Prevent tests to fail if data in DB
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
These tests have always passed for me running cypress on UI, before and after this patch.
CLI tests however, were flaky and often times failed.
It appears this is because each module refresh was flooding the server with HTTP requests causing the whole thing to bottleneck (?)
With this patch, all widget requests are stubbed for every widget test, and tests now pass on CLI consistently. Passed 10/10 for me, whereas before it failed more often than passed.
Run this:
for i in {1..10}; do cypress run --spec t/cypress/integration/ERM/ModuleDashboard_spec.ts; done
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
prove t/db_dependent/api/v1/erm_counts.t
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Sponsored-by: UKHSA (UK Health Security Agency)
Signed-off-by: Jeremy Evans <Jeremy.Evans@ukhsa.gov.uk>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Test plan:
1) Enable ERMModule
2) Visit ERM:
<staff_url>/cgi-bin/koha/erm/erm.pl
3) Notice you now have a 'dashboard' homepage showing 4 different widgets
4) Click 'Open Widget Picker'. Remove and add different widgets. Close
5) Drag and drop the widgets around the dashboard, from left to right, top to bottom
6) Click the context menu (vertical dots) on each of the widgets. Click 'move to right', 'move to left', 'remove'.
7) On the 'Licenses needing action', notice there is a 'Settings' item option. Click that, test different settings values. Close settings.
8) Using the widget picker or context menu for the respective widgets, remove all widgets except one. Click away i.e. 'Agreements' on the left menu. Go back to home by clicking 'Home'. Notice your dashboard is as you left it.
9) Enable all widgets, rearrange them by dragging and dropping in different places. Repeat the previous exercise. Notice the dashboard keeps the widgets arranged as you left them.
10) Add some test data, on k-t-d run:
bash <(curl -s a0347d4199/bug_39320_test_data.sh)
11) Refresh the homepage dashboard, check the 'Counts' are correct and links work, check that the 'Latest SUSHI Counter jobs' table now populates and links to the data provider as well as respective job.
12) For 'Licenses needing action', test different settings values and confirm the table filters the results as expected.
13) As before, click away and comeback to ERM home. Notice your 'Licenses needing actions' settings are kept as you left them.
14) For the 'eUsage report' widget, notice it says 'No saved reports are availble to run'. Create a new usage report at:
<staff_url>/cgi-bin/koha/erm/eusage/reports
15) Click the 'Create report' tab, input the mandatory fields and enter a report name at the bottom of the page. Click 'Save report'.
16) Come back to ERM home. Notice the 'eUsage report' widget now shows a dropdown with a disabled 'Run' button.
Pick the report you just created. Notice the 'Run' button is now enabled. Click 'Run'. Notice you get the report results screen (will show nothing as there is no usage data in this test exercise).
17) k-t-d: Run cypress tests (reset_all beforehand to clear test data):
cypress run --spec t/cypress/integration/ERM/ModuleDashboard_spec.ts
Sponsored-by: UKHSA (UK Health Security Agency)
Signed-off-by: Jeremy Evans <Jeremy.Evans@ukhsa.gov.uk>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
The prettier plugin for TT has been adjusted to correctly parse our TT
files: most of them does not have an opening head but have a closing
head, then have an opening body but no closing body.
To avoid the parser to fail we are commenting those lines in a
pre-process step in the plugin, then uncomment them in the tidy script.
Which means we actually cannot parse correctly formatted HTML files.
https://gitlab.com/koha-community/prettier-plugin-template-toolkit/-/blob/main/src/parser.ts?ref_type=heads#L242
export const preprocess: Parser<Node>["preprocess"] = (text) => {
return text
.replace("</head>", "<!--</head>-->")
.replace(/<body(.*)/, "<!--<body$1-->")
.replace("</body>", "<!--</body>-->");
};
https://git.koha-community.org/Koha-community/Koha/src/branch/main/misc/devel/tidy.pl#L168
# Revert the substitutions done by the prettier plugin
my $content = read_file($file);
$content =~ s#<!--</head>-->#</head>#g;
$content =~ s#<!--<body(.*)-->#<body$1#g;
$content =~ s#<!--</body>-->#</body>#g;
(yes it's a bit messy, but our templates are messy)
Signed-off-by: Catherine Small <c.small@rcs.ac.uk>
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This patch adds comprehensive unit tests for the new valuebuilder plugin
functionality. The tests verify that plugins can provide valuebuilders
through the get_valuebuilder() method and that FrameworkPlugin can
correctly load and integrate with plugin-based valuebuilders.
