting
subquery, reproducing the original 500 with no diagnostic.
This patch makes _build_count_subquery decline to build a subquery
(return undef) when the correlation can't be safely resolved, instead
of emitting SQL that may be broken. The caller already has an existing,
tested degradation path for "no DBIC relationship info" (prefetch the
association directly; to_api falls back to the Perl-level ->count call;
sorting on it is rejected with a clean 400) - unresolved nested
correlations now go through that same path rather than risking a 500.
Also:
- Extract the repeated "_resultset vs _result" result-source lookup
into a shared _result_source() helper (was duplicated in
dbic_merge_prefetch and _build_count_subquery).
- Document the one-level-deep limitation in _build_count_subquery's POD.
- Add a regression test pinning the 2-level-nested-+count case: it must
fall back to a plain prefetch (no COUNT subquery) rather than error
or emit unsafe SQL.
Test plan:
1. ktd --shell
2. prove t/Koha/REST/Plugin/Query.t t/db_dependent/Koha/Object.t \
t/db_dependent/Koha/REST/Plugin/Objects.t \
t/db_dependent/api/v1/acquisitions_orders.t \
t/db_dependent/api/v1/acquisitions_vendors.t \
t/db_dependent/api/v1/ill_requests.t \
t/db_dependent/api/v1/tickets.t \
t/db_dependent/api/v1/patrons.t
=> All pass, no regressions.
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
When nested +count embeds (e.g. biblio.items+count) are used alongside a
has_many prefetch (e.g. biblio.suggestions or the implicitly added
biblio.biblioitem) and pagination, DBIC's limiting subquery optimization
places +select scalar refs in the inner subquery where the parent join
alias (e.g. 'biblio') is not yet available.
This patch fixes _build_count_subquery to resolve the correlation back to
the main table (me) via the FK chain. Instead of generating:
(SELECT COUNT(*) FROM items
WHERE items.biblionumber = biblio.biblionumber)
It now generates:
(SELECT COUNT(*) FROM items
WHERE items.biblionumber = me.biblionumber)
This works because me.biblionumber IS the FK to biblio, holding the same
value. The main table columns are always available regardless of DBIC's
query structure.
Changes:
- Add $main_source parameter to _build_count_subquery for FK resolution
- Thread main_source through _parse_prefetch and dbic_merge_prefetch
- When parent_alias != 'me', resolve through parent relationship condition
- Update Query.t assertion to match the new correlation expression
Test plan:
1. Apply the regression tests from the previous commit
2. Run:
$ ktd --shell
k$ prove t/db_dependent/api/v1/acquisitions_orders.t
=> FAIL: The Bug 42995 subtest fails with 500 errors
3. Apply this patch, repeat from step 2
k$ prove t/Koha/REST/Plugin/Query.t \
t/db_dependent/Koha/Object.t \
t/db_dependent/Koha/REST/Plugin/Objects.t \
t/db_dependent/api/v1/acquisitions_orders.t \
t/db_dependent/api/v1/acquisitions_vendors.t \
t/db_dependent/api/v1/ill_requests.t \
t/db_dependent/api/v1/tickets.t \
t/db_dependent/api/v1/patrons.t
=> SUCCESS: Tests pass!
4. Sign off :-D
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
This patch adds regression tests for the 500 Internal Server Error
triggered when GET /api/v1/acquisitions/orders is called with nested
+count embeds (e.g. biblio.items+count) combined with a has_many
prefetch (e.g. biblio.suggestions.suggester) and pagination.
The root cause is that DBIC's limiting subquery optimization places
+select scalar refs in the inner subquery where the parent join alias
(biblio) is not yet available, resulting in:
DBD::mysql: Unknown column 'biblio.biblionumber' in 'WHERE'
The regression was introduced in:
commit e51334604f
Bug 41950: Make +count embeds sortable via SQL COUNT subqueries
Test plan:
1. Apply this patch
2. Run:
$ ktd --shell
k$ prove t/db_dependent/api/v1/acquisitions_orders.t
=> FAIL: The new subtest fails with 500 errors instead of 200
3. Sign off :-D
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
[FAIL] t/db_dependent/Koha/Plugins/C3.t
FAIL file_permissions
File must have the exec flag
FAIL test_no_warnings
'use Test::NoWarnings' must be present in this test file.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
If a patron is deleted and there's a checkout associated with them, an api call to {url}/api/v1/checkouts?checked_in=true
will result in an error. This patch updates the api to allow for a null value to be returned for patron_id
To test:
Before applying the patch:
1. In Administration->System preference enable the RESTBasicAuth system preference.
2. Checkout an item to a patron (I used item barcode: 39999000000252 and patron barcode 23529000121682)
3. Check the item back in
4. In your brower go to {sandbox/ktdurl}/api/v1/checkouts?checked_in=1
5. Observe there is a patron ID in the JSON response
6. Delete the patron
7. Repeat step 4
8. Observe an error because a "null" value was returned
9. Apply patch
Sandbox:
In sandbox management Actions->Apply patches
Once complete Actions->Build CSS and JS
Actions->Restart services
KTD:
ktd --shell
git bz apply 41889
yarn build
restart_all
10. Repeat step 4
11. Observe the checkout loads, but now has a null patron id.
12. Sign off
On KTD can also run the script: qa
to run tests.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
The tests were lacking a 'real' scenario proving the
sorting on a _counts works at the DB level
prove t/Koha/REST/Plugin/Query.t
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
Paul Derscheid
Martin Renvoize
Nick Clemens
Tomás Cohen Arazi
Pedro Amorim
Jonathan Druart
Andrii Nugged
Matt Blenkinsop
Olivia Reynolds
Fridolin Somers
Kyle M Hall
Aleisha Amohia
Shi Yao Wang
Lari Taskula
Hammat Wele
Laura_Escamilla
David Cook
Sanjar Tulkinov Anvar o'g'li
Baptiste Wojtkowski
Casey Conlin
Cameron E. Tidd
Lucas Gass
Emily Lamancusa
Lisette Scheer
