AddBiblio does not return a title; the biblioitemnumber is stored in the
title variable.
The variables for biblioitemnumber are not used and can be removed.
Test plan:
Run t/db_dependent/Reserves.t
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
These tests do not cover correctly getletter, but it is not the goal of
this patch. Superlibrarian behaviour must be tested as well.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
TEST PLAN
---------
prove t/db_dependent/Letters.t
-- there will be a message: "C4::Context->userenv not defined!"
apply this patch
prove t/db_dependent/Letters.t
-- there will no longer be that message.
run qa test tools
Signed-off-by: Lee Jamison <ldjamison@marywood.edu>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Bug 18471 added number formating in order receive.
I found a bug generated by this.
Koha::Number::Price defines methods to (un)format price depending on CurrencyFormat syspref.
The bug is that for CurrencyFormat US, some params are not defined in this module so there are retrieved from system locale. If system locale is not en_US, the params will be wrong. For example fr_FR will define ',' as decimal and so saved unit price is divided by 1000.
To recreate:
- On a server with locale 'fr_FR'
- With Plack mode
- Set CurrencyFormat on 'US'
- Open a basket
- Place an order for an item
- Close basket
- Receive order with changing 'Actual Cost' to '1,234'
- Note the 'Actual Cost' is now '1.23'
- Run unit test t/Number/Price.t without and with patch
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
It is important to highlight that tests have been skipped instead of having them pass
on the absence of the required fr_FR.UTF-8 locale installed on the system.
This patch does that check and effectively skips them.
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
This patch adds unit tests to t/Number/Price.t in order to valid the problem.
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
This patch makes changes to the tag moderation template and JavaScript
to fix handling of tags with double or single quotes. This patch also
moves the tags moderation JavaScript out of the template and into a
separate JS file.
To test you should have multiple tags awaiting moderation, including
tags which contain double and single quotes.
- Go to Tools -> Tags.
- In the list of tags pending approval, test approving and rejecting
tags, including those containing single or double quotes.
- The state of the "Approve" or "Reject" buttons should correctly
change according to the action you chose.
- The label in the status column should update correctly.
- In the "Check lists" form, submitting approved, rejected, and
unclassified terms should result in the correct message.
Signed-off-by: Lee Jamison <ldjamison@marywood.edu>
Works correctly based on test plan. Tested using single- and
double-quoted tags. Passes QA Tools.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Just as we show this distinction in OPAC, this patch adds a type column
in the Your lists tab that displays Private or Shared. It always contains
Public in the other tab.
Test plan:
[1] Check if you see Shared for a private lists with shares in staff.
[2] Run t/db_dependent/Utils/Datatables_Virtualshelves.t
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Lee Jamison <ldjamison@marywood.edu>
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
This function only serves to check if the shelf name is not empty. It
does not even work since it incorrectly refers to f.addshelf (while it
should test #shelfname).
We can solve this and improve consistency by doing the same as in the
staff template with html5 required.
Test plan:
[1] Without this patch it was possible to add a list without a name in
OPAC (with js error: TypeError: f.addshelf is undefined).
[2] With this patch, verify that you cannot do this anymore.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Lee Jamison <ldjamison@marywood.edu>
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
If we we move from Private to Public or vice versa, this should reflect
on the Anyone remark from the first patch.
If we go back to a private list without shares, we should show it.
Handled in a js sub AdjustRemark, triggered by onchange of the category
combo or the permission combo.
Test plan:
[1] Edit a private list without shares in OPAC. Toggle category and/or
permissions. Is the remark shown or hidden accordingly?
Note: Include a test with OpacAllowPublicListCreation enabled.
[2] Edit a shared list or public list in staff. Toggle category/permissions.
You should never see the remark.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
This is a follow-up of report 18228 that simplifies the permissions for
lists. It reapplies the idea behind bug 10865 by adding a remark next to
the Anyone permission if it is not actually effective. A next patch will
make it behave dynamically.
Note: The Anyone permission is not relevant if it is a private list that
has no shares. Note that looking at the prefs OpacAllowPublicListCreation
and/or OpacAllowSharingPrivateLists is not decisive. You should look at
list permissions and shares in the database; turning off the Sharing pref
does not automatically remove all shares in the system. It only blocks
creating new shares.
We only need changes in opac-shelves.tt and virtualshelves/shelves.tt.