The tests use a dedicated TestValuebuilder plugin to ensure realistic
testing scenarios without relying on mocks, following the pattern
established in other plugin hook tests.
Test plan:
1. Run the new test file:
prove t/db_dependent/Koha/Plugins/Valuebuilder_hooks.t
2. Verify all tests pass (9 tests total)
3. Run all plugin tests to ensure no regressions:
prove t/db_dependent/Koha/Plugins/
4. Verify all plugin tests still pass (113 tests across 19 files)
5. Test the specific functionality:
- Verify plugins without get_valuebuilder() method are not found as valuebuilders
- Verify plugins with get_valuebuilder() method are correctly discovered
- Verify FrameworkPlugin can load plugin-based valuebuilders
- Verify JavaScript generation works with plugin valuebuilders
- Verify plugin enable/disable affects valuebuilder availability
6. Confirm FrameworkPlugin integration tests still pass:
prove t/db_dependent/FrameworkPlugin.t
Signed-off-by: Catherine Small <c.small@rcs.ac.uk>
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This adds unit tests for adding support for the new LSL field:
- GIR segment extraction tests in t/Edifact.t for
sub_location_code field mapping
- Validation function tests in t/db_dependent/Koha/EDI.t
for location/collection code validation
- Copy function tests in t/db_dependent/Koha/EDI.t for
location/collection code assignment
- gir_segments generation tests in t/db_dependent/Koha/Edifact/Order.t
for preference handling
Tests cover:
- LSL field extraction from GIR segments alongside existing LSQ fields
- Independent LSQ/LSL preference mapping (location/ccode/empty combinations)
- Validation of location and collection codes against authorised values
- Error handling and message formatting
- Backwards compatibility with existing LSQ-only configurations
- Edge cases including missing data and invalid preference values
- LSL maps to location, LSQ maps to ccode
- LSL maps to ccode, LSQ maps to location (swapped configuration)
- Only LSL field present (LSQ empty)
- Only LSQ field present (LSL empty)
- Both LSL and LSQ map to location (edge case)
- Both fields disabled
Test plan:
1. Apply this patch
2. Run: prove t/Edifact.t
- Should pass with 8 new LSL-related tests included
3. Run: prove t/db_dependent/Koha/EDI.t
- Should pass with 16 new validation function tests
4. Run: prove t/db_dependent/Koha/Edifact/Order.t
- Should pass with 14 new gir_segments LSL/LSQ tests
5. Verify all existing tests still pass - no regression
Sponsored-by: Westminster City Council <https://www.westminster.gov.uk/>
Sponsored-by: Royal Borough of Kensington and Chelsea <https://www.rbkc.gov.uk/>
Sponsored-by: OpenFifth <https://openfifth.co.uk/>
Signed-off-by: Hannah Dunne-Howrie <hdunne-howrie@westminster.gov.uk>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This commit adds a unit test for the syspref override permission checking
functionality to ensure:
- Users without debug permission cannot override system preferences via URL parameters
- Users with debug permission can successfully override system preferences
- Superlibrarians maintain their ability to override system preferences
- The permission checking logic works correctly with haspermission()
The test creates three test patrons:
1. Regular patron with only catalogue permission (should be denied)
2. Patron with debug permission (should be allowed)
3. Superlibrarian (should be allowed)
Tests verify both the permission checking and the actual environment variable
setting that controls syspref overrides.
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This patch adds api unit tests for the extended_attributes functionality in the
ERM agreements API to ensure that:
1. Extended attributes are correctly returned when embedded
2. The internal record_table field is not exposed via the API
3. Both single GET and list endpoints work correctly with extended_attributes
Test plan:
1. prove t/db_dependent/api/v1/erm_agreements.t
2. Verify all tests pass
Signed-off-by: Jan Kissig <bibliothek@th-wildau.de>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This changes "Koha administration" to "Administration" on the
staff interface home page and the administration module home page.
Everywhere else in the staff interface, such as browser page titles
and breadcrumbs, administration is used.
This improves consistency, and is how it is commonly referred to.
It also change some other occurences in the code base, such
as code comments.
Test plan:
1. Note the inconsistencies in how we name and refer to the
administration module:
1.1 Staff interface home page: Koha administration
1.2 Administration module home page: Koha administration, but note
that the breadcrumb and browser page title is "Administration".
