This script is called by a cron job, but was never marked as executable,
so will error every time it's called, which is no good. This also adds
an appropriate #! line to the top of the file.
To test:
* apply the patch
* make a package
* ensure that the installed
/usr/share/koha/bin/cronjobs/automatic_renewals.pl
is mode 755 and does something when run.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Publication year was useless for MARC21, as it was using the
biblioitems.publicationyear (UNIMARC) instead of biblio.copyrightdate
(MARC21).
This patch tries to fix the search options to work for both MARC
flavours.
To test:
In Reports > Statistics wizards > Catalog:
- Test the publication year search options on a UNIMARC and
a MARC21 installation.
- Make sure search results make sense for both.
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
The changes are fine with me, but this script is certainly not bug free.
Also note that all its SQL should be moved out according to our
coding guidelines. When doing so, a rewrite would be appropriate.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
TEST PLAN:
1. Load 167 staged files to Koha.
2. Try to make an acquisiton from a staged file.
3. Wait 16s for the acqui/addorderiso2709.pl -view to load.
AFTER PATCH:
1. Load 167 staged files to Koha.
2. Try to make an acquisiton from a staged file.
3. Wait 1.6s for the acqui/addorderiso2709.pl -view to load.
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Prior to this patch, the koha packages depended on mysql-client. This
meant that it was tricky to use things such as the MariaDB libraries
instead.
This patch recommends mysql, but will accept mariadb and other things
that are marked by debian as able to replace mysql.
Signed-off-by: wajasu <matted@34813.mypacks.net>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Minor code tidy to clean up qa script warning.
http://bugs.koha-community.org/show_bug.cgi?id=9165
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
A small enhancement to clear existing synced passowrd should this
config option be enbled. This followup is related to bug 12831
http://bugs.koha-community.org/show_bug.cgi?id=9165
Signed-off-by: Robin Sheat <robin@catalyst.net.nz>
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
This adds a configuration option to LDAP that prevents it from storing
user's passwords in the local database. This is useful when users of
hosted Koha wish to prevent any form of offsite password storage for
security reasons.
Notes:
* if the option is not included in the koha-conf.xml file, then the
current default behaviour of saving the password locally is retained.
* this has no impact on passwords that are already in the database.
They will not be erased.
To use:
* edit the koha-conf.xml for a system that uses LDAP for
authentication.
* in the <ldapserver> configuration, add:
<update_password>0</update_password>
* feel a greater sense of security.
To test:
1) have a Koha system that authenticates using LDAP.
2) note that when a user logs in, their password is saved (hashed) in
the database.
2.5) it is important to note that, for whatever reason, a user's
password is not stored on a login where their account is created,
only when they log in after being created. Thus perhaps log in and
log out a couple of times to be sure.
3) add the <update_password>0</update_password> option to the
<ldapserver> section of koha-conf.xml.
4) login with a new user (or erase the password from the database for
an existing user) and note that the password field is not populated.
5) log out and log back in just to be sure, check the password field
again.
Sponsored-By: National Institute of Water and Atmospheric Research (NIWA)
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Local only logins should continue to function when LDAP is enabled.
This was not the case after bug 8148 [LDAP Auth should FAIL when ldap
contains a NEW password]. For this case, we need to diferentiate
between local accounts and ldap accounts. This is somewhat challenging
and thus this patch is only part of the story.
The other half can be achieved with bug 9165
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
checkpw_ldap should return 0 if it is not an anonymous bind, and authentication
fails. This is better explained on the bug comments. This is just a regression
test for the revised functionality.
To test:
- Run
$ prove t/db_dependent/Auth_with_ldap.t
=> FAIL: it fails because C4::Auth_with_ldap doesn't match the expected behaviour
- Apply the bugfix from Martin
- Run
$ prove t/db_dependent/Auth_with_ldap.t
=> SUCCESS: tests now pass.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
This patch could have only been
- name => $values[0]->{name},
+ name => $letter->{name},
Other changes are just indentation and variable names (send an hashref
$letter to the template and use the Branches TT plugin to display the
branch name)
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
C4::Letters::getletter does not set mtt in value ( i.e. { email => "my
email notice} ) at the contrary of the get_letter routine defined in
tools/letters.pl.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
This patch fixes a major issue introduced by the
commit 5c4fdcf Bug 11742: A letter code should be unique.
The interface should let the possibility to create a default template
letter and some specific ones, with the same letter code (letter.code).
The patches submitted on bug 11742 tried to fix an issue based on a
(very bad) assumption: letter.code should be considered as a primary key and
should be uniq.
This patch reintroduces this behavior.
Note that the interface will block a letter code used in different
module (this is consistent not to have the same letter code used for different
needs).
This patch is absolutely not perfect, it just tries to change as less
change as possible and to use new tested subroutines.
Test plan:
1/ Verify that the problem raised on bug 11742 does not appears anymore.
2/ Verify there are no regression on adding, editing, copying, deleting
letters.
3/ Verify you are allowed to create a default letter template with a letter
code and to reuse for a specific letter (i.e. for a given library).