Test plan:
[1] Verify that you see the remark now in OPAC and staff.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
To test:
* Apply this patch
* Define some rules in the interface using the left hand buttons
* Define some rules in the interface using the right hand buttons (scroll to see)
* Delete some rules in the interface using the left hand buttons
* Delete some rules in the interface using the right hand buttons (scroll to see)
* Choose a rule to edit using the left hand buttons
* Edit the rule, saving using the left hand button
* Edit a rule, but clear the values using the left hand buttons
* Choose a rule to edit using the right hand buttons (scroll to see)
* Edit the rule, saving using the right hand buttons (scroll to see)
* Edit a rule, but clear the values using the right hand buttons (scroll to see)
Signed-off-by: Guillaume <guillaume_lair@ville-limoges.fr>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
This patch adds the required entry in Makefile.PL so the clubs/ dir is correctly installed
during the make install step.
To test:
- Run:
$ cd kohaclone
$ perl Makefile.PL
- Follow any install method and options
- Run
$ make
$ ls blib/INTRANET_CGI_DIR/ | grep clubs
=> FAIL: No clubs dir
- Run:
$ make clean
- Apply this patch
- Run:
$ perl Makefile.PL
- Follow any install method and options
- Run
$ make
$ ls blib/INTRANET_CGI_DIR/ | grep clubs
=> SUCCESS: The clubs/ dir is correctly picked for inclussion in the intranet cgi dir
- Sign off :-D
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Inventory tool performance seems to be worst since 16.11.
I think it is because of authorized values computing changed by Bug 17249.
For each subfield of each item, we try to get the authorized value description with Koha::AuthorisedValues->search_by_marc_field.
But this method does not use cache like Koha::AuthorisedValues->get_description_by_koha_field.
I propose to use Koha::AuthorisedValues->get_description_by_koha_field and also to look for authorized value description only for item fields used in TT : location, notforloan, itemlost, damaged, withdrawn.
I have experimented inventory time on 100 items from 5s to 1s.
Test plan :
- Without patch
- Perform inventory with barcode file
- Check results and mesure execution time
- Apply patch
- Reperform inventory with same barcode file
- Check results is the same
- Compare execution time
- Run prove t/db_dependent/Items/GetItemsForInventory.t
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
If you manually visit the following links when you only have
permission to run reports, you'll still be able to access the ability
to create and edit reports:
/cgi-bin/koha/reports/guided_reports.pl?phase=Create%20report%20from%20SQL
/cgi-bin/koha/reports/guided_reports.pl?phase=Edit%20SQL
This patch ties these 2 unaccounted for phases to the create_reports
permission.
With patch, issue no longer can be reproduced.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
There are complex conditions to know if a librarian can use a fund, we
cannot do that with a SQL query, we need to use C4::Budgets::CanUserUseBudget
When this patch is applied then all funds are visible to both
superlibrarian patrons and patrons with all acq permissions
Signed-off-by: Alex Buckley <alexbuckley@catalyst.net.nz>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Bug 18314 causes t/db/SIP/Message.t to fail (quite often) since
TestBuilder fills login_attempts with a random number. (Note: Only
when FailedLoginAttempts is non-zero.)
Trivial fix: TestBuilder should have a zero default for login_attempts.
Test plan:
Do not yet apply this patch.
Set pref FailedLoginAttempts to say 3.
Run t/db_dependent/SIP/Message.t. Might fail on the password test (CQ).
Apply this patch.
