Colin Campbell [Wed, 14 Aug 2013 14:26:34 +0000 (15:26 +0100)]
Bug 10728: fix log noise generated by subscription-renew.pl
Error log is showing warnings because mode is undefined
and we are doing string comparisons on it.
Set it to a default value (we were already assigning it a variable
which we were not using) and use the the result in the comparison
To test, after applying the patch:
[1] Use the renew link from the subscription detail page to renew
a subscription.
[2] Verifying that doing this did not add a warning containing
'subscription-renew.pl: Use of uninitialized value $mode in string eq '
to the Apache error log.
Bug 10642: fix inappropriate uses of $sth->finish() in C4::RotatingCollections.pm
From the man page
finish()
Indicate that no more data will be fetched from this statement handle
before it is either executed again or destroyed.
You almost certainly do not need to call this method.
Adding calls to "finish" after loop that fetches all rows is a common
mistake, don't do it, it can mask genuine problems like uncaught fetch errors.
To test:
Verify that prove -v t/db_dependent/RotatingCollections.t passes
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, passes UT provided by bug 10653
Marc Veron [Sun, 7 Oct 2012 17:54:44 +0000 (19:54 +0200)]
Bug 8887: only display desired result when doing exact syspref search
This patch fixes an issue where entering the complete name of a system
preference when doing a syspref search in the staff interface resulted
in the display of *all* preferences belonging to the desired one's
module.
Since providing a more specific search string should result in getting
back more specific results, the previous behavior was not intuitive.
Test scenario:
a) In sysprefs, do a search with partial match (e.g. intranetcolor)
-> Result shows one entry: intranetcolorstylesheet
b) Do an exact search. e.g. intranetcolorstylesheet
-> Result shows all Staff Client preferences
Apply patch, test again. Now b) behaves like a).
Signed-off-by: Mirko Tietgen <mirko@abunchofthings.net> Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, works as advertised.
Jonathan Druart [Mon, 17 Jun 2013 12:56:22 +0000 (14:56 +0200)]
Bug 10481: FIX No enrollment fee when changing patron category.
When a patron changes to a category with enrollment fee, they
are not generated.
Test plan:
- Choose a category without fee (e.g. Kid)
- Add an enrollment fee for another category (e.g. Young adult)
- Choose a kid and change his category to "Young adult".
- Note the warning message "Fees & Charges: Patron has Outstanding fees
& charges of XX" on the check out page.
This patch also moves all instances of adding the enrollment fee
to a new routine in C4::Members, AddEnrolmentFeeIfNeeded(), so
additional tests include:
- Register a new patron and give it a category that has
an enrollment fee. Verify that the fee is charged.
- Renew the patron. Verify that the additional fee is charged.
- Register a new patron with a child patron category.
- Use the 'update child to adult' menu option to change the
patron's category to one that is fee-bearing. Verify that the
enrollment fee was charged.
Signed-off-by: Owen Leonard <oleonard@myacpl.org> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit f8eb19163bd4e59e99b6dae90a5712b6a7be52f6) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 98397ed4b10b03fb0b00e9550b571c3cc47407dc) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 98397ed4b10b03fb0b00e9550b571c3cc47407dc)
This patch changes the MARC21 007 plugin editor so that it uses
1 character instead of 4 characters for position 01.
TEST PLAN:
1) Create new record/edit an existing record
2) Using the tag editor for the 007 field, choose "Videorecording"
3) Click OK
4) Change the second character in the 007 from "c" to "d"
5) Open the tag editor again
6) Note that position 01 will still say "c - Videocartridge" rather
than "d - Videodisc".
7) Exit tag editor
8) Apply patch
9) Reload the page (Shift + Refresh is always useful)
10) Open the tag editor again
11) Note that position 01 will now say "d - Videodisc"
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 73ac970f734b0f04fff5561aa35bb9d9d47aa943) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Fix worked as expected.
(cherry picked from commit 8a407d8d02463cd15423735dc6030da2c8f5e3a7) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 8a407d8d02463cd15423735dc6030da2c8f5e3a7)
Fridolyn SOMERS [Mon, 17 Jun 2013 15:01:33 +0000 (17:01 +0200)]
Bug 10483: fix item uniqueness check on order form
When syspref "UniqueItemFields" is defined, the item uniqueness is
tested in acquisition by an AJAX call to check_uniqueness.pl. This
patch fixes an issue where check_uniqueness.pl wasn't looking
at the correct CGI parameters.
Test plan :
- Select an existing item with barcode
- Add "barcode" to "UniqueItemFields" syspref (use space as separator
for multiple values)
- Set "AcqCreateItem" syspref to "Create an item when placing an order"
- Go to an acquisition basket
- Create a new order from empty
- Enter existing barcode in item form and save
=> You get an alert that barcode already exists and order is not saved
- Enter a non-existing barcode in item form and save
=> Order and item are created
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 1c9135262b9233f59fd05e34c90980e7dc8be338) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit df428a49e5c8b9106771d308bd1c1dea62d64429) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit df428a49e5c8b9106771d308bd1c1dea62d64429)
Kyle M Hall [Fri, 16 Aug 2013 11:21:10 +0000 (07:21 -0400)]
Bug 10664 [QA Followup] - fix error in overdue_notices.pl if there is no active currency
Fixes the following koha-qa.pl error:
FAIL misc/cronjobs/overdue_notices.pl
OK pod
OK forbidden patterns
OK valid
FAIL critic
# Variables::ProhibitConditionalDeclarations:
Sophie Meynieux [Wed, 31 Jul 2013 06:04:25 +0000 (08:04 +0200)]
Bug 10664: fix error in overdue_notices.pl if there is no active currency
If there is no active currency, when running overdue_notices.pl
you get a sofware error "Can't use an undefined value as a HASH
reference" on GetCurrency() function call.
With this patch, if there is no active currency, fines are formated
by default 0.00 and notices are correctly generated.
Test plan :
- have at least one borrower with overdue that should trigger a notice
- verify that there is no active currency (Adminitration > Currency &
Exchange rates)
- run misc/cronjobs/overdue_notices.pl -v -n (to get the notice directly
on output)
=> without the patch you get the software error
=> with the patch, notices are correctly generated
If an active currency is defined, the script overdue_notices.pl runs
the same with or without patch
Fridolyn SOMERS [Tue, 28 May 2013 13:43:05 +0000 (15:43 +0200)]
Bug 10362: ensure that all messages and alerts are shown upon checkin
If an item is returned and has mutiple error or warnings (e.g.,
it both needs to be transferred and the patron was debarred
in the course of the return), because of a bug in the template,
not all of the messages would be displayed.
This patch changes the template to show both messages and alerts.
Test plan :
- Perform an checkout that will a checkin :
* need a transfert (return in a different branch)
* set user debarred (depends on issuing rules)
- Perform check-in
=> You get tranfert message and debarment alert
Colin Campbell [Mon, 12 Aug 2013 15:15:37 +0000 (16:15 +0100)]
Bug 10712: Save missing config variables to install log
The variables for AUTH_INDEX_MODE and ZEBRA_TOKENIZER were not
being saved to the koha_install_log file. As a result when using the
--prev-install-log option in Makefile.PL the user was still prompted
to select them.
This patch adds the appropriate variables to the list saved in
the install log.
To test:
On a (non-package) system installed/upgraded without the patch running
'perl Makefile.PL --prev-install-log /path/to/koha-install-log' will
still prompt you to choose authorities indexing mode and select
between chr and icu, after upgrading with this patch applied the same
command should get the responses from the install log and not require
user intervention.