1.3 Breadcrumbs and browser page titles for any of the administration
areas are just "Administration".
1.4 More > Administration.
2. Apply the patch.
3. Note that on the staff interface home page and the administration
home page are now labelled "Administration".
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Roman Dolny <roman.dolny@jezuici.pl>
Signed-off-by: Paul Derscheid <paul.derscheid@lmscloud.de>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Amended-by: Jonathan Druart
tidy
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
To test:
1. prove t/db_dependent/Koha/Patron.t
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Adding a new guarantee and storing it fails with "Creating a patron"
selected in TrackLastPatronActivityTriggers when guarantors are set
to be mandatory. The problem comes from Koha::Patron->store called
in update_lastseen. Since we validate guarantor in Koha::Patron->store
but don't store them in there, new call to store fails because it
doesn't receive guarantor info as parameter.
As a solution, we should validate guarantors in separated method.
Validation can't be moved e.g. to Koha::Patron::Relationship->store
since this could lead to situations where guarantee is stored without
guarantor. This patch adds new method validate_guarantor and adjusts
old guarantor requirements tests to match this new method.
To test:
1. Select option "Creating a patron" in "TrackLastPatronActivityTriggers" syspref.
2. Enable syspref "ChildNeedsGuarantor" and make sure you have options in
"borrowerRelationship" syspref.
3. Create a guarantee patron, fill all required fields and add guarantor for patron
(either add Patron guarantor or fill Non-patron guarantor inputs).
4. Attempt to save patron.
=> Error box is displayed with text "The following fields are wrong. Please fix them.".
=> Logs show error "Patron creation failed! - [Child patron needs a guarantor]".
5. Apply this patch.
6. Create a guarantee patron again and save.
=> Patron should be saved correctly.
7. Create a new guarantee patron but this time don't add patron guarantor or fill
non-patron guarantor inputs.
8. Attempt to save patron.
=> Save should fail with error "A child patron needs a guarantor."
9. Add patron guarantor who can also be a guarantee and save.
=> Save should fail with error "A guarantor cannot be a guarantee."
Also prove t/db_dependent/Koha/Patron.t
Sponsored-by: Koha-Suomi Oy
Signed-off-by: Magnus Enger <magnus@libriotech.no>
Works as advertised.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Use should use coded_values for the library and item type column so that
we can search using the code and the description.
With this patch we now have a dropdown list with the libraries and item
types in the table's headers.
Test plan:
0. Turn on AllowHoldItemTypeSelection
1. Create a hold and select an item type
Note: I think there is a bug if you select it from "Hold next available
item", use "Hold details". Please confirm?
2. Go to the holds history view
3. Filter and search by item type description and code. Same for
library.
Use the dropdown list in the headers.
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This QA follow-up adds comprehensive test coverage for edge cases that
the original patch handles:
1. Empty string values ('') - the default value in sysprefs
- Verifies that empty strings are treated as "not set"
- Confirms charge calculations are skipped (performance optimization)
- Ensures patrons are not blocked
2. Zero values (0) - valid numeric limit
- Verifies that zero triggers charge calculation (valid number)
- Confirms patrons are not blocked (0 means "no limit")
- Documents intended behavior that 0 disables blocking
3. Positive numeric values - normal case
- Confirms charges are calculated and blocking works correctly
These tests validate that looks_like_number() is the correct approach
rather than simple truthiness checking, as it properly distinguishes
between:
- Empty string '' (not a number, skip calculation)
- Zero 0 (is a number, calculate but don't block)
Test plan:
1. Apply this patch
2. Run the tests:
prove t/db_dependent/Koha/Patron.t
3. Verify all tests pass
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This test ensures that when extended attribute validation fails during
patron creation via the API, no welcome email is sent. This validates
the fix where extended_attributes() validation was moved before the
welcome email is enqueued.
Test plan:
1. Apply this patch
2. Run: prove t/db_dependent/api/v1/patrons.t
3. Verify all tests pass, including the new test that confirms no welcome
email is sent when extended attribute validation fails
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
For SIP based payments, no branchcode is recorded for the corresponding
credit. Koha should be recording the institution id (i.e. branchcode)
for SIP payments.
