David Cook [Tue, 27 Feb 2024 06:05:24 +0000 (06:05 +0000)]
Bug 35955: Cache CSRF token in template plugin
This change uses the Koha::Cache::Memory::Lite cache to
cache the CSRF token, so that it is only generated once,
and is re-used by the Koha::Template::Plugin::Koha object
throughout the entire template processing for the HTTP request.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org> Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
David Cook [Thu, 15 Feb 2024 02:49:18 +0000 (02:49 +0000)]
Bug 36098: Add Koha::Session module to ease session handling
This patch adds a Koha::Session module that makes it easier
to work with Koha sessions without needing the full C4::Auth module.
Test plan:
0. Apply the patch
1. Run the following unit tests:
prove ./t/db_dependent/Auth.t
prove ./t/db_dependent/Auth_with_cas.t
prove ./t/db_dependent/Koha/Session.t
2. Observe that they all pass
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com> Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Jonathan Druart [Tue, 30 Jan 2024 08:02:19 +0000 (09:02 +0100)]
Bug 35935: Ensure login branch will be used after incorrect login
If a different branch is selected after an incorrect login, the previous
branch will be used.
To recreate:
* login with foo/bar, select CPL => FAIL
* login with koha/koha, select another branch => OK but CPL is picked!
It was caused by a dup of "branch" in CGI param list (and first was
picked).
This patch patch also removes "koha_login_context" to not have it twice.
You can also open the source of the page to confirm that form#loginform
contains "branch" and "koha_login_context" in hidden inputs.
Signed-off-by: Magnus Enger <magnus@libriotech.no>
Tested in KTD. Works as advertised. Signed-off-by: Julian Maurice <julian.maurice@biblibre.com> Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Jonathan Druart [Fri, 26 Jan 2024 07:58:17 +0000 (08:58 +0100)]
Bug 35918: Fix auto library connect (AutoLocation)
This code is a bit weird, its purpose it to auto select the library depending on the IP.
A problem appears if the same IP is used, then the user's choice will
might be overwritten randomly by another library.
To recreate the problem:
Turn on AutoLocation
Use koha/koha @CPL for test
And the following config:
*************************** 1. row ***************************
branchcode: CPL
branchname: Centerville
branchip: 172.18.0.1
*************************** 2. row ***************************
branchcode: FFL
branchname: Fairfield
branchip: 172.18.0.1
*************************** 3. row ***************************
branchcode: FPL
branchname: Fairview
branchip: 172.18.0.4
Connect and select CPL. Randomly FFL will be picked instead.
Signed-off-by: Magnus Enger <magnus@libriotech.no>
Tested this on top of 35890 and 35904 because git bz said they were required dependencies.
Figured out the IP Koha was seeing me as coming from in /var/log/koha/kohadev/plack.log.
Added that IP to the branchip for Centerville, Fairfield and Fairview. Set AutoLocation = Yes.
After this I could recreate the problem: If i left the "Library" field in the login screen
at "My Library" I got logged into a random library selected from the three i had set
branchip for. Applying the patches fixed this, as expected.
Tests pass, with AutoLocation off.
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Jonathan Druart [Thu, 25 Jan 2024 09:35:41 +0000 (10:35 +0100)]
Bug 35904: Make C4::Auth::checkauth testable easily
This patch suggests to add a new flag do_not_print to
C4::Auth::checkauth to not print the headers and allow to test this
subroutine more easily.
We do no longer need to mock safe_exit and redirect STDOUT to test its
return values.
There are still 3 left:
1.
733 # checkauth will redirect and safe_exit if not authenticated and not authorized
=> Better to keep this one, not trivial to replace
2.
806 # This will fail on permissions
This should be replaced but testing $template->{VARS}->{nopermission}
fails, I dont' think the comment is better.
3.
828 # Patron does not have the borrowers permission
Same as 2.
2. and 3. should be investigated a bit more.
This patch also move duplicated code to set patron's password to a
subroutine set_weak_password.
Test plan:
Read the code and confirm that everything makes sense.
QA: Do you have a better way for this? Yes it's dirty!
Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Martin Renvoize [Wed, 16 Jun 2021 10:04:42 +0000 (11:04 +0100)]
Bug 34426: Add xt/find-missing-csrf.t
Signed-off-by: David Cook <dcook@prosentient.com.au> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl> Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Andreas Jonsson [Mon, 12 Feb 2024 11:16:02 +0000 (11:16 +0000)]
Bug 35930: Add guards for plugins_enabled
The 'new' method in Koha::Plugins returns undefined if
plugins are disabled. Therefore, calls to this method
must be guarded by a check that plugins actually are enabled.