The values can be seen written into misc/koha-install-log
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 40e6f6a8f22b8af7adef79a0ece80a61a69ccc50) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 521d260a5931119c688fd872c30041ffe1c5c692) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 521d260a5931119c688fd872c30041ffe1c5c692)
Bug 10643: fix inappropriate uses of $sth->finish() in C4::ClassSource.pm
This patch gets rid of finish() and replace prepare_cached by prepare.
From the man page
finish()
Indicate that no more data will be fetched from this statement handle
before it is either executed again or destroyed.
You almost certainly do not need to call this method.
Adding calls to "finish" after loop that fetches all rows is a common
mistake, don't do it, it can mask genuine problems like uncaught fetch errors.
To test:
prove t/db_dependent/ClassSource.t
t/db_dependent/ClassSource.t .. 1/24 prepare_cached(SELECT * FROM `class_sort_rules` WHERE `class_sort_rule` = ?) statement handle DBI::st=HASH(0x92dbedc) still Active at C4/ClassSource.pm line 249
prepare_cached(SELECT * FROM `class_sources` WHERE cn_source = ?) statement handle DBI::st=HASH(0x926876c) still Active at C4/ClassSource.pm line 137
attempting to use non-existent class sorting routine routine1
prepare_cached(SELECT * FROM `class_sort_rules` WHERE `class_sort_rule` = ?) statement handle DBI::st=HASH(0x92dbedc) still Active at C4/ClassSource.pm line 249
prepare_cached(SELECT * FROM `class_sources` WHERE cn_source = ?) statement handle DBI::st=HASH(0x926876c) still Active at C4/ClassSource.pm line 137
t/db_dependent/ClassSource.t .. ok
All tests successful.
Files=1, Tests=24, 0 wallclock secs ( 0.02 usr 0.00 sys + 0.12 cusr 0.01 csys = 0.15 CPU)
Result: PASS
Bug 10621: use correct from-address for subscription alert emails
From-address and to-address were the same (patron's email) for
subscription alerts. This patch changes 'from' the branch or
kohaadminemailaddress
To test
- add a subscription in staff/serials in case you don't have any
- enable patron notifications or the subscription
- in the OPAC, subscribe to the serial
- in staff/serial, receive an issue of the serial
Before applying the patch, the email that is supposed to be sent
has the patron's email as 'from' and 'to' (and is likely to fail).
If you follow the steps after applying the patch, the email alert
should have the 'from' address of the patron's branch or
kohaadminemiladdress -- which should also work fine with the MTA/SMTP
you have set up for messaging.
Owen Leonard [Fri, 28 Jun 2013 19:10:14 +0000 (15:10 -0400)]
Bug 10516 - Improve titles, breadcrumbs, and confirmation messages for Z39.50 server adds and updates
The template for adding/editing Z39.50 servers shows the same title,
breadcrumbs, etc. for add and edit operations. This patch creates
separate text for each case.
To test, try both adding and editing a Z39.50 server. The page title,
breadcrumb, and heading should correctly reflect the operation you
perform.
Signed-off-by: Liz Rea <wizzyrea@gmail.com> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Works nicely and all strings are translatable.
Mathieu Saby [Fri, 3 May 2013 21:07:37 +0000 (23:07 +0200)]
Bug 10189: Translate values in UNIMARC 128b/c cataloguing plugins to English
Cataloguing plugins for UNIMARC 128b and 128c fields are only in Ffrench.
This patch translates them.
Source : http://blue.lins.fju.edu.tw/mao/marc/unicon.htm
Note : 128b and 128c are deprecated in last version of UNIMARC Manual (field 145 used instread).
But they are still used in French Sudoc network.
To test : in a UNIMARC english Koha, edit a record and use 128b and 128c
cataloguing plugins. Check everything is in english.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Thx for translating those!
Verified new strings get parsed into the po files.
Templates still contain lots of tabs, those can be fixed separately.
Colin Campbell [Tue, 9 Jul 2013 10:18:49 +0000 (11:18 +0100)]
Bug 10507: improve warning about duplicate patron attribute value
If a warning about a duplicated patron attribute is returned,
the value was being passed back for display in the warning
message with the same name as the variable in errors
As the message is displayed in the context of errors
the error value 1 was being displayed not the value
duplicated as intended.
Pass the value to the template with a unique name.
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
I fixed another tiny typo while signing off: "is already in use" instead
of "is already is use."
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, works as advertised.
Katrin Fischer [Mon, 1 Jul 2013 21:44:39 +0000 (23:44 +0200)]
Bug 4907: make 'Organize by' filter in suggestions reflect selected option
When using the 'Organize by' option to change the display of the
suggestions, the option is applied, but the pull down always jumps
back to the first entry 'Status'.
This patch fixes that.
To test:
- Try out all settings and verify that after the page reloads the selected
option is applied to your suggestions and the pull down shows
the option you selected.
Note: Until bug 10519 is properly fixed the tabs will not show
the correct descriptions, but the URL will show you the selected
option as parameter: displayby=acceptedby
Bug 5858: Followup for INSERT IGNOREs in marc21_framework_DEFAULT.sql
Changes 8 INSERTs into INSERT IGNOREs.
This allows current installs to benefit easier from the new tags/subfields.
Note that running with --force will not achieve the same! The multi-value
inserts will still be aborted, though execution continues.
Test plan:
Run the file or do a new install.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Ran the file on a current install manually.
Bug 5858: update default MARC21 framework to April 2013 revision
Applies to new installs, although with the follow-up
patch, users can run the installer/data/mysql/en/marcflavour/marc21/mandatory/marc21_framework_DEFAULT.sql
script through the mysql client to add new tag and
subfield entries.
To test:
1) Delete MARC21 default framework
2) Apply patch
3) Test that new default framework loads without problem
Signed-off-by: Mason James <mtj@kohaaloha.com> Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Run this file with and without this patch.
This patch adds 24 tags (338 vs 314) and 323 subfields (3951 vs 3628).
Verified that last QA comments were incorporated.
Followup Bug 9353: Missing subfields on MARC21 frameworks
This followup changes updatedatabase.pl.
Only touch frameworks that have the missing subfields.
If both patches are applied, the test plan changes.
To test:
1) dev master system with default and example MARC21 frameworks
2) Go to Home > Administration > MARC frameworks
3) Click on 'MARC structure' of any framework
4) Search for tag 029 (or 863/4/5)
5) Click 'Subfields', there are none.
6) Apply both patches
7) run 'perl installer/data/mysql/updatedatabase.pl'
8) Repeat steps 2-4
9) Click 'Subfields', now they are.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Update works nicely. Checked various frameworks and
results were ok. Also new fields are hidden. Signed-off-by: Jared Camins-Esakov <jcamins@cpbibliography.com>
(cherry picked from commit da4bc6a6315a7f5aca1bb3105549758b8acd7592) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit da4bc6a6315a7f5aca1bb3105549758b8acd7592)
Solved conflict in updatedatabase.pl
This two are necessary for Bug 5858 Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit 459422c8bad09a8301cf47a7b39abf15a3b265ff)
863abcdefghijklmnopqstvwxz68 (same for 864/5)
(http://www.loc.gov/marc/holdings/hd863865.html)
To test:
1) dev master system with default and example MARC21 frameworks
2) Go to Home > Administration > MARC frameworks
3) Click on 'MARC structure' of any framework
4) Search for tag 029 (or 863/4/5)
5) Click 'Subfields', there are none.