Test plan:
1) Apply this patch
2) prove t/db_dependent/SIP/Transaction.t
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
To test:
Enable pseudonymization, may need to add bcrypt settings (see bug 28911)
Perform some circulations
View the jobs (Admin->Manage jobs)
Filter for Pseudonymize statistic
Note the progress is 0/1 even when job finished
Apply patch, restart all
Do more circulations
Confirm jobs now say 1/1 complete
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This adds to the unit tests in t/db_dependent/Koha/Patrons/Import.t. It confirms that repeating the same patron import (with extended patron attributes) will not generate duplicated borrowersLog entries.
Test plan:
1. prove t/db_dependent/Koha/Patrons/Import.t
Sponsored-by: Auckland University of Technology
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Just adding the fifth parameter here, since that actually should
be the next published date. But note that it is just a placeholder,
since the test does not use the associated date substitutions.
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Before this patch, we pass the previous issue date to generate the next seq - this works for the pattern as we are incrementing.
When the patter is translated ot words, we need to do this from the date of the next issue, so we must pass in the next
published date
This patch changes the call to GetNextSeq from serial-collection.pl to ensure the new issue publication date is used to
generate the numbers.
To test:
0 - Follow the plan to recreate from the initial bug description and recreate
1 - Apply patch
2 - On the serial used in previous patch go to 'Serial collection'
3 - Generate next issue
4 - Confirm the numbering is correctly generated after this patch
Signed-off-by: Wesley Owen <wesleyowen@churchofjesuschrist.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
The test was looking for .current class but the Vue component
uses aria-current="page" attribute instead.
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
.dt-select-filter does no longer exist
Here we rely on DT's API to retrieve the correct visible column index.
However we cannot for Selenium tests, not ideal but it fixes the
failure.
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Test plan:
1) Choose an existing patron and note their first and second names as well as their date of birth
2) In the interface, create a new patron and enter all of those details to be the same as your patron from step 1
3) Submit the form, you will get a warning saying that there is a duplicate
4) In the OPAC, self register a new patron using the same details again
5) The form should submit with no error or warning
6) Apply patch, yarn api:bundle and restart_all
7) Repeat steps 2 - 5
8) Step 3 should still have the same behaviour
9) Step 5 should now show a warning at the top to say that there is a duplicate, please contact a member of staff
10) Find the 'PatronDuplicateMatchingAddFields' syspref and select an extra field there, ensuring that your patron from step 1 has data for that field
11) In either the interface or the OPAC, register a new patron but this time change the data of the new field you added in step 10
12) No warning should show for a duplicate patron
13) prove t/db_dependent/api/v1/patrons.t
14) prove t/db_dependent/Koha/Patrons.t
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Pamela Bird <pbird@cityofchesapeake.net>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Add test on empty marc object.
Remove MARC21 module and create with parent MARC object.
Test plan:
Run t/db_dependent/Koha/Biblio/Metadata/Extractor/MARC/MARC21.t
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Test plan:
Run t/db_dependent/Koha/Biblio/Metadata/Extractor/MARC.t
Verify that addbiblio.pl still opens as expected when modifying an
existing record or adding a new record.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This enhancement ensures the due date stored in items.onloan is changed when a recall causes a checkout due date to change.
To test:
1. Enable UseRecalls syspref and set corresponding circulation rules:
- Recalls allowed
- Recalls per record
- On shelf recalls allowed - set to "If any unavailable"
- Recall due date interval - set to 1
- Recall pickup period
2. Check out an item to a patron (Patron A)
2a. Make sure this patron has a primary email address set
2b. Check the patron's messaging preferences - make sure this patron has the 'Email' checkbox checked for the 'Return recalled item' notice
3. Go to Tools -> Notices & slips and find the RETURN_RECALLED_ITEM notice. In the Email notice, change "[% checkout.date_due | $KohaDates %]" to "[% item.onloan | $KohaDates %]" and Save.
4. Log into the OPAC as another patron
5. Search for the item checked out in step 2, and place a recall on that item
6. Search for that item in the staff interface
=> Note that the item status says, "Checked out to <patron>: due <tomorrow>"
7. Create the following SQL report in the reports module, using the barcode of the item that was recalled:
SELECT onloan FROM items WHERE barcode='xxxxxxxxx'
8. Run the report
=> The onloan value is still set to the item's original due date, not the updated due date of tomorrow
9. Go to Patron A's account again and view their Notices tab. Click on the 'Notification to return a recalled item'
=> The original due date is used in the notice.
