Implemented invalid_username_or_password template
variable being handed to Auth.tt. Removed indentation
changes and modification to return value of checkpw
Test plan:
See comment #2
Alternative test plan:
- Clear session cookies or close / open your browser
- Go to [MY SERVER]/cgi-bin/koha/installer/install.pl
- Verify that message appears if wrong username/pw provided
- Verify that you can log in with databas administrator account
Tested with alternative test plan, works as expected. Title and
description in commit message adapted.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 87eb80136d)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Bug 18028 removed the install_misc directory but
install_misc/UpgradeBackup.pm was still used by the 'upgrade' rule of
make.
Other files from install_misc were useless to it may be better not to
reintroduce this directory with only 1 file.
Test plan:
`make`
`sudo make install`
`make upgrade`
Signed-off-by: Josef Moravec <josef.moravec@gmail.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Mason James <mtj@kohaaloha.com>
In aqbudgetperiods.pl a commented line is removed that contains a
wrong userenv hash key. Should be branch instead of branchcode.
In aqbudgets GetBudgetHierarchy is called with the same wrong userenv
hash key. Should be userenv->{branch}.
This made another bug visible: if you call GetBudgetHierarchy with a
branch and without owner, the where clause should take into account
that the branchcode can be empty (not null).
Test plan:
[1] Run Budgets.t
[2] Run aqbudgets.pl from Administration
[3] Add the show_mine=1 parameter in the URL
[4] Change owner of one of the funds and repeat step 2 and 3.
t/Budgets.t returns green. Followed test plan, works as expected.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Mason James <mtj@kohaaloha.com>
ref($sub) eq 'ARRAY' should always be true, what we want is to call
can_edit_subscription if there is at least 1 serial for this
subscription.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Mason James <mtj@kohaaloha.com>
Currently we have subscriptions with 300+ serials received.
It takes a lot of time to show serials-collection.pl (~20s).
This is especially troublesome when receiving serials, since after receival we get redirected to that page.
We no longer can receive daily serials in the allotted timeframe.
This quick and dirty fix prevents checking the subscription editing for each serial, but instead checks it for the first serial only.
This reduced page load time by ~18s
TEST PLAN:
1. Receive ~300 serials (or just a bunch :) )
2. Observe the gradual slowing of the receival action.
AFTER THIS PATCH:
1. Receive ~300 serials more (or just a bunch)
2. Observe a significant performance improvement.
Signed-off-by: Paul POULAIN <paul.poulain@biblibre.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Mason James <mtj@kohaaloha.com>
The error codes come from CanBookBeIssued.
The warnings speak for themselves. Note that the GNA message is similar
to the one used in opac-user.tt (gonenoaddress is translated to the patron
as contact information not up-to-date).
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Mason James <mtj@kohaaloha.com>
Empty tags should be removed, and message id
Note that not all error codes are covered here.
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Mason James <mtj@kohaaloha.com>
Only RENEW_ISSUE should not be considered as blocker, others should.
Note that this code is not robust and the whole script as well as the
return of CanBookBeIssued should be rewrittent completely.
We need a small patch to easy backport to stable releases
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Mason James <mtj@kohaaloha.com>
The "confirmed" flag is used to know if that user confirmed a situation
that needs a confirmation.
But if the issue/renew is impossible the CanBookBeIssued and the
'impossible flags' should be checked.
Otherwise a patron can checkout and renew bypassing the circulation
rules (Understand 'no limit' here...)
Test plan:
Want to renew?
Checkin $barcode, then
/cgi-bin/koha/sco/sco-main.pl?patronid=$cardnumber&barcode=$barcode&confirmed=1&op=checkout
Want to bypass the checkin?
Same url...
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 85bd15a83f)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Conflicts:
opac/sco/sco-main.pl
The AddRenewal subroutine currently uses the circulation rules for the branch
stored in the Issues table (which is the holding branch) when calculating the new due date.