6) Apply the patch
7) Replace old frameworks, easiest by command line:
mysql -uuser -ppass kohadb -e 'delete from biblio_framework; delete from marc_tag_structure; delete from marc_subfield_structure'
mysql -uuser -ppass kohadb < installer/data/mysql/en/marcflavour/marc21/mandatory/marc21_framework_DEFAULT.sql
mysql -uuser -ppass kohadb < installer/data/mysql/en/marcflavour/marc21/optional/marc21_simple_bib_frameworks.sql
8) Repeat steps 2-4
9) Click 'Subfields', now they are.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Updated frameworks are ok.
Note: Only en frameworks are updated, so if other languages need
those fields too, we need some followups. Signed-off-by: Jared Camins-Esakov <jcamins@cpbibliography.com>
(cherry picked from commit d0f3a49d9672cd018a51168b05db7a9d1f17f250) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit d0f3a49d9672cd018a51168b05db7a9d1f17f250) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit e22738e3b16593ef5282b493818fe220fb889d24)
Owen Leonard [Wed, 5 Jun 2013 18:03:51 +0000 (14:03 -0400)]
Bug 10346 - "Add multiple copies" should be labelled "Add multiple items"
The Koha interface standardizes on the term "item" instead of "copy," so
the "Add multiple copies" button would be more correctly labeled "Add
multiple items."
To test, view the add item screen for an existing or new record. You
should see a button labeled "Add multiple items."
Signed-off-by: Cedric Vita <cedric.vita@dracenie.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 79df5211d6faff8ab31b46c815ed50c70093a057) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 7000aa561c7fb33c422f9b60c4f7590a6071ee08)
Owen Leonard [Mon, 10 Sep 2012 12:34:33 +0000 (08:34 -0400)]
Bug 7598: make strings in acquisitions JavaScript translatable
English strings in acq.js cannot be processed by the translation script.
This patch moves the strings to the templates to be set as JS variables
for use by the script.
To test:
- On the uncertain prices page, enter an invalid price and click save.
You should see a JavaScript alert pointing this out.
- When adding a fund to an existing budget, enter an amount which
exceeds the amount in the parent budget. You should see a JavaScript
alert pointing this out. There is a similar function which has been
corrected, checkBudgetParent(), but I don't know how to trigger it for
testing.
The closebasketgroup function has been updated but as far as I can tell
it's not actually in use.
Signed-off-by: Srdjan <srdjan@catalyst.net.nz> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit f69e13665da007a6ab28e4c1d5a7d1a50f3ac3ac) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit d8eeb697c873b1ba9366bbce275a14d004931f40)
David Cook [Thu, 27 Jun 2013 06:02:08 +0000 (16:02 +1000)]
Bug 9801: display facet labels in search results only when there are facet values
Currently, the headings for facets (such as location) are showing even
if there are no actual values/links for that facet type.
This patch simply checks if the facet array contains any facets before
printing the heading and HTML for each facet value/link.
Test Plan:
Before applying:
0) If you want to check the location facet (which is easiest),
enable singlebranch mode in the globals system preferences.
1) Do a search for records/items that you have no shelving location
specified (or simply remove the shelving location for a small range
of records that you can find in a search query).
2) Note that "Location" appears amongst the facets even though there
are no values/links present. (If you look at the HTML source, you'll
see some empty HTML tags have been printed out.)
Apply patch.
3) Flush your cache and refresh your page.
4) You should no longer see a "Location" facet (or those empty HTML
tags) on the facet sidebar.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Template changes only.
Checked according to test plan, made sure location facet
still shows up when values exist, otherwise it's hidden.
It might make sense to add ids to the facets so libraries
can decide to hide some of them easily. Or alternatively
add a system preference to make that possible.
David Cook [Thu, 13 Jun 2013 06:46:46 +0000 (16:46 +1000)]
Bug 10457: fix SQL error when using 'export select items data' in serials claims page
Currently, if you use the "check all" checkbox next to the "Vendor"
table header, and try to "Export selected items data", you'll
see the following SQL error in the Apache error log:
DBD::mysql::st execute failed: You have an error in your SQL syntax;
check the manual that corresponds to your MySQL server version for
the right syntax to use near 'on ORDER BY title' at line 12 at
/.../lib/C4/Serials.pm line 1828.
This is because the Javascript that produces the re-direct URL is
including the value of the checkall checkbox, which is "on",
as a "serialid".
We need to only take checkboxes with a name of "serialid" when
crafting the re-direct link.
This patch adds the [name=serialid] attribute selector to the Jquery
that creates the URL to lateissues-excel.pl from claims.pl.
Test Plan:
Before applying the patch:
1) Go to Serials > Claims (found on the left navigation bar of Serials)
2) Choose a vendor that has serials to claim (or create some late
serials for a vendor so that they will show up)
3) Click on the checkbox to the left of the "Vendor" table heading.
4) Click "Export selected items data" at the bottom of the page
5) Check the Apache error log. Notice that you'll get a SQL error
like the one in the message above.
Apply the patch.
Repeat Steps 1-4.
Note that you should now have a CSV file downloading rather than
getting a SQL error.
Owen Leonard [Mon, 8 Jul 2013 18:37:03 +0000 (14:37 -0400)]
Bug 10553: make public lists available on OPAC login page
If you are not logged in to the OPAC, looking at the login page, and you
click the Lists button to see public lists it says there are none. This
patch corrects Auth.pm so that it loads the list of public lists in this
situation.
To test you must have at least one public list. Make sure you are logged
out of the OPAC and visit the login page (/cgi-bin/koha/opac-user.pl).
Clicking the "Lists" button should show you a list of public shelves.
Signed-off-by: Liz Rea <liz@catalyst.net.nz>
works as described, and list button is not shown when opacpublic is disabled.
bug 10559: remove spurious "statement handle still active" warnings
Generating (e.g.) overdue notices can result in spurious warnings in
the cronjob logs:
$ ./misc/cronjobs/overdue_notices.pl -t -library CPL
prepare_cached(SELECT * FROM issues WHERE itemnumber = ?) statement handle DBI::st=HASH(0x54a7828) still Active at C4/Letters.pm line 589
This patch removes the warning by making sure that the relevant statement
handle is finished after fetching its first row of results.
To test:
[1] Set up an overdue loan such that running overdue_notices.pl will
trigger the generation of a notice.
[2] Run overdue_notices.pl -t and note the warning message.
[3] Apply the patch.
[4] Run overdue_notices.pl -t again and note that the warning message
is no longer displayed.
[5] Check the message_queue table and verify that the overdue
notices generated in steps 2 and 4 have the same text.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Srdjan <srdjan@catalyst.net.nz> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 2bab95cac5f43b034adec36ec62a8c8f7f520627) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 5e7452758949e037fcf286cead721b559dd7d0e3) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 5e7452758949e037fcf286cead721b559dd7d0e3) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit cca7c4fbde1ba5074937251d9fd3c8b3d86deef7)
Bug 10595: don't display OpacTopissue page when system preference is turned off
When the syspref is turned off, the link is hidden but the page is still
accessible. This patch checks the syspref and redirects to 404 if it is disabled.
Test plan:
Before applying the patch
Enable OpacTopissue
Observe there is a link to opac-topissues.pl next to Advanced search etc.
Follow that link, observe that opac-topissues.pl is loading
Disable OpacTopissue
Observe the link is gone
Manually open /koha/opac-topissues.pl and check that it is loading even though the feature is disabled
Apply patch
(OpacTopissue still diabled)
Manually open /koha/opac-topissues.pl and observe that you are being redirected to the 404 page.