10. Apply the patch and restart services
11. Reset to run the tests again:
- Check in the item, but do not confirm the recall in the pop-up
- Go to the biblio detail page and cancel the recall
12. Repeat steps 2 to 5.
13. Run your report again (step 8)
=> The onloan value should now show the updated due date of tomorrow
14. View Patron A's latest 'Notification to return a recalled item' notice (step 9)
=> The updated due date should be used in the notice.
15. Confirm tests pass
- t/db_dependent/Koha/Recalls.t
Sponsored-by: Auckland University of Technology
Signed-off-by: Brendan Gallagher <brendan@bywatersolutions.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
The filtering condition was not correct, the presence of a line in
the transfer table is not enough to now if the item is in transit.
Here we filter reusing the same conditions as
Koha::Item::Transfer->in_transit
Test plan:
Have at least 2 items for a given bibliographic record
Transfer item 1
Transfer item 2
Receive/Fullfill the transfer for item 1
Item 1 is now available and item 2 is in transit
Go to items record and filter holdings table with status "In transit"
=> Only item 2 is displayed
Filter holdings table with status "Available"
=> Only item 1 is displayed
Signed-off-by: Anneli Österman <anneli.osterman@koha-suomi.fi>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Attempt to call undefined import method with arguments (new ...) via package C4::Calendar (Perhaps you forgot to load the package?) at t/db_dependent/Hold.t line 25.
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Attempt to call undefined import method with arguments (split_callnumber) via package C4::ClassSplitRoutine::LCC (Perhaps you forgot to load the package?) at t/Labels.t line 22.
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
perl -wc misc/devel/create_test_data.pl
Attempt to call undefined import method with arguments ("schema") via package "Koha::Database" (Perhaps you forgot to load the package?) at /kohadevbox/koha/t/lib/TestBuilder.pm line 5.
Attempt to call undefined import method with arguments ("_type") via package "Koha::Biblios" (Perhaps you forgot to load the package?) at /kohadevbox/koha/t/lib/TestBuilder.pm line 7.
Attempt to call undefined import method with arguments ("_type") via package "Koha::Items" (Perhaps you forgot to load the package?) at /kohadevbox/koha/t/lib/TestBuilder.pm line 8.
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Tomás Cohen Arazi <tomascohen@theke.io>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Test plan:
Run the test without applying the follow-up patch.
The test should fail. This shows that we need the fix and will help
avoid future regressions.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Addressing QA lines of feedback has changed the forms. Cypress tests need to be updated accordingly.
cypress run --spec t/cypress/integration/SIP2/Accounts.ts --config video=false,screenshotOnRunFailure=false
cypress run --spec t/cypress/integration/SIP2/Institutions.ts --config video=false,screenshotOnRunFailure=false
cypress run --spec t/cypress/integration/SIP2/SystemPreferenceOverrides.ts --config video=false,screenshotOnRunFailure=false
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
This follow-up removes the security vulnerability of storing plaintext
passwords in the sip_accounts table. The SIP2 system now relies entirely
on Koha's secure password authentication.
Changes:
- Remove plaintext password comparison from login_core() and telnet_transport()
- Remove login_password column from sip_accounts table
- Update database schema files and atomic update
- Remove password field from API definitions and Vue form
- Update all tests to remove login_password references
Authentication now works in two stages:
1. Access control: Check if userid exists in sip_accounts (allowlist)
2. Authentication: Validate password against hash in borrowers table
This maintains SIP access control while eliminating plaintext password
storage. Staff users authenticate with their Koha password (securely hashed).
Test plan:
1. Run updated tests to verify they pass
2. Verify SIP2 authentication works with staff user credentials
3. Confirm no plaintext passwords are stored
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Signed-off-by: Andrew Fuerste Henry <andrew@bywatersolutions.com>
Signed-off-by: Lisette Scheer <lisette@bywatersolutions.com>
Signed-off-by: Martin Renvoize <martin.renvoize@openfifth.co.uk>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>
Jonathan Druart
Pedro Amorim
Pedro Amorim
Lucas Gass
Martin Renvoize
David Nind
Tomás Cohen Arazi
Lari Taskula
Emmi Takkinen
Kyle M Hall
Nick Clemens
Alex Buckley
Marcel de Rooy
Owen Leonard
Matt Blenkinsop
Aleisha Amohia