This patch replaces using the branch from the Issues table with the branch
specified by the HomeOrHoldingBranch syspref.
To test:
1. Set up 2 branches, Branch1 and Branch2
2. Set up a loan rule in Branch1 for DVDs with a 21 day loan period and a
21 day renewal period.
3. Set up a loan rule in Branch2 for DVDs with a 14 day loan period and a
14 day renewal period.
4. Checkout a DVD belonging to Branch1 while logged into Branch2. It will
receive the correct 21 day loan period.
5. Renewing the same DVD while logged into either Branch1 or Branch2
will give a 14 day due date, rather than 21 days.
6. Checkout a DVD belonging to Branch2 while logged into Branch1. It will
receive the correct 14 day loan period.
7. Renewing the same DVD while logged into either Branch1 or Branch2
will give a 21 day due date, rather than 14 days.
8. Apply the patch and repeat steps 4-7. The correct due date should be given
when the item is renewed, regardless of where it is checked out or renewed.
This update removes reassignment of $branch variable.
Signed-off-by: Cédric Vita <cedric.vita@dracenie.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 7d35bdf58a)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Hie, Tools > Batch record deletion seems broken.
Any deletion returns error :
Bibliographic record YYY was not deleted. An error occurred. (The error was: {UNKNOWN}: DBD::mysql::db begin_work failed: Already in a transaction at /usr/share/perl5/DBIx/Class/Storage/DBI.pm line 1560. at /home/koha/src/C4/Biblio.pm line 3468 , see the Koha log file for more information).
Looks like it is because of Bug 18242 which added a transaction in C4::Biblio::_koha_delete_biblio_metadata : $schema->txn_do.
The script batch_delete_records created a transaction with $dbh->{AutoCommit} = 0;
This patch fixes by using also Koha::Schema in batch_delete_records to manage transaction.
It also removes "$dbh->{RaiseError} = 1", this behavior is managed in Koha::Database.
Test plan :
- Go to Staff interface : Tools > Batch record deletion
- Enter a few existing biblionumbers
- Click on "Continue"
- Click on "Delete selected records"
=> Without patch you get a DB error
=> With patch you get confirmation message
- Try to get the biblios to confirm they are deleted : /cgi-bin/koha/catalogue/detail.pl?biblionumber=xxx
- Test with and without Plack
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 0f32bd9999)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
If a fine is created for a lost item but the itemnumber is not supplied,
the system will return it.
The item should not be mark as returned if there is no item linked to
the fine.
Test plan:
1. Turn StoreLastBorrower on
2. Create a manual invoice for a lost item, do not supply a barcode
3. Pay the fines 'Pay fines > Pay'
=> Without this patch applied you get
Can't call method "last_returned_by" on an undefined value at
/home/marc/koha/C4/Circulation.pm line 2188.
=> With this patch applied, you must not get the error.
To test:
1) Go to columns_settings.pl --> Circulation --> issues-table and note that you cannot change the settings for checkin.
2) Go to columns_settings.pl --> Patrons --> issues-table and note that you cannot change the settings for checkin.
3) Apply the patch.
4) Go to both tables again and note that you can now change the settings.
Followed test plan, works as expected.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 9098a4d3fa)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
==TEST PLAN==
1) Make a subscription with a record
2) Search and view the record in OPAC
3) Click on the subscription tab and click more details
4) The breadcrumb will say 'Detals for' without the record name
5) Apply patch
6) Refresh the page
7) The breadcrumb will display the name of the record
(fixed small error)
Signed-off-by: Sonia BOUIS <sonia.bouis@univ-lyon3.fr>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 39bcba540a)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
On Home > Patrons > Accounts for... and Home > Patrons > Pay fines for...,
make the amounts display as defined in syspref 'CurrencyFormat'
To test:
- Apply patch
- Give a patron a high fine or credit (e.g. 12345.67)
- Got to pages Home > Patrons > Accounts for... and
Home > Patrons > Pay fines
- Verify that amounts display following syspref 'CurrencyFormat'
(e.g. 12'345.67 for syspref set to 360'000.00 (CH)
Signed-off-by: Mirko Tietgen <mirko@abunchofthings.net>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit db467287d1)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Conflicts:
koha-tmpl/intranet-tmpl/prog/en/modules/members/pay.tt
We need to standardize on terminology, this changes all instances of
patron types to be patron categories. To text, just view the patch
Signed-off-by: Cédric Vita <cedric.vita@dracenie.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 9a4fd7b270)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
This patch restores the branchname in OPAC subscription detail view.