Enable OpacTopissue
Follow the link or manuall open opac-topissues.pl and check that it is working
bug 10549: make the ILS-DI services advertise that they return UTF-8
This patch changes a few occurences of ISO-8859-1 to UTF-8
within the XML generation of the ILS-DI module.
To test:
- Activate ILS-DI system preference
- Go to [youropac]/cgi-bin/koha/ilsdi.pl
- Check all examples in the documentation for the correct
encoding
- Check GetAvailability gives you the correct encoding and
check the source for the correct encoding
David Cook [Fri, 5 Jul 2013 02:00:36 +0000 (12:00 +1000)]
Bug 10541: enable cross-browser AJAX in additem.js
Currently, the additem.js is using "indexOf" to search
for a value in an array. While this works in Chrome, Firefox,
and IE > 9, it fails miserably in IE 8 and 7 (which don't have
the indexOf method). This means that users aren't able to add
items using the acquisitions module!
Instead of using "indexOf", we should be using the jQuery function
$.inArray. It was added in jQuery v1.2 (3.8.0 uses v1.3.2 so even
our oldest supported release can use this method). It's perfectly
cross-browser compatible...works in Chrome, Firefox, and every
version of IE that I've tried (i.e. 7, 8, 9).
Test Plan:
Before applying patch:
0) Switch to Internet Explorer 7, or 8, or 9, or 10.
If you're using IE 9 or 10, you'll need to change the Document Mode to
IE7 standards or IE8 standards.
You can do this by opening Internet Explorer 9 or 10, pressing F12 (or
clicking on the gear in the top right corner and choosing
F12 Developer Tools), and then clicking on "Document Mode" on the
top toolbar. There, you can change to IE7 or IE8 standards.
N.B. This is not always a perfect emulation in every case, but this
time it does show you the bug.
1) Set the system preference AcqCreateItem to "receiving an order"
2) Go to Acquisitions
3) Either:
a) Receive a shipment for a basket with items
b) Create a new basket, create an order, close the basket, and
then do 3a)
4) In the "Item" fieldset, fill out some fields such as barcode,
Date acquiried, Public note, etc.
5) Click "Add" at the bottom of the fieldset
6) Note that while the item may have been added, the "Item" fieldset
is not being shown again. You may also notice a Javascript error
appearing in a pop-up window or you might see a yellow warning flag
on the bottom status bar.
APPLY THE PATCH
7) Do a full refresh of the page (hold down shift and press the refresh
button on the browser next to the address bar), and try adding items
again.
8) Note that you receive no warnings and that items are added correctly
as they would be in Firefox or Chrome.
OPTIONALLY
9) To be sure that I haven't broken anything, go through the same steps
in IE9 (with IE9 standards) or Chrome or Firefox. Everything should be
working.
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
Tested in IE10 in IE7 mode and IE9 mode. Also tested in Firefox.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, works as advertised.
C4:Circulation:GetUpcomingDueIssues is used in the advance_notices.pl
script. This patch corrects an error in its handling of the maxdays
parameter that resulted in it picking up *all* upcoming due loans and
recently overdue loans.
Test plan :
- Create an issue with a date due in the paste
- Create an issue with a date due in two days
- Launch advance notices with due date in max 2 days : perl misc/cronjobs/advance_notices.pl -c -n -v -m=2
=> You get a warn "found 0 issues"
- Launch advance notices with due date in max 3 days : perl misc/cronjobs/advance_notices.pl -c -n -v -m=3
=> You get a warn "found 1 issues"
Signed-off-by: Mathieu Saby <mathieu.saby@univ-rennes2.fr>
I did the following test :
- 1 book to check in 2 days
- 2 books to check in in the past
before applying the patch :
$perl ../misc/cronjobs/advance_notices.pl -c -n -v -m=2
getting upcoming due issues at ../misc/cronjobs/advance_notices.pl line 203.
found 1 issues at ../misc/cronjobs/advance_notices.pl line 205.
I changed the value of "-m" : 0, 1, 2, 3, 4
=> always 1 issue found (the book to check in in 2 days)
after applying the patch :
$perl ../misc/cronjobs/advance_notices.pl -c -n -v -m=2
found 0 issues
for m = 0, 1, 2 => 0 issues
$perl ../misc/cronjobs/advance_notices.pl -c -n -v -m=3
found 1 issues
for m = 3,4,5 => 1 issues (the book to check in in 2 days)
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, works as advertised.
Katrin Fischer [Wed, 10 Jul 2013 20:43:41 +0000 (22:43 +0200)]
Bug 10563: Fix broken 'Edit' link on check expiration page
This fixes the broken 'Edit' (subscription) link on the
check expiration page in serials.
To test:
- Add a subscription
- Go to the check expiration page
- Search for your subscription, use an expiration date in the
far future
- Verify the edit link on the result list doesn't work right
before applying the patch, but works after
Sophie Meynieux [Fri, 26 Apr 2013 13:06:40 +0000 (15:06 +0200)]
Bug 10127 : Printoverdues.sh should tar only pdf files
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
From the bug report:
At the end of printoverdues.sh, an archive is generated
containing all files present in directory given as parameter.
As archives are generated in this given directory, they
contain previously generated tar files, creating oversized files.
Only .pdf files should be archived
All tests and QA script pass.
Worked ok in my tests, only packign PDF in the given directory.
This routine is not in used and does not make sense. It should not be
used later.
Signed-off-by: Srdjan <srdjan@catalyst.net.nz> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, not references to get_branch_code_from_name found.
Owen Leonard [Tue, 2 Jul 2013 13:59:21 +0000 (09:59 -0400)]
Bug 10422 - Remove references to unused and non-existent wizard.css
doc-head-close.inc contains a reference to "wizard.css" which doesn't
exist. The option to include it never evaluates as true, and that
section can be removed from the template.
To test, apply the patch and view any page in the staff client.
Everything should be styled as before. A search of the Koha source
should return no references to "wizard.css" or a "wizard" variable.
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
No occurrence of "wizard" in pl/pm files.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl
Jason Etheridge [Fri, 8 Mar 2013 15:41:06 +0000 (10:41 -0500)]
Bug 9770: fix sorting of Dewey call numbers that contain prefixes
C4::ClassSortRoutine::Dewey can pad the wrong part of a call number internally.
The subroutine get_class_sort_key tokenizes a call number string (splitting on
periods and whitespace) and counts the number of tokens that solely contain
digits. If there is only one such digit group, a comment in the code states
that it will pad said digit group. However, the bug is that the code assumes
said digit group is the first token, when this may not be the case.
In practice, this can cause poor sorting when used a call number is in the form
of PREFIX _space_ 3DIGITS.
To test:
[1] Create two item records whose class scheme is set to
'ddc' (Dewey) and whose call numbers contain prefixes, e.g.,
J DVD 700.1 ABC and J DVD 850 DEF.
[2] Use the inventory tool to produce a list of item items that include
the two created in step 1. Obsere that that items are sorted
in the incorrect order, with "J DVD 850 DEF" coming before
"J DVD 700.1 ABC". Alternatively, run the following SQL
to see the incorrect sort order:
SELECT cn_sort, itemcallnumber
FROM items
WHERE itemcallnumber LIKE 'J DVD%'
ORDER BY cn_sort;
[4] Apply this patch.
[5] Run misc/maintenance/touch_all_items.pl to force cn_sort to be
recalculated.
[6] Repeat step 2 and verify that the call numbers are now sorted
corrected.