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
To test:
Package up a branch with this patch
install that package
create a site - sudo koha-create --create-db testdisable
enable a site - sudo koha-enable testdisable
check it's enabled - sudo koha-list --enabled
* it should show up
disable a site - sudo koha-disable testdisable
Do this for both debian squeeze/wheezy and ubuntu 12.04 and 14.04, if you can. I'd like to see a sign off from a debian (sq/wh)eez(e/y) or ubuntu 12 user, because I could only test reliably on ubuntu 14.04.
* make sure apache restarts and no errors are produced
check it's disabled - sudo koha-list --enabled
* it should not show up
check the site is still there - sudo koha-list
* it should still be there
check that the config file has the Include for disabling uncommented
* the line Include /etc/koha/apache-shared-disable.conf should not have a # in front.
Re-enable the site - sudo koha-enable testdisable
* the line Include /etc/koha/apache-shared-disable.conf should have a # in front.
And the final question - does the site work? All other functions unchanged?
Signed-off-by: Mason James <mtj@kohaaloha.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Works as expected. code reads better too.
Edit: I added a missing space in one line.
The work from KnowledgeWare Technologies need to be
added to about page.
To test:
1) Apply the patch
2) Go to About page > Translations
3) Credits for Arabic must show updated information
(agreed between involved parties)
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
They all agreed on the attribution.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
No problems found.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
- Adds translations for new permissions
- Adds ISO code to the sample currencies
- Removes a comment line about branchrelations from
sample_libraries (en)
To test:
- Run the German web installer with all sample data
selected
- Verify it runs without problems
- If you speak German - verify the translations make sense
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
No errors from the web installer. I verified that all the changes I saw
in the diff appeared correctly in the permissions. Currencies all
appeared to be correct after installation. Translations appeared to be
very German-looking.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
These 2 bugs are in conflict.
The first one always join the issue table, the second one join on this
table too if the OnSiteCheckouts pref is enable.
So DBI raises an error if the pref is enabled (2 joins on the same
table).
This patch removes the conditional join.
Test plan:
Go on a detail record page with items and verify that items are list and
that the error no more appears in the log file.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Reproduced the problem, the patch fixes it, no noticeable regression found.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Works as described, items are visible again.
Passes tests and QA script.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
This patch adds the missing multiple-select.png image.
Test plan:
- search the pref 'advsearch'
- without this patch, you get a 404 error for the image
- with this patch, you don't get the error but a wonderful image on the
right of the "multiple" pref
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Without any parameter, dt_from_string should not raise a warning
message.
Test plan:
Verify that the test file t/DateUtils.t displays a warning:
Use of uninitialized value $date_string in pattern match (m//) at
Koha/DateUtils.pm line 58
if the change in dt_from_string is not applied (manually edit the file).
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Email addresses in git seem to be case sensitive.
The list of shortlog -ns was good, but git shortlog -e still made a
difference.
Additionally, reworded initial comments slightly.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
If a library is using Talking Tech for phone notices, any waiting hold
phone notice will show up twice!
This is because Koha generates on at the time the hold is set to
waiting, and then the cronjob TalkingTech_itiva_outbound.pl generates
it's own notice as well.
The former notice will always have a status of 'pending', as the
TalkingTech_itiva_inbound.pl script will update the notice the outbound
script created.
The solution is to prevent Koha from creating a phone notice for waiting
holds if TT is enabled, and let the cron script do it.
Test Plan:
1) Enable Talking Tech from the system preferences
2) Set a hold waiting phone notice in the notices and slips editor
3) Choose a patron, enable hold phone notices for that patron
4) Place a hold for a patron, and check it in so it's marked as waiting
5) Note the phone notice generated for the patron
6) Apply this patch
7) Repeat step 4
8) Note that this time, a phone hold waiting notice is not generated
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Amends condition with an additional or statement. Shoudn't affect
anything but phone notices. Change appears logical.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
The feed message transport type exists but nobody knows how it can be
used. This patch assumes nobody uses it and removes it.
Before removing the entry in the DB, the updatedb entry will check if no
foreign key exists in order to avoid a delete in cascade.
This path does *not* deletes the occurrences in the templates.
Test plan:
Apply this patch, execute the updatedb script and verify the
message_transport_types table does not contain the 'feed' entry if it was not
in use.
Signed-off-by: Nick Clemens <nick@quecheelibrary.org>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Patch will remove the confusing option to configure message texts for feeds.
Checked on a database with an existing feed type notice (not deleted) and
on one without any references to the feed message transport type (deleted!)
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Without patch:
-------------
Make payment for patron who has fines
Select the Pay Amount button and add a note in the note box.
Select confirm
Result: The note does not display in list
With patch:
----------
Result: The note displays in list
Bonus testing: The note is included in system logs as well (Home:Tools:Logs)
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
This patch fixes the issue for bootstrap. It was earlier fixed for the prog theme.
Solution: Prevent jQuery.fn.highlight = function(pat) to be called with empty pat
by skipping empty values in highlight loop.