Additionally, it adds the subscriptons count to the tab title to make
it consistent with the other tab titles.
To test:
- Search a subscription in OPAC catalog
- Go to Subscriptions tab (see screenshot in initial comment)
- Verify that library name does not display
- Apply patch
- Verify that library name displays properly in Subscriptons tab
- Verify that the tab title displays the subscriptions count
e.g.: Subscriptions (2)
Signed-off-by: Claire Gravely <claire_gravely@hotmail.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit c6463c258f)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Conflicts:
koha-tmpl/opac-tmpl/bootstrap/en/modules/opac-detail.tt
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit f14de0007c)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
The headers of the tables displayed in patroncards/manage.tt are hardcoded in the .pl and thus cannot be translated.
This simple fix reuse the translate_card_element tool already in the code to make the strings translatable.
Applied patch and verified that table headers on manage.pl display the same as before.
Signed-off-by: Marc Véron <veron@veron.ch>
Re-tested, works as expected.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit bb1492e9f6)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Good catch from Jonathan. See comment11.
Authorities detail should pass a CSRF token to authorities-home when
deleting a record without linked biblios.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Passing the token with GET is not a good way to do, but nothing quick to
replace that.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit d2ee53fb5f)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Without this patch, it is possible to delete authority records with URL
manipulation.
Like: /cgi-bin/koha/authorities/authorities-home.pl?op=delete&authid=[XXX]
Test plan:
[1] Go to Authorities. Search for some authorities (without links).
[2] Delete an authority. Should work.
[3] Apply patch.
[4] Construct an URL like above to delete another authority. Should fail.
Under Plack this results in an internal server error, the log tells
you: Wrong CSRF token.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Amended the test plan.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 5a7dc0749f)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Instead of passing the homebranch to the slip, let's pass
TransferWaitingAt (the value we display)
To test:
1 - Check in an item that will initiate a transfer (hold or automatic
item return)
2 - Note the transfer message and receipt match
3 - Remove the reason for transfer (bot don't cancel transfer)
i.e. cancel hold or change homebranch of item to current library
4 - Check the item in again
5 - Note the message displays the transfer destination and the slip
shows the homebranch
6 - Apply patch
7 - Repeat 5 but note slip and message match
8 - Test any edge cases for correctness
9 - sign off
Signed-off-by: Josef Moravec <josef.moravec@gmail.com>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit f454013ec9)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Resolves complaints from qa tools.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 53f2b56fa1)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Conflicts:
C4/Serials.pm
- Remove useless comments
- Use Koha::Objects::find instead of Koha::Objects::search
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit c26afc2568)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Test plan:
* run without the patch, the test will fail
* run with the patch, the test will pass
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit d79a264b69)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Without this patch, there is an inconsistency in subscriptions:
* a new subscription has no history end date set (as expected)
* if you renew it, a date is set.
This patch removes the setting of the history end date during renewal
Test plan:
* create a subscription, go to subscription-detail.pl page, tab "Summary". No History end date is set
* renew the subscription => a date has been set
* apply the patch
* create another subscription
* renew it => no history end date has been set
* sign-off
Signed-off-by: beroud <isabelle.beroud@univ-lyon3.fr>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 1843ee148b)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Fields with underscore like B_address do not print on patron cards.