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit dba36a7a1216238a260ea5fbe2218627487e9f19) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit f05642589d94c25ef1b1f01b4a2606fe455400b8) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit f05642589d94c25ef1b1f01b4a2606fe455400b8) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit 404a931efa9365ae084f088111382eb2076a0a7e)
Jason Etheridge [Fri, 21 Jun 2013 19:31:32 +0000 (15:31 -0400)]
Bug 9770: test case for sorting of Dewey call numbers that contain prefixes
This adds a test for C4::ClassSortRoutine::Dewey to check that the
call number "JR DVD 800.1" sorts before "JR DVD 900"
To test:
[1] Apply just this patch.
[1] Run prove -v t/ClassSortRoutine_Dewey.t
[2] Test #7 should fail.
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes test plan and QA script. Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit bce45b4bf55e82345efed2850d9cb5fd77f3c483) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 926e455f19852c1c7b69aa00bd587caff8bccd14) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 926e455f19852c1c7b69aa00bd587caff8bccd14) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit 74b2f3fb4acec7b4adf54f32bb4eabcf51ca3353)
David Cook [Wed, 12 Jun 2013 01:24:09 +0000 (11:24 +1000)]
Bug 10448: can now change framework after duplicating bib record
Changing the framework in the MARC editor immediately after duplicating
a bib record no longer clears the fields.
This patch changes the Changefwk Javascript function so that it passes
the "op" value and the "biblionumberdata" (as the biblionumber) from
addbiblio.pl back to itself, when submitting the form in order to
change the framework.
The reason we need to do this is because the form in addbiblio.tt
is hard-coded to always submit an "op" value of "addbiblio". Currently,
we need to have it hard-coded to "addbiblio", because all the magic
happens in addbiblio.pl when there is an "op" of "addbiblio". If we
always passed the "actual" "op" value, such as "duplicate", nothing
would ever happen when we clicked "save". It seems to me that this
is a flaw in the design of addbiblio.pl.
If we pass the "op" and "biblionumber" when changing frameworks, we're
able to tell addbiblio.pl that we're still wanting to "duplicate" this
"X" biblionumber. However, by having the form still hard-coded to
"addbiblio", when we hit save, the form will do the magic, check if
it's a duplicate, and save the record (or prompt for action if it
is a duplicate).
--
I also noticed that if you make changes to a record, then change
the framework before saving, your changes get cleared (since the
original record from the database is loaded when the page reloads).
It seems to me that this is a bug. Changing the framework should
change the layout while preserving the content. I think most users
would assume that when changing the framework.
This patch also introduces another hidden input into addbiblio.tt and
the Changefwk Javascript called "changed_framework". Basically, if
the Changefwk Javascript is run, it tells addbiblio.pl that the
framework is changed, and it uses the posted data from the form
(which we have been modifying) instead of reloading the record from
the database.
--
Test Plan:
A) Before Applying Patch:
To Show That Changing the Framework Erases All Fields When
Duplicating a Record:
1) Go to any bib record
2) Go to Edit > Edit as new (duplicate). You should see filled
in fields.
3) Change the framework to any other framework than the one
that is currently specified.
4) Note that every single field is now blank
To Show That Changing the Record then Changing the Framework
Ignores Changes, When Editing a Record
5) Go to any bib record
6) Go to Edit.
7) Change the title of the record to "I've changed the title".
8) Change the framework to any other framework than the one that is currently specified.
9) Look at the title. You'll notice it is the original title, and NOT "I've changed the title".
B) Apply the Patch
Also, clear your memcache and shift+refresh your screen. You don't want to use cached templates/javascript.
C) After Applying the Patch
Repeat Steps 1-3 and 5-8.
You should now notice that changing the framework when duplicating the item does not clear all the fields.
You should also notice that any changes you make prior to changing the framework will still exist after changing it.
Owen Leonard [Thu, 11 Jul 2013 14:28:55 +0000 (10:28 -0400)]
Bug 10514: improve visibility of Add item link on new order form
This patch converts the "Add" and "Clear" links
to the standard "submit/cancel" format inside a fieldset. This gives
them a little more visual weight.
Based on the changes made by Liz Rea and Jonathan Druart.
To test:
- create a basket
- add a record to it
- scroll down - the link to add item and cancel should both be more
prominent now.
- Click "Add item" - it should add an item.
Signed-off-by: Liz Rea <liz@catalyst.net.nz>
I still feel weird about the button, but as two people have said they'd rather have the button, I'm alright with it I guess.
Really what I want is people to notice it's there and click it at the appropriate time. I hope this will help that issue.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Leaves the translation problems, but that needs more work and is
out of the scope of this bug.
Tested Add and Update functionality works correctly.
Robin Sheat [Wed, 5 Jun 2013 05:50:29 +0000 (17:50 +1200)]
Bug 10411: add cache control headers for static resources (Debian package)
This adds caching directives to the static resources so that the browser
doesn't ask about them all the time.
By default, Debian apache doesn't have the required module (mod_expires)
enabled, and so this patch will have no effect. In order to enable this
function, run:
a2enmod expires
and restart apache.
Test Plan:
* Using firebug or equivalent, load a page.
* Note that none of the images, css, or js files have a Cache-Control or
Expires header set.
* Add this patch, redeploy the package with it (or overwrite the
apache-share.conf file) enable the 'expires' module, restart apache.
* Force-reload the page
* Note that the images, css, js now have Cache-Control and Expires set
for 12 hours into the future.
Katrin Fischer [Mon, 1 Jul 2013 19:58:57 +0000 (21:58 +0200)]
Bug 10490: Overdue fines cap can't store decimal values
When overduefinescap was added to the issuingrules the datatype
given was decimal. This translates in MySQL to decimal(10,0).
This doesn't allow you to store decimal values and therefore
values like 4.5 are saved as 5 in the database.
To test:
On a current installation:
1) Try to enter 4.5 as Overdue fines cap. Verify that the value
is not correctly saved.
2) Apply patch and run database update.
3) Try adding/changing an issuing rule setting Overdue fines cap
to 4.5 again.
4) Verify the value is saved correctly.
Create a new Koha installation from scratch:
1) Verify that the issuingrules table has been created correctly
and that you can add/mofidy issuingrules correctly.
Because this bug can create data loss, the old database update
has also been changed to avoid this problem for people updating
at a later point in time. Checkout an older version of Koha
pre 3.09.00.027.
1) Run the database updates.
2) Verify again, that adding/modifying issuingrules works correctly.
Signed-off-by: Srdjan <srdjan@catalyst.net.nz> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 16c57940441aac62c1c90538f3c7b7cd386e6d92) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 47df7c9641321b70cd132790d4cfeab500bc34be)
Robin Sheat [Fri, 30 Nov 2012 02:48:50 +0000 (15:48 +1300)]
Bug 10560: make it possible for default interface language to not be English
If you enable another translation, and disable English, then if you dont
have a cookie set, or your browser is not set to that language, you will
get English. So you can not disable English in either the staff client
or the OPAC.
This patch fixes the language selection to do the right thing.
To test you must have at least one other language installed besides
English. Apply the patch and disable the en translation. Koha should
fall back to one of the enabled translations.
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
I added a patch description and test plan, missing from the
original patch.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
I have tested with various combinations of activated languages
and have found no regression. If the cookie is set, the right
language is shown accordingly. Else the first language in the
list seems to be picked. It did never fall back to English
in my tests, when English was explicitly deactivated.