Run t/db_dependent/SIP/Message.t again. Does not fail anymore.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Test plan:
prove -v t/db_dependent/Koha/Patrons.t
Subtest: renew_account
1..30
ok 1 - 2016-02-29T00:00:00 + 12 months must be 2017-02-28T00:00:00
ok 2 - 2016-02-29T00:00:00 + 12 months must be 2017-02-28T00:00:00
ok 3 - With BorrowerLogs, Koha::Patron->renew_account should have logged
ok 4 - today + 12 months must be 2017-03-31T00:00:00
ok 5 - today + 12 months must be 2017-03-31T00:00:00
ok 6 - Without BorrowerLogs, Koha::Patron->renew_account should not have logged
ok 7 - today + 12 months must be 2017-03-31T00:00:00
ok 8 - today + 12 months must be 2017-03-31T00:00:00
ok 9 - 2016-04-30T00:00:00 + 12 months must be 2017-04-30T00:00:00
ok 10 - 2016-04-30T00:00:00 + 12 months must be 2017-04-30T00:00:00
ok 11 - 2016-10-30T00:00:00 + 12 months must be 2017-10-30T00:00:00
ok 12 - 2016-10-30T00:00:00 + 12 months must be 2017-10-30T00:00:00
ok 13 - With BorrowerLogs, Koha::Patron->renew_account should have logged
ok 14 - today + 12 months must be 2017-11-30T00:00:00
ok 15 - today + 12 months must be 2017-11-30T00:00:00
ok 16 - Without BorrowerLogs, Koha::Patron->renew_account should not have logged
ok 17 - today + 12 months must be 2017-11-30T00:00:00
ok 18 - today + 12 months must be 2017-11-30T00:00:00
ok 19 - 2016-12-30T00:00:00 + 12 months must be 2017-12-30T00:00:00
ok 20 - 2016-12-30T00:00:00 + 12 months must be 2017-12-30T00:00:00
ok 21 - 2017-06-30T00:00:00 + 12 months must be 2018-06-30T00:00:00
ok 22 - 2017-06-30T00:00:00 + 12 months must be 2018-06-30T00:00:00
ok 23 - With BorrowerLogs, Koha::Patron->renew_account should have logged
ok 24 - today + 12 months must be 2018-07-31T00:00:00
ok 25 - today + 12 months must be 2018-07-31T00:00:00
ok 26 - Without BorrowerLogs, Koha::Patron->renew_account should not have logged
ok 27 - today + 12 months must be 2018-07-31T00:00:00
ok 28 - today + 12 months must be 2018-07-31T00:00:00
ok 29 - 2017-08-31T00:00:00 + 12 months must be 2018-08-31T00:00:00
ok 30 - 2017-08-31T00:00:00 + 12 months must be 2018-08-31T00:00:00
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
From DateTime::Duration pod:
""
For positive durations, the "end_of_month" parameter defaults to wrap.
For negative durations, the default is "limit". This should match how
most people "intuitively" expect datetime math to work.
""""
We need end_of_month => limit for positive durations as well.
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Add problematic cases to highlight the problem.
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
From jenkins output:
Subtest: CanBookBeIssued + Koha::Patron->is_debarred<7c>has_overdues
1..8
not ok 1
[SKIP]
I executed it several times and display the different $error, $alerts and question keys.
GNA and RESTRICTED were sometimes set, which block the issue.
Reading the code it seems that some patron's attributes must be removed to avoid the checkin rejection.
Test plan:
Execute the tests several times and notice that it fails randomly
With this patch they should always pass.
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
If finesMode is not set to production, only 1 fine will be created (the renewal
one will not). This is what assumes the tests.
If set to 'production', the tests will fail because the fines will not
be deleted (because of the DBIx::Class) warning.
Now we mock the value before charging.
prove t/db_dependent/Circulation.t
t/db_dependent/Circulation.t .. 16/95 DBIx::Class::Storage::DBI::select_single(): Query returned more than one row. SQL that returns multiple rows is DEPRECATED for ->find and ->single at t/db_dependent/Circulation.t line 491
t/db_dependent/Circulation.t .. 56/95
# Failed test 'Can auto renew, OPACFineNoRenewals=10, patron has 10'
# at t/db_dependent/Circulation.t line 670.
# got: 'auto_too_much_oweing'
# expected: 'auto_renew'
# Looks like you failed 1 test of 6.
Test plan:
prove t/db_dependent/Circulation.t
should return green whatever the value of finesMode
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
That breaks randomly tests on Jenkins (Koha/Patrons.t)
It certainly affects data in other languages but I did not find a quick
and easy way to modify this file (replace all guarantorid values with
NULL)
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Easy fix
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
As the DB structure is not initialized for sessions, running this tests raise an irrelevant warning.
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Mirko Tietgen <mirko@abunchofthings.net>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Mirko Tietgen <mirko@abunchofthings.net>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Mirko Tietgen <mirko@abunchofthings.net>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
$patronid is not necessarily set or does not match a valid cardnumber.