To reproduce:
- Create patron card layout using fields with underscore in their name
(e.g. <surname><B_address> )
- Print (export) patron card
- Verify that fields without underscore are replaced by their value,
but fields with underscore do not replace but show the field name
To test:
- Apply patch
- Try to reproduce and verify that fields with underscore are replace
as expected
Signed-off-by: Christopher Brannon <cbrannon@cdalibrary.org>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit a723de2ae0)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
The combobox on the left side of late orders is filled by sub
GetBooksellersWithLateOrders. The same change as in the first patch
must be made here to include suppliers with late orders without a
price.
Bonus: Sort the list.
Test plan:
[1] Run t/db_dependent/Bookseller.t.
[2] Go to late orders. Use the filter on suppliers.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Srdjan <srdjan@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit b76eac9ca2)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
This patch removes the filter of unitpricesupplier <> 0 and
unitpricelib <> 0 from GetLateOrders as used in lateorders.pl.
This allows you to claim late orders without a price entered.
Test plan:
[1] Run t/db_dependent/Acquisition.t
[2] Check if you see an order with no price in lateorders.pl. (If needed,
remove prices from a few orders.)
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Srdjan <srdjan@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 71f267fd3f)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
The zero status should be added in @notforloans. It is a legitimate status
for available books. The zero status description can be translated in
the template.
The 'Change item status' is confusing. We mean that we saw a notforloan
status that we do not expect (read: does not exist in the table).
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 664d653417)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
In inventory tool, there is a test on scanned items notforloan value : if the value is not one of authorized values, the inventory results displays the problem "Change item status".
The bug is that if notforloan filter is not used, the notforloan values array is not filled and so all scanned items are marked with this problem.
Test plan :
- Check you have in default framework an item subfield mapped with items.notforloan and with an authorized value catergory, for example 'NOTFORLOAN'
- Check the authorized value category has at least value '0'
- Create a new value in this category, for example '9'
- Create a biblio with default framework
- Create an item with barcode='000AAA1', callnumber='ZZZAAA1' and notforloan=0
- Create an item with barcode='000AAA2', callnumber='ZZZAAA2' and notforloan=9
- Delete new value created above, for example '9'
- Create a file 'barecodes.txt' containing 2 lines '000AAA1' and '000AAA2'
- Go to inventory tool : /cgi-bin/koha/tools/inventory.pl
- Choose file in 'Barcode file'
- Enter item callnumber between 'ZZZ' and 'ZZZZ'
- Check 'Compare barcodes list to results'
- Submit
=> without patch, you get 2 results with problem 'Change item status' even for item with notforloan=0
=> with patch, you get 1 result with 'Change item status' on item with notforloan=9
Signed-off-by: Nick Clemens <nick@quecheelibrary.org>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 358fb744c7)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
This patch updates SIP checkin to use reserve_id as param, to fix regression
introduced by Bug 14695.
To test:
Given 1 patron and 1 biblio with two items attached:
1) create two record level holds on one biblio on same user
Using SIP client or telnet:
2) checkin item x on pickup branch, observe item is waiting and hold 1 is filled
3) checkin item y on pickup branch, observe item y now is waiting and x is available
4) apply this patch and repeat 1-3, now expected behaviour should be
that both holds are filled and both items waiting
Signed-off-by: Srdjan <srdjan@catalyst.net.nz>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit 72de37e24b)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
To test:
1 - Add reserves.reservenotes to HOLDPLACED message
2 - Enable emailLibrarianWhenHoldIsPlaced OpacHoldNotes sysprefs
3 - Place a hold via OPAC with a note
4 - view the messagequeue and note the reservenotes is blank
5 - Apply patch
6 - Place a hold with a note
7 - view the messagequeue and note the reservenotes is populated
Followed test plan, works as expected.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit 3b794b1312)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
If CircAutoPrintQuickSlip is set to clear screen, the $borrowernumber variable
is reset but $borrower.