In parcels.pl, and other acquisition pages, the funds are not sorted
by name in combo-box. With a great number of funds, it is difficult
to find one.
This patch adds a default value to $orderby arg of C4::Budgets::GetBudgets.
Test plan :
- Create a new fund with a name beginning with 'z' and set you as owner.
- Create a new fund with a name beginning with 'a' and set you as owner.
- Go to a vendor and click "Receive shipments"
- Look at fund combobox
=> Funds are sorted by name
Signed-off-by: Silvia Simonetti <s.simonetti@cineca.it> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
When testing make sure your funds would sort the other way
around when sorting by code instead of description.
Bug 9519: fix code for Italian language search limiter
'ind' rather than 'ita' was set as the ISO639-2 language
code for Italian in the fr-FR, it-IT, pl-PL, ru-RU, and uk-UA
installation SQL.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
I checked nb-NO and de-DE - language code there had already
been corrected.
Fridolyn SOMERS [Fri, 31 May 2013 08:17:08 +0000 (10:17 +0200)]
Bug 10384: fall back to normal authentication if LDAP server is not available
This fixes a bug where, assuming LDAP authentication is enabled, if a user
tries to log in while the LDAP server is down, the following fatal error
is displayed:
Can't call method "bind" on an undefined value at C4/Auth_with_ldap.pm line 134, <DATA> line 558.
This patch catches this error to allow normal authentication when LDAP connexion fails.
Test plan :
- Configure LDAP connexion with a host not having LDAP. ie :
<useldapserver>1</useldapserver>
<ldapserver id="ldapserver">
<hostname>localhost</hostname>
<base>dc=test,dc=com</base>
<user>cn=Manager,dc=test,dc=com</user>
<pass>passwd</pass>
<replicate>0</replicate>
<update>0</update>
<auth_by_bind>0</auth_by_bind>
<mapping>
<firstname is="givenname" ></firstname>
<surname is="sn" ></surname>
<branchcode is="branch" >MAIN</branchcode>
<userid is="uid" ></userid>
<password is="userpassword" ></password>
<email is="mail" ></email>
<categorycode is="employeetype" >PT</categorycode>
</mapping>
</ldapserver>
- Try to connect with mysql user (defined in koha-conf.xml)
- Try to connect with a user defined in borrowers
You may try to connect with working LDAP connexion
Jonathan Druart [Thu, 25 Apr 2013 07:47:53 +0000 (09:47 +0200)]
Bug 10088: FIX adding an order from a staged file.
Bug 9902 introduced an issue in the C4::Items::PrepareItemrecordDisplay
routine. The existence of $defaulvalue hashref should be tested before
getting to the branchcode key.
Test plan:
Before applying the patch, an error occurred when you try to create an
order from a staged file.
After applying the patch, the error does not appear anymore.
Signed-off-by: Mathieu Saby <mathieu.saby@univ-rennes2.fr> Signed-off-by: Mason James <mtj@kohaaloha.com> Signed-off-by: Jared Camins-Esakov <jcamins@cpbibliography.com>
(cherry picked from commit f2e44716f60578b9663862d7b2ba3b16f7067143)
Kyle M Hall [Wed, 17 Apr 2013 13:33:55 +0000 (09:33 -0400)]
Bug 9533: batch item modification form now shows default field values only by request
This patch makes the pre-population of the fields for the batch
item editor with the default values from the default framework
optional and off by default.
Test Plan:
1) Apply patch
2) Add default values for cost and replacement cost to the
default framework.
3) Browse to Tools / Batch item modification
4) Choose a file, or entire some barcodes in the text area
5) Ensure that the 'Populate fields with default values from default
framework' is *not* checked
6) Click 'Continue'
7) Observe that the fields for cost and replacement cost are blank
8) Click the 'back' button on your browser
9) Check the checkbox to enable the default value
10) Click 'Continue'
11) Observe that the fields for cost and replacement cost now contain
the default values you assigned them in the default MARC framework
Signed-off-by: David Cook <dcook@prosentient.com.au> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Switch works correctly, all tests and QA script pass.
Note: there is a bug with default values for all non text input
fields. Default values won't be applied for those fields as the
pull downs are not preselected correctly with the default values
defined in the framework.
Fridolyn SOMERS [Wed, 26 Jun 2013 15:52:40 +0000 (17:52 +0200)]
Bug 9113: fix handling of certain characters by item batch modification tool
If you're using the batch modification tool and entering a call number
like "E+ 123 ABC", the tool removes the + and puts a space in its place,
e.g., "E 123 ABC"
This is because the form is posted via Ajax by background-job-progressbar.js.
Values are URI-encoded using escape(), but this method does not escape some
characters: * @ - _ + . /
Also, "+" is considered as a space in a URI.
This patch replaces escape() by encodeURIComponent() which encodes every character.
Test plan :
Perform an items batch modification by setting '* @ - _ + . /' in a field
(notes for example) and see that all characters are saved correctly.
Marc Veron [Fri, 14 Jun 2013 14:37:23 +0000 (16:37 +0200)]
Bug 5894: Display all titles when confirming copy of items from cart to list
Adds missing code that splits biblios before calling template.
To test:
- Add some items to cart
- Open cart, select some or all items
- Add to list; with patch, following window shold show something
like "Add 3 items to a list:", followed by list of items
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, works as advertised.
Fridolyn SOMERS [Fri, 7 Jun 2013 08:17:54 +0000 (10:17 +0200)]
Bug 10428: undefined check in add additem.pl
In additem.pl, in the code that generates item form you find:
@values = $itemrecord->field($tag)->subfield($subtag) if ($itemrecord && defined($itemrecord->field($tag)->subfield($subtag)));
This patch adds the check of $itemrecord->field($tag) undef.
I did not found a way to test it because item tag is always defined
when editing an item, but it would be safer/prettier to add this check.
Bug 10029 - CAS authentication fails in OPAC reserve
If OPAC reserve page is accessed without being logged-in, login form is displayed as well as a CAS authentication link (if enabled). A click on this link will lead to CAS server but one comming back to Koha, page shows an error : "ERROR: No biblionumber received".
This is because CAS link only contains the query path "/cgi-bin/koha/opac-reserve.pl", not the query parameters.
This patch adds query parameters to URI sent to CAS.
Test plan :
- Enable CAS
- Go to opac without been logged-in
- Try to place hold on a record
=> You get to /cgi-bin/koha/opac-reserve.pl?biblionumber=XXX showing authentication page
=> Check that CAS link contains query param "biblionumber"
- Click on CAS link and log in
=> Check you return well logged-in to reserve page with biblionumber param
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
I have followed the test plan as far as I could and the links
contain the biblionumber now, which they didn't before.
I couldn't check the CAS login, but my normal login worked
as expected.
All tests and the QA script pass.
Kyle M Hall [Mon, 17 Jun 2013 16:28:13 +0000 (12:28 -0400)]
Bug 10484: serials-edit.pl not checking for barcode field before checking for barcode subfield
The script serials-edit.pl is not checking for the existence of the
barcode field before checking for barcode subfield for autoBarcode =
incremental. If the barcode field doesn't exist, the script dies with
errors.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes tests and QA script.
Checked that adding items on serial receive still works.
Marcel de Rooy [Mon, 3 Jun 2013 13:27:50 +0000 (15:27 +0200)]
Bug 10386: improvements to VirtualShelves.t
Most important: Does no longer delete all shelves!
Checks if there are ten borrowers for testing. But even works without them :)
When creating or modifying lists, takes name clashes into consideration.