Test plan:
* Code inspection of patch, alternatively
* Activate the ill system by installing a backend such as
koha-illbackend-libris:
https://github.com/Libriotech/koha-illbackend-libris
* Make sure plugins are disabled in koha-conf.xml
* In the staff interface, go to ILL requests.
* The page should load without getting an error 500.
PA amended commit message: This is not related to ILL backends being plugins or not
This is about ILL batches, where checking for metadata enrichment plugins was missing 'enable_plugins' guard
Additionally, unrelated to batches, it's also about ILLAvailability, where checking for ILL availabililty plugins was missing enable_plugins guard
Signed-off-by: Pedro Amorim <pedro.amorim@ptfs-europe.com> Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io> Signed-off-by: Hans Pålsson <hans.palsson@hkr.se> Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Nick Clemens [Fri, 26 Jan 2024 14:10:01 +0000 (14:10 +0000)]
Bug 35518: Check authentication and set userenv before fetching userenv variables
Currently we get the userenv before we have set it correctly for the session
To test:
1 - Sign in as a user with fast cataloging permission
2 - Bring up a patron, type gibberish into barcode field to get a fast cataloging link
3 - Check the link, it should have your current signed in barcode
4 - Sign in to a different browser with a different user and at a different branch
5 - Bring up a aptron in circulation and type gibberish into barcode field to get a fast cataloging link
6 - It may have your branch, but it may also have the other user's branch from the other window
7 - Keep entering gibberish to get a link until one user has the correct branch
8 - Then switch to the other browser, and keep entering gibberish, watch the branchcode change
9 - Apply patch, restart all
10 - Test switching between browsers. generating fast cataloging links
11 - Users should now consistently have the correct branch
Signed-off-by: David Nind <david@davidnind.com> Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Andreas Jonsson [Thu, 8 Feb 2024 10:57:03 +0000 (11:57 +0100)]
Bug 36034: (bug 34893 follow-up) fix capture of return values from checkpw
Adapt code to the change of return value type of checkpw
introduced in bug 34893
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org> Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Martin Renvoize [Thu, 11 Jan 2024 11:14:27 +0000 (11:14 +0000)]
Bug 29002: Remove whitespace only lines
Jonathan highlighted some trailing whitespace.. I only see a few cases
where a line only contains whitespace and I didn't see these caught by
the QA script at the time of submission.
This patch introduces some tests on the current (and new) behavior for
the `checkpw` function.
I needed it to better understand if an edge case was actually possible
(it wasn't).
Found a really minor annoyance for the internal check with expired
password not returning the $patron object for consistency with the other
use cases.
I think this method deserves (at least) changing the return value to a
sane data structure. But that's not target for backporting to stable
releases. So a separate bug.
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Kyle M Hall [Fri, 1 Dec 2023 11:29:19 +0000 (06:29 -0500)]
Bug 34893: (QA follow-up) Tidy code for qa script
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net> Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net> Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Kyle M Hall [Fri, 22 Sep 2023 18:20:59 +0000 (14:20 -0400)]
Bug 34893: ILS-DI can return the wrong patron for AuthenticatePatron
Imagine we have a set of users. Some of those users have a NULL userid. We then call AuthenticatePatron from ILS-DI for a patron with a NULL userid, but a valid cardnumber. We call checkpw, which returns the cardnumber and userid. We then call Koha::Patrons->find on the userid *which is null*, meaning the borrowernumber returned is not the correct one, but instead the earliest patron inserted into the database that has a NULL userid.
Test Plan:
1) Give three patrons a userid and a password
2) From the database cli, set all patrons's userid to null
Run this query: update borrowers set userid = null;
3) Call AuthenticatePatron with username being the 1st patron cardnumber,
and password being the password you set for that patron
http://localhost:8080/cgi-bin/koha/ilsdi.pl?service=AuthenticatePatron&username=kohacard&password=koha
4) Note you get back a borrowernumber for a different patron. Refresh the page and the number is correct.
5) Do the same with the 2nd patron. Same issue at 1st and correct number after.
6) Apply this patch
7) Restart all the things!
8) Do the same with the 3rd patron.
9) Note you get the correct borrowernumber! :D
10) prove t/Auth.t t/db_dependent/Auth_with_ldap.t t/Auth_with_shibboleth.t t/db_dependent/Auth_with_cas.t
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net> Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Kyle M Hall [Wed, 29 Nov 2023 17:18:32 +0000 (17:18 +0000)]
Bug 34893: Add unit tests
Signed-off-by: Victor Grousset/tuxayo <victor@tuxayo.net> Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
projects / koha.git / log