In 16.11.x and before that causes the app to crash, because
$borrower->{flags} is set to a hashref by GetMemberDetails, if
$borrowernumber is set.
This case is better handled in master (17.05) but side-effects could
happen: if $borrowernumber is reset, $borrower should be reset too.
The way to drive this behaviour is terrible and lot of things should be
done to clean this area. This patch is a quick and easy fix to make it
backportable easily.
Test plan:
On master, no change expected
On 16.11 and before:
0. Set CircAutoPrintQuickSlip=clear screen
1. Create a new patron
2. Set permission for this patron to 'reservesforothers' (or something
else).
3. Go on the circulation tab
4. Do not fill the barcode input and submit
=> Without this patch you will get an error
Can't use string ("64") as a HASH ref while "strict refs"....
=> With this patch apply you will get a blank screen (expected
behaviour).
I would recommend to test this patch with the other value of
CircAutoPrintQuickSlip as well
borrower_message_preferences cannot be truncated because of the foreign.
DBMS fails with
"Cannot truncate a table referenced in a foreign key constraint"
To avoid that we should remove the FK check and truncate the other table
as well.
I am wondering if we really need a truncate here
DELETE FROM borrower_message_preferences;
should do the job, but leave it as it because of the param name.
Test plan
perl misc/maintenance/borrowers-force-messaging-defaults --doit --truncate
Should no longer raise the error message
Signed-off-by: Mark Tompsett <mtompset@hotmail.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(cherry picked from commit dd88c8f710)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
The parameter change in Koha::Token should be applied to the calling
scripts.
Test plan:
Confirm that the different forms of the scripts modified by this patch
still work correctly.
Test the problematic behavior:
Open 2 tabs with in same user's session, go on the edit patron page
(memberentry.pl).
Log out and log in from the other tab.
Submit the form
=> Wrong CSRF token should be raised
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>
Adds a internal routine to handle default values for the parameters
id and secret.
Also adds a parameter session_id for generate_csrf and check_csrf. This
session parameter is combined with the id parameter when generating or
checking a token.
Test plan:
Run t/Token.t
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>
Currently the CSRF token generated is based on the borrowernumber, and
is valid across user's session.
We need to restrict the CSRF token to the current session.
With this patch the CSRF token is generated concatenating the id
(borrowernumber) and the CGISESSID cookie.
Test plan:
Run t/Token.t
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>
Signed-off-by: Mason James <mtj@kohaaloha.com>
To make sure we will not never get a race conditions for these kinds of
notices, we need to add a LOCK on the message_queue table.
This does not smell the best way to do that, but I faced deadlock issues
when I tried to use "UPDATE FOR"
https://dev.mysql.com/doc/refman/5.7/en/innodb-locking-reads.htmlhttps://dev.mysql.com/doc/refman/5.7/en/lock-tables.htmlhttps://dev.mysql.com/doc/refman/5.7/en/commit.html
To test this patch, or another solution, you need to apply manually this
change:
my $message = C4::Message->find_last_message($borrower, $type, $mtt);
unless ( $message ) {
+ sleep(1);
C4::Message->enqueue($letter, $borrower, $mtt);
} else {
And repeat the test plan from first patch.
Do not forget to truncate the message_queue table.
Followed test plans, works as expected.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit be156d9ad9)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
There is an obvious race condition when CHECKIN and RENEWAL are
generated from circulation.pl calling svc/renew or svc/checkin in AJAX.
The 2 first queries will try to get the id of the last message
(find_last_message) and if it does not exist, they will insert it.
Theorically that could be lead to have several "digest" messages for a
given patron.
I did not recreate more than 2 messages, from the third one at least one
of the two firsts existed in the DB already.
This patch just simplifies the code to make the SELECT and INSERT or
UPDATE closer and limit the race condition possibilities.