Small change to _CheckShelfName in VirtualShelves module. Making it possible to
check a name for a list whose owner has been set to NULL. Note that a test
like field=? with undef for placeholder will not work in MySql.
Test plan:
How do you test a test? Well, you could run it on various databases..
But for real hacking, you could also add some debug lines.
I tested this by forcing 10 undefs in @borrowernumbers.
And by overwriting the return value of randomname with an existing name.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 93e87ca0b61611fc35238bf13eb226f477cf9c4e) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit a97b62379f82a94157916959109c7c1a75582d8a) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit a97b62379f82a94157916959109c7c1a75582d8a)
Fixed minor conflicts:
C4/VirtualShelves.pm Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit 5e9db705a5466ff508d45a9bbfcabd9744efa53f)
Jonathan Druart [Mon, 20 May 2013 14:56:38 +0000 (16:56 +0200)]
Bug 10290: UT: VirtualShelves.t needs to create its own data
Try before the patch:
prove t/db_dependent/VirtualShelves.t
And after, it should produce:
t/db_dependent/VirtualShelves.t .. ok
All tests successful.
Files=1, Tests=82, 6 wallclock secs ( 0.03 usr 0.00 sys + 0.46 cusr 0.03 csys = 0.52 CPU)
Result: PASS
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 44905ac99cee8ddca6c63752d948a74afae20194) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Tests are good :-D. They look fine and work as expected.
(cherry picked from commit e918450e6ad6e661474479f8890f6b141fe4e15a) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Missed this one from previous release
(cherry picked from commit e918450e6ad6e661474479f8890f6b141fe4e15a) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit a16542961f60f5a77d0af86d605a890a5b23782d)
do some validation of the KohaOpacRecentSearches cookie
Add validation of the value of the KohaOpacRecentSearches. In
particular, this patch avoids the generation of an internal server
error when the OPAC is presented with an old cookie that uses the
old Storable-based serialization.
This patch also moves parsing of the cookie value into a
new routine in C4::Auth, ParseSearchHistoryCookie, and adds
a test case.
To test (in conjunction with the previous patch):
Exercise the OPAC search history functionality, after
turning on the EnableOpacSearchHistory syspref:
- As an anonymous user, conduct a variety of searches,
including ones that include non-ASCII characters
- Check the search history and verify that all searches
are listed
- Apply this patch and the previous one.
- Do *not* clear the KohaOpacRecentSearches cookie
- Check the search history and verify that no searches
are listed any more
- As an anonymous user, conduct a variety of searches,
including ones that include non-ASCII characters
- Check the search history and verify that all searches
are listed
- Log into the OPAC
- Verify that current and past searches are listed in
search history.
use JSON rather than Storable for the OPAC search history cookie
To test:
Exercise the OPAC search history functionality, after
turning on the EnableOpacSearchHistory syspref:
- Clear the KohaOpacRecentSearches cookie
- As an anonymous user, conduct a variety of searches,
including ones that include non-ASCII characters
- Check the search history and verified that all searches
are listed
- Log into the OPAC
- Verify that current and past searches are listed in
search history.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com> Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Magnus Enger [Tue, 25 Jun 2013 06:39:23 +0000 (08:39 +0200)]
Bug 10379 - Followup: add koha-rebuild-zebra -q to the man page
The first patch on this bug forgot to add the new -q option to
koha-rebuild-zebra to the man page for that command. This patch
fixes it.
To test, run something like these commands:
$ xsltproc /usr/share/xml/docbook/stylesheet/docbook-xsl/manpages/docbook.xsl \
debian/docs/koha-rebuild-zebra.xml
$ man -l koha-rebuild-zebra.8
This should display the man page. (Remember to remove koha-rebuild-zebra.8
when you are done :-)
$ prove xt/verify-debian-docbook.t
All tests should pass.
Robin Sheat [Thu, 30 May 2013 04:12:13 +0000 (16:12 +1200)]
Bug 10379: silence zebra rebuild cron when no instances are enabled
This adds a -q option to koha-rebuild-zebra that prevents it outputting
help if no instances are provided on the command line. This means that
if you've just installed the package, you won't get messages from cron
telling you how to use the script.
Test plan:
* install koha-common 3.12 or master onto a system with no enabled
instances.
* note that you or the sysadmin will start getting emails every 5
minutes.
* replace /etc/cron.d/koha-common and /usr/sbin/koha-rebuild-zebra with
the versions from this patch.
* note that the emails stop flowing.
Signed-off-by: Magnus Enger <magnus@enger.priv.no>
"sudo koha-rebuild-zebra -q" gives no output, as expected.
Followed the test plan, all looks good. Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit c31e553a72d40bf0b1c099284ecbb4b95833e186) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit cc9b3522b192b8510a4846cd7a4b3aca3e8a0a07) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit cc9b3522b192b8510a4846cd7a4b3aca3e8a0a07) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry oicked from commit 22341e5fe36e54523464b633de6a500bc706e248)
Kyle M Hall [Tue, 19 Feb 2013 16:41:59 +0000 (11:41 -0500)]
Bug 9541: make OPAC login page respect OPAC_CSS_OVERRIDE
When clicking the login link for opac-user.pl in a multiple branch
scenario the environment variable for OPAC_CSS_OVERRIDE was ignored from
the koha-conf.xml file. It seems like is is working on every page in
the opac except for the login page.
Test Plan:
1) Set up a Koha server with 2 separate catalog configurations
( e.g. opac1.kohatest, opac2.kohatest )
2) Set the OPAC_CSS_OVERRIDE directive for separate css files
in each opac
3) Browse to the opac login page, note the css is not applied
4) Apply this patch
5) Reload the page, note the css is now applied
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 71b79d59908cd85130afd8600eff0f93fc79b8a9) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 69a8048eeecacff99b64fa722fa8850891476031)
Jonathan Druart [Mon, 10 Jun 2013 12:05:32 +0000 (14:05 +0200)]
Bug 10439: fix bug preventing the library for a notice from being changed
2 javascript functions should be apply only in a specific case, not for all
views.
Test plan:
- Switch on the IndependentBranches pref and log
in as a superlibrarian.
- Try to change the library for a notice. Before the patch,
no change is applied.
- Try to create a new notice, changing the library from
default in the process. Before the patch, the notice
is not created.
The bug can also be reproduced when IndependentBranches is
off, in which case the staff user need not be a superlibrarian.
Signed-off-by: Srdjan <srdjan@catalyst.net.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 77a5e3e365a50b02c9f224d731e05de90fc667c9) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Test plan works as advertised.
(cherry picked from commit a2e89edf43b2946da0986a5c83b04a60dea04061)
Bug 10464: Patron search on placing a hold now supports autocomplete
The patron search box for placing a hold request on a bib in the
staff interface now supports autocomplete if the CircAutocompl
system preference is enabled.
This patch grabs the autocomplete handling javascript stuff from patron-search-box.inc
and puts an adjusted version on reserve/request.tt. It sets form and input
IDs to match those configured in the autocomplete call, adjusted where needed to
avoid name collisions with the patron-search-box.inc that is present everywhere.
This patch also doubles the width of the patron search input field.
To test:
- On a biblio record, go to the Holds tab, mine looks like
/cgi-bin/koha/reserve/request.pl?biblionumber=2
- A search box is shown for patron searching.
- Search for a patron: it works as usual.
- Make sure CircAutocompl is "Don't try"
- Apply the patch
- Click the Holds tab again, and do the same search: functionality should remain intact.