To test:
Without patch, do a search as in comment #1 odr #3
Result: Endless loop (Script not responding)
Appply patch:
No endless loop, page displays OK
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
To reproduce you need to search a string with 2 spaces.
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
I didn't manage to reproduce the problem, but I couldn't find
any regression either trying multiple searches.
Trusting Marc's and Jonathan's testing.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Modified:
serials/subscription-detail.pl - fix split command
Testing:
I Apply the patch
0) Select a serial subscription;
1) Click Edit -> Edit subscription;
2) Click Next;
3) Click Test prediction pattern;
4) Select 3 Vols;
5) Save subscription;
6) Select Planing tab;
7) Valide "Irregularity: 3 issues".
Signed-off-by: Aleisha <aleishaamohia@hotmail.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Note: You have to select more than 1 issue as not to
be published to see the problem.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
If you use a claimissue notice to claim serials, the generated letter
will be
<order>Title1, Author1</order>
<order>Title2, Author2</order>
...
<order>TitleN, AuthorN</order>
This patch geds rid of these tags.
Test plan:
1/ Create a claimissue notice with something like:
<<LibrarianFirstname>>
<<LibrarianSurname>>
The following issues are in late:
<order><<biblio.title>>, <<biblio.author>> (<<biblio.serial>>)</order>
2/ Generated late serial issues.
3/ Send notifications to vendor.
4/ The order tags should not exist anymore in the sent email.
You can see bug 5342 for a more detailled test plan.
Note for QA: This should have been done in GetPreparedLetter, but I did
not find a better way to do.
Signed-off-by: Paola Rossi <paola.rossi@cineca.it>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Works as described. Tested having the <order> tags on one line
and also for a multi-line layout.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Whenever a librarian is prompted with a form during the checking process
( hold to fill, transfer, etc ) the date to backdate returns to is forgotten!
Test Plan:
1) Begin returning items with a remembered backdate
2) Check in an item that need transfered or has a hold
3) Complete the action requested
4) Note your set return date is now lost
5) Apply this patch
6) Repeat steps 1-3
7) Note your backdate is now remembered!
Signed-off-by: Sean McGarvey <smcgarvey@pascocountyfl.net>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Updates the en permission file and adds a database
update to fix the permisson.
The update is done using the description, to not
untranslate non-English permission descriptions.
To test:
- Run the database update on your existing installation
and check the typo is gone
- Run the web installer and check there are no errors
and permissions are loaded correctly (and without typos)
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Test plan works correctly.
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Fix a typo. Not test plan required, just a look at default UNIMARC framework.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Due to a disconnect between how booleans are handled in Perl vs JSON,
all checkins done from the checkouts table will have fines forgiven!
This is because the parameters sent to svc/checkin are coded in JSON
and the JSON false is treated in Perl as the string 'false' which
evaluates to true!
Test Plan:
1) Find a patron with an overdue item that has a fine
2) Return the item via the checkouts table on circulation.pl or moremember.pl
3) Note the fine is forgiven
4) Apply this patch
5) Repeat steps 1-2
6) Note the fine is no longer forgiven
Signed-off-by: Liz Rea <liz@catalyst.net.nz>
Verified bug, fix corrects the problem.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Also tested waiving fines form the check-in page still works.
Passes tests and QA script.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
The suggestion table does not contain a 'branch' column name.
The script should pass 'branchcode' to C4::Suggestion::NewSuggestion.
Test plan:
0/ Enable the AllowPurchaseSuggestionBranchChoice pref
1/ Create a suggestion at the OPAC should not raise a DBIx::Class error.
I could reproduce the bug.
With patch bug is gone.
Signed-off-by: Marc Véron <veron@veron.ch>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Works as described, thx for the quick fix.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Signed-off-by: Magnus Enger <digitalutvikling@gmail.com>
Fixes names and mappings for contributors from Oslo Public Library.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
According to the manual, "Items will stay in the PROC location until
they are checked in".
This is not the actual behavior. Right now items will only change from
PROC to CART, and that is only if InProcessingToShelvingCart is enabled.
Some libraries want to use the PROC to permanent location feature,
without using the CART.
Additionally, the location is only removed if using returns.pl, but
that is not what the manual says either. What if the library uses
SIP2 devices for handling returns? This should be taken into
account.
Test Plan:
1) Apply this patch
2) Set an item's current location to PROC, and it's permananet location
to a different location.
3) Check the item in any way you wish
4) Note the shelving location is updated to the permanent location
5) prove t/db_dependent/Circulation/Returns.t
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
I tested this with items which had items.location set to 'PROC' and
items.permanent_location set to NULL, '', and a real value, and it
worked correctly in all cases. I tested with check-ins from returns.pl
and from the table of checkouts in circulation and the PROC location was
correctly removed in both cases.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Works as described, passes tests and QA script.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
There are still 9 patches of Fridolin under an old email address.
This patch also takes them into account.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
Tested by generating the release notes. I noticed that
the 'company' list often has a domain name instead
of a company name, maybe something for another patch.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Katrin Fischer <katrin.fischer.83@web.de>