Test plan:
0. Set RenewalSendNotice and circ rules to have a lot of renewals available
1. Use batch checkouts (or one by one) to check out several items to a
patron
2. Empty message_queue (at least of this patron)
3. Renew them all at once ("select all" link, "renew or check in"
button)
4. Check the message_queue
Without this patch you have lot of chances to faced a race condition and
get at least 2 messages for the same patron. This is not expected, we
expect 1 digest with all the messages.
With this patch apply you have lot of chances not to face it, but it's
not 100% safe as we do not use a mechanism to lock the table at the DBMS
level.
Tested both patches together, works as expected.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit 607b14516a)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Sets the vendors currency in the edi generated order
The currency used is agreed between vendor and library
and the value in the quote is optional
Edifact potentially allows the currency to be overwitten
by specifying another currency in the CUX segment but
we know currently of no supplier doing this
Signed-off-by: Alex Buckley <alexbuckley@catalyst.net.nz>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
A similar bad template check from C4::Auth::get_template_and_user
should be applied in C4::Templates::gettemplate.
Before this patch it would be possible to expose files like:
my $template = C4::Templates::gettemplate(
'/etc/passwd', 'intranet', CGI::new, 1
);
print $template->output;
Note that the is_plugin flag in the above call is the culprit. This patch
provides a quick security fix without touching get_template_and_user, and
can be backported to stable branches.
I will provide an enhanced and centralized check on report 17989, also
removing the is_plugin flag.
Note: We allow .pref here too for use in admin/preferences.pl.
Test plan:
[1] Run t/db_dependent/Auth.t (triggering get_template_and_user and
gettemplate).
[2] Run t/db_dependent/Templates.t again (see first test plan).
The tests should no longer fail.
[3] Open a page on opac or intranet.
[4] Open a systempreferences tab.
[5] Add a book to the cart and send it ([opac-]sendbasket uses gettemplate).
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit fbbd4b40f0)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
A trivial test, similar to the ones in Auth.t.
Without the check in gettemplate (added in the second patch), the passwd
file will be exposed and the test fails.
Test plan:
Run t/db_dependent/Templates.t without second patch. The two tests in the
last subtest should fail.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit 5508897d5a)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Conflicts:
t/db_dependent/Templates.t
This directory is no longer maintained and contain outdated information
on how to install Koha.
On the Internet there are tutos using them and that can create confusion
to new user (yes it could be confusing as well to execute a file that
does no longer exist).
NOTE: last tweak was mid 2016, but before that 2015. So I agree with
the deletion.
Signed-off-by: Mark Tompsett <mtompset@hotmail.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit a6f50cbcb9)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Bug 17731 removed the -x option of rebuild_zebra but koha-rebuild-zebra
still cals the script with this option.
"Warning: You passed -x which is already the default and is now deprecated"
Test plan:
sudo koha-rebuild-zebra -f
should no longer raise the warning
Signed-off-by: Mason James <mtj@kohaaloha.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit fe898bc641)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Test plan:
- Add a new patron attrbute and mark it searchable
- Populate a new patron with 'potato' in that field
- Add/edit another patron to have email potato@invalidemail.com'
- Perform a patron search with query 'potato' (in standard fields)
=> Both patrons are returned
- Perform a patron search with filters 'Email' and query 'potato'
=> Only 1 patron is returned and you are redirected to the patron detail page.
Followed test plan, works as expected.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit 5823d4ca6a)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
Signed-off-by: Brendan A Gallagher <brendan@bywatersolutions.com>
(cherry picked from commit f86500351a)
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Conflicts:
t/db_dependent/Utils/Datatables_Members.t
Alex Buckley
Jonathan Druart
Marcel de Rooy
Olli-Antti Kivilahti
David Kuhn
Fridolin Somers
Christopher Brannon
Luke Honiss
Marc Véron
Grace McKenzie
Blou
Nick Clemens
Julian Maurice
Paul Poulain
Benjamin Rokseth
Mason James
Colin Campbell