- Change CircAutocompl to "Try"
- Click the Holds tab again
- Autocomplete should show you the possible options, and once you choose one result,
it should work as usual.
Sponsored-by: Universidad Nacional de Córdoba Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Adding a nice new autocomplete feature.
Colin Campbell [Thu, 30 May 2013 17:23:55 +0000 (18:23 +0100)]
Bug 766: remove CGI::scrolling_list from request.pl
Let the template take care of generating the display.
Lets at least part of request.pl be more succinct.
To test, place a hold and search for the patron using a partial name.
You should get a dropdown select list of patrons and be able to select
the patron to hold for from there.
The functionality should remain unimpaired after this patch is applied.
David Cook [Fri, 14 Jun 2013 03:03:19 +0000 (13:03 +1000)]
Bug 10466: fix glitch in OPAC display of titles saved to list
When displaying (in the OPAC) the set of records saved to a list,
an apparently random number (typically 33 or 34) is sometimes
displayed at the end of the publication description. In particular,
this can occur when XSLT is *not* being used to display search results;
this patch corrects the problem.
Now for the technical details:
This patch checks to see if "size" is undefined. If it is, we add a
blank (i.e. "") value to it in place of undef.
If we do not do this, calling "itemloo.size" will return the size
of the "itemloo" hash, rather than the value for the "size" key.
This is because "size" is a virtual method in Template Toolkit. It's
uncertain why the value is retrieved for the "size" key when there is
a defined value and why TT doesn't use the method instead, and that
it uses "size" as a method only if there is either no "size" key or
if the value tied to the "size" key is null/undef. This might be a
feature or it might be a bug in TT...
In the meantime, we will check to see if it's undefined. If it is,
we'll give it a value.
This bug has been identified in the opac-search.pl, search.pl and
addbooks.pl pages before. To address it, we're currently checking
if there is a "size" key, and if not...we're adding one with a blank
value.
This patch takes up that same idea, although I think it might be better
to rename the variable before passing it to TT in case the behaviour
of TT changes in the future in regards to how it handles virtual
methods.
N.B. Obviously, this only affects users not using XSLTs.
--
Test Plan:
Before applying the patch:
0) Make sure you have opac search result XSLT turned off
1) Find bib records that do not have a 300$c (Dimensions) value.
2) Find bib records that do have a 300$c (Dimensions) value.
(N.B. These values should be stored in the `size` column of biblioitems).
3) Add items from both sets of records to a List
4) Note that records without a 300$c will display a number at the end
of the "Publication" description/string. It should be something like
33 or 34 in most cases.
5) Note that records with a 300$c don't display this number. They just
show the value from 300$c.
Apply the patch.
6) Clear your cache, refresh the page, etc.
7) Note that the number (e.g. 33 or 34) has disappeared from the end
of the "Publication" description/string.
Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Added a small comment at the end of this one line.
Robin Sheat [Mon, 15 Jul 2013 14:06:40 +0000 (02:06 +1200)]
Bug 10590 - parameterise the limit option
The limit option was previously substituted directly into the query. The
previous patch on bug 10590 filters it on input, but there's no reason
not to have it made to work properly in the query for added safety.
To test:
[1] Go to the top checkouts report (http://OPAC/cgi-bin/koha/opac-topissues.pl)
[2] Run the report several times, varying the filters on
number of results and item type.
[3] Verify the the list of top checkouts appears to be correct.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 89cf013a6fadcb1347151798f3fdab0d8c75cd15) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 2e77edf8dfa0350a6e20805e7fa39b97608c804e) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 2e77edf8dfa0350a6e20805e7fa39b97608c804e) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit f5ee81028dda33027daf3c921b38bf06218f9dfa)
Bug 10590 - in opac-topissues limit param is not protected
In opac-topissues page, the limit URL argument is directly added to SQL query.
This patch adds protections : limit must only contain digits and must be lower than 100.
Test plan :
- Edit URL to : /cgi-bin/koha/opac-topissues.pl?limit=10&branch=&itemtype=&timeLimit=999&do_it=1
=> You get the results of 10 most cheched-out of all time
- Edit URL to : /cgi-bin/koha/opac-topissues.pl?limit=&branch=&itemtype=&timeLimit=999&do_it=1
=> You get the results of 10 most cheched-out of all time
- Edit URL to : /cgi-bin/koha/opac-topissues.pl?limit=9999&branch=&itemtype=&timeLimit=999&do_it=1
=> You get the results of 100 most cheched-out of all time
- Edit URL to : /cgi-bin/koha/opac-topissues.pl?limit=WHERE&branch=&itemtype=&timeLimit=999&do_it=1
=> You get the results of 10 most cheched-out of all time
Signed-off-by: Robin Sheat <robin@catalyst.net.nz> Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 57866d6b67c3f8b29290150f21c71395315a73fe) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 5867909f8866b4e68340b88817bfb95f51bc3c9e) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 5867909f8866b4e68340b88817bfb95f51bc3c9e) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit c0cb1f8b0e8aa53707447ee7c27fd15fc0a7b80c)
Fridolyn SOMERS [Thu, 6 Jun 2013 08:57:29 +0000 (10:57 +0200)]
Bug 10425 - Help page missing for Local Use preferences
When viewing Local Use system preferences tab, the help button will lead to an error page.
This is because Local Use tab uses not the same script has other tabs : systempreferences.pl.
This patch adds help/systemprefences.tt
Test plan :
- Go to Local Use preferences : /cgi-bin/koha/admin/systempreferences.pl
- Click on Help link (at right top)
- See you get the same page as other preferences tab help
Marcel de Rooy [Thu, 13 Jun 2013 08:20:27 +0000 (10:20 +0200)]
Bug 10458: Remove obsolete Z3950 module and test
Are not used. Contain several FIXMEs.
Removing them makes life easier.
Test plan:
Actually, you cannot test this.
But for confidence: do a Z3950 search in cataloguing and acquisition.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit 63ba05183e627bbe7278b9064d22e93d39b65461) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 1dbd1d938aabb72021cd8157e111d3f7555fff9d) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 1dbd1d938aabb72021cd8157e111d3f7555fff9d)
Owen Leonard [Mon, 10 Jun 2013 15:14:41 +0000 (11:14 -0400)]
Bug 10297 - categorycode and branchcode should be required when adding patrons
categorycode and branchcode should show as required under all
circumstances, not just when they are included in the
BorrowerMandatoryField system preference. This patch adds the correct
classes and hints to those fields.
To test, view the patron entry form for a new or existing patron.
Category and Library should show that they are required.
Galen Charlton [Wed, 9 Jan 2013 01:32:47 +0000 (17:32 -0800)]
bug 9370: improve splitting of LC call numbers for labels
Use the Perl module Library::CallNumber::LC to parse and split
LC call numbers when generating spine labels.
For example, QH541.15.C6 C25 2012 should be split as follows:
QH
541.15
.C6
C25
2012
To test, create an item with call number QH541.15.C6 C25 2012
and classification source LC, then create a spine label for that
item using a layout of type 'biblio' that has the split call numbers
option enabled. The call number should be split as indicated above.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
(cherry picked from commit edb8bddff074769a95744b6a7fa59eff61e0cfc1) Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
(cherry picked from commit 10f5822daabf34ede50b2b35b5e02324482af9ac) Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
(cherry picked from commit 10f5822daabf34ede50b2b35b5e02324482af9ac) Signed-off-by: Chris Hall <followingthepath@gmail.com>
(cherry picked from commit 56ea58ecda26f6ca4e4e8e437296b2aa2a2419d0)
projects / koha.git / log