These cases must be taken into account to avoid the script to crash and
raise the following error:
Can't call method "unblessed" on an undefined value at
(...)/koha/opac/sco/sco-main.pl line 117
Test plan:
Hit sco/sco-main.pl and confirm that the error is gone with this patch
applied
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
From
commit 0492fcf87f
Bug 17855: Onboarding tool
+++ b/koha-tmpl/intranet-tmpl/prog/en/modules/intranet-main.tt
@@ -1,5 +1,3 @@
-[% USE Koha %]
-[% SET footerjs = 1 %]
+[% USE Koha %] has been re-added by
commit 151cdcd133
Bug 18673: News author does not display on staff client home page
But not the footerjs = 1
Test plan:
From the main page of the staff interface, delete a news item
Without this patch applied you will not get a JS confirmation message
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
Trivial fix.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
We also need just one rollback at the end here.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
There is an action_logs entry via logaction() without transaction to be rolled
back in t/db_dependent/Log.t. This leaves an entry in action_logs after
the test is over.
To replicate:
1. prove t/db_dependent/Log.t
2. Observe a new entry in action_logs table
To test:
1. Apply patch
2. prove t/db_dependent/Log.t
3. Observe there are no new entries in action_logs
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Test plan:
1. On an up to date branch go to Patrons and try to browse for a patron
by last name
2. Notice that the page processes indefinately
3. Apply patch
4. Repeat step 1
5. Notice the patron results are displayed
Signed-off-by: David Cook <dcook@prosentient.com.au>
Works as described.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Close the p tag and add a br tag to visually divide the two sections
Data problems and Warnings regarding system configuration.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Discovered the problem when testing a duplicate key in deleteditems.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
This is a trivial fix, but finding the cause for a failing checkin in
SIPServer surely was not.
Without this simple change, I could not checkin an item while the SIP server
just closed the connection and syslog contained nothing useful.
The error can be easily reproduced without this patch by:
use Modern::Perl;
use C4::SIP::ILS::Item;
my $item = C4::SIP::ILS::Item->new('some_barcode');
where some_barcode should refer to an item checked out.
You should trigger the error (with your due date):
2017-07-28 23:59:00 is not an object at line 3.
Why the ILS/Item.pm module still compiles without a warning, is not clear
though.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Colin Campbell <colin.campbell@ptfs-europe.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Zebra doesn't care about cases, but ES does and default mapping is
'control-number'
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
See MARC specification.
Note that we have a preceding text before all 776s too.
Test plan:
[1] Set field 776 ind2 to 8. And add text in $i.
[2] Look for this text on OPAC detail and staff detail.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Felix Hemme <felix.hemme@thulb.uni-jena.de>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Adds code to opac and staff detail XSLT in order to show 776$t (or $a) and
a link based on 776$w (record control number). $w should be in the form:
(MARCOrgCode)Record-number
For instance: (NL-AmRIJ)12345
Note: You should have enabled default opac/intranet XSLT view.
Test plan:
[1] Add two 776s (one with $t and $w and one with $t only)
[2] Verify that you see "Additional physical formats" on opac-detail. Does
the link for $w work?
[3] Check intranet detail too.
[4] Empty 776$t and put text in $a. Check detail page again.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Felix Hemme <felix.hemme@thulb.uni-jena.de>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Same problem in t/db_dependent/Search/History.t.
To replicate:
1. Check the row count of borrowers, branches, categories, sessions, sms_providers
tables
2. prove t/db_dependent/Search/History.t
3. Repeat step 1
4. Observe borrowers the following tables have increased in row count:
- borrowers
- branches
- categories
- sessions
- sms_providers
To test:
1. Before applying the patch, go through steps at "To replicate" plan
2. Apply patch
3. Go through steps at "To replicate" plan
4. Observe step 4 no longer applies and those tables have the same number of
rows as before executing the test.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Test t/db_dependent/Auth.t seems to have an ineffective test data cleanup.
Data generated by TestBuilder is left in borrowers, branches, categories,
sms_providers and sessions tables after the test.
To replicate:
1. Check the row count of borrowers, branches and categories tables
2. prove t/db_dependent/Auth.t
3. Repeat step 1
4. Observe borrowers the following tables have increased in row count:
- borrowers
- branches
- categories
- sessions
- sms_providers
To test:
1. Before applying the patch, go through steps at "To replicate" plan
2. Apply patch
3. Go through steps at "To replicate" plan
4. Observe step 4 no longer applies and those tables have the same number of
rows as before executing the test.
This issue has been happening in REST tests as well, and this solution is
directly copy-pasted from t/db_dependent/api/v1/cities.t
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
To test:
1. Before applying patch, check the amount of branches in database
2. prove t/db_dependent/SIP/Message.t
3. See that a new branch is stored
4. Apply patch
5. prove t/db_dependent/SIP/Message.t
6. See that a new branch is no longer stored
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
