Galen Charlton [Tue, 22 Oct 2013 00:14:47 +0000 (00:14 +0000)]
Bug 11108: fix duplicate display of basket whose receipt has just been cancelled
After receiving an order basket, before finishing receiving the shipment,
one has an option to cancel the receipt. This patch fixes a bug where
a basket whose receipt was just cancelled displays on both the pending orders
and already received tabs.
This patch also fixes a minor unitialized variable warning.
To test:
[1] Create a basket with at least one order and close it.
[2] Receive the order, then on the row in the 'Already received'
table, click the cancel receipt link.
[3] In the page that displays, the basket just cancelled displays
on both tables. Clicking the cancel receipt link again results
in an error message.
[4] Apply the patch.
[5] Repeat steps 1 and 2. This time, the cancelled basket displays
only in the pending orders table, as expected.
[6] Verify that after applying the patch, the following no longer
is logged in the Apache error log:
parcel.pl: Use of uninitialized value in string eq at acqui/parcel.pl line...
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Ed Veal <ed.veal@bywatersolutions.com> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes test plan, all tests and QA script.
Works as described.
Owen Leonard [Fri, 25 Oct 2013 14:39:03 +0000 (10:39 -0400)]
Bug 11133: fix markup errors in OPAC place hold form HTML
There are markup errors in the OPAC place hold form which prevent it
from validating successfully. These errors don't seem to cause any
visual problem but can potentially cause problems when debugging
template logic.
To test, apply the patch and validate the page using your preferred
validator (since it's a page which requires login you'll have to upload
or paste the source). The page should validate successfully. Test when
placing holds on:
- a single item
- multiple items
- items which you cannot place on hold
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.
Tested according to test plan and validated souce code with
W3C Validator.
Owen Leonard [Sun, 20 Oct 2013 17:39:06 +0000 (10:39 -0700)]
Bug 11057: Bootstrap theme: fix display of active language in language chooser
There is a problem with how an active single language is styled, making
the alignment look incorrect. This patch amends the markup and the CSS
to make it look better. This patch does not address the position of the
language menu.
To test you must have at least one additional single translation
installed. Apply the patch, refresh the page, and confirm that the
display of the active language is improved.
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Improved view of current language. No koha-qa errors.
1) Downloaded a couple of xx-YY-bootstrap.po from
http://translate.koha-community.org/projects/314
2) Installed sample languages
3) Enable them and opaclanguagesdisplay
4) Checked correct alignment of selected language
Signed-off-by: Ed Veal <ed.veal@bywatersolutions.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
The selected language has not a padding, contrary to others.
Not perfect but it fixes the descripted behavior.
Martin Renvoize [Thu, 5 Sep 2013 14:49:55 +0000 (15:49 +0100)]
Bug 10829: fix browsing across pages breaks for OPAC bibs-linked-to-authority searches
Test plan
1. Do an authority search in the OPAC (A search for 'an:49' perhaps..
that should bring up results for authority number 49, hopefully there
are multiple pages of results, if not try again and find one that
gives you multiple pages of results)
2. Click the final result on any page of results (there must be more
results to follow for this to work)
3. Try clicking the 'Next' button in the 'Browse results' pane on
the right of the page.
4. It should take you to the next result in the search list (i.e
the first result from the next page of results) - Double check it is
correct, we found that it was often not.
5. Apply patch and run same process, this time you should get the
correct paging.
Signed-off-by: Koha Team Amu <koha.aixmarseille@gmail.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
I reproduce the issue and I confirm this patch fixes it!
Fridolyn SOMERS [Mon, 3 Dec 2012 13:08:07 +0000 (14:08 +0100)]
Bug 9192: let UNIMARC_sync_date_created_with_marc_biblio.pl create fields if needed
Prior to this patch, this script would only modify the date
created and modified fields; this patch changes the behavior
so that the fields can be created if they're not already present
in the record.
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz> Signed-off-by: Mason James <mtj@kohaaloha.com>
Bug 9192: UNIMARC_sync_date_created_with_marc_biblio.pl field creation (followup)
Add tests :
if field < 10 it must not have a subfield
if field > 9 is must have a subfield
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Both patches applied, no koha-qa errors.
Colin Campbell [Thu, 10 Oct 2013 13:09:19 +0000 (14:09 +0100)]
Bug 11034: remove incorrect return for BiblioAutoLink
BiblioAutoLink was being called expecting an array to
be returned. The subroutine in fact returns a scalar
As the variables that the return was stored in, went
out of scope immediately. It is probably better to be
explicit that we are throwing the return away by not
going through the motions of storing it in a variable
To test:
[1] Add or save a bib record using the cataloging editor.
[2] Verify that no regressions have occurred
[3] (optional) If BiblioAddsAuthorities is on, verify that
relevant authority records are linked to from the bib.
biblioitems.serial is not a foreign key; it's a Boolean indicating
whether the biblio record is a serial, which in turn influences how
serial items are displayed.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Joy Nelson <joy@bywatersolutions.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Kyle M Hall [Mon, 21 Oct 2013 20:46:26 +0000 (16:46 -0400)]
Bug 10493: (follow-up) fix various QA issues
Adds a new Template Toolkit filter EncodeUTF8 to encode strings
to utf8 for correct display of diactritics.
Adds the new JavaScript function removeFocus() to staff-global.js
Use this function to remove the focus from any element for
repeated scanning actions on errors so the librarian doesn't
continue scanning and miss the error.
Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
This works as described - you have to actually look at the
error and pick what you want to do or confirm it. I think
maybe tying the action to a shortcut (c = confirm or similar)
would be nice, so you can get away with only using the keyboard.
Owen Leonard [Thu, 1 Aug 2013 19:37:46 +0000 (15:37 -0400)]
Bug 10493: (follow-up) fix minor template issues
This template followup fixes some minor template errors and adds some
style to dialog buttons matching those seen on the circulation page.
To test, submit barcodes to trigger the various error conditions:
Non-existant barcode, over the renew limit, restricted patron, item on
hold. Override and Ignore buttons should be properly styled and
functional.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Kyle M Hall [Wed, 19 Jun 2013 09:28:01 +0000 (05:28 -0400)]
Bug 10493: Add renewal script
This patch adds a renewal tool that functions similar to the returns where a
librarian can continuously scan items for renewal. This script blocks
renewals that are impossible, and allow the same renewal overrides
as circulation.pl
Test plan:
1) Apply the patches for bug 8798
2) Apply this patch
3) Browse to /cgi-bin/koha/circ/renew.pl
4) Enter an invalid barcode, you should get an error message
5) Enter a valid, but not checked out barcode, you should get an error
message.
6) Enter a valid barcode that is checkout out and should be renewable,
you should get a success message.
7) Enable AllowRenewalLimitOverride
8) Enter a barcode for an item that has been renewed too many times
9) You should get a warning which you can override.
10) Disable AllowRenewalLimitOverride
11) Repeat steap 8
12) You should get a blocking error message
11) Enter a barcode for an item with unfilled holds on it,
you should get an overridable warning
Signed-off-by: Owen Leonard <oleonard@myacpl.org> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Passes all tests and QA script, some issues have been
addressed in follow-ups.
Jonathan Druart [Wed, 2 Oct 2013 07:58:14 +0000 (09:58 +0200)]
Bug 5336: (follow-up) multiple QA fixes
- add a message if the search returns no result
- keep selected values if the search returns no result (for fund and
order status)
- remove plurals in order status
- move the order status column in the search results table
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
New tests also pass:
t/db_dependent/Acquisition/close_reopen_basket.t
1) Database update - I end up with too many partials. My test cases:
- New orders, basket still open
OK Expected: new, Result: partial
- Ordered orders, basket and basketgroup closed
OK Expected: ordered, Result: partial
- Partially received orders
OK Expected: partial/complete, Result: partial/complete
- Received orders
OK: Expected: complete, Result: complete
- Cancelled orders
* cancelled from open basket before order
OK: Expected: cancelled, Result: cancelled
* cancelled from closed basket before receive
OK: Expected: cancelled, Result: cancelled
All aqorders where updated with the correct status.
I have saved my 'pre-updatedatabase' and can repeat the test anytime you have a follow up.
2) Testing search functionality
a) Order search - result list
Order search shows a new column, I think it would be a bit better if the
status there was singular instead of plural - new order, partially received
order etc. - maybe we could even leave out the 'order'? (minor)
The column seems a bit lost in the middle, not sure where it would make more
sense (just saying)
b) Order search - advanced search form
The advanced search form now shows a new filter: Order status
All new status can be searched.
- an "empty" search will find all but cancelled orders
- searching for cancelled orders works correctly and shows results
Search works correctly, no regressions found.
If no result is found a message is shown.
All search input is kept, so you can modify your search terms easily.
3) Testing setting of status after applying the patch
a) Create a new order with 2 items - status is new. OK
b) Close the basket - status is ordered. OK
c) Receive both items - status is completed. OK
d) Undo receipt - status is ordered. OK
e) Receive only 1 item - order is split up into 2 orders:
- status is partial OK
- status is completed OK
f) Undo receipt of received item - order is combined into 1 again
- status is ordered OK!
g) Receive only 1 item again - status ordered/partial.
h) Delete order.
- status is completed
- status is cancelled
OK!
i) Undo receipt of 1 item again.
Refresh page.
This results in the following behaviour, that has been reported as
bug 10984. After refreshing the page follow message is shown:
Cannot cancel receipt. Possible reasons :
- The order line you trying to cancel was created from a partial
receipt of another order line which is already received.
Try to cancel this one first and retry.
- The order line you trying to cancel was created from a partial
receipt of another order line which has been deleted.
Cancellation is not possible.
BUT: The receipt is undone, but you are left with a
line with 2 items, a cancellation date and the status ordered.
Because of the cancellation date the order is not visible in pending orders.
The status is correct - so I feel this should not stop this patch.
Jonathan Druart [Tue, 1 Oct 2013 12:06:11 +0000 (14:06 +0200)]
Bug 5336: (follow-up) fix case where partial status was not correctly set
An order (o1) is considered as partially received if (at least) another
order (o2) has the o1.ordernumber == o2.parent_ordernumber and the
quantity received is < to the quantity (to receive) and o1.ordernumber
!= o1.parent_ordernumber.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Jonathan Druart [Wed, 18 Jan 2012 15:05:40 +0000 (16:05 +0100)]
Bug 5336: Order search (search and UI enhancements)
You can now search orders by
- order status
- fund
The patch series also adds a new field, aqorders.orderstatus, which can
contain following values:
new
ordered
partial (for partially received orders)
complete
cancelled
To test: Search and check if results are consistent in histsearch.pl Signed-off-by: Cedric Vita <cedric.vita@dracenie.com> Signed-off-by: Pierre Angot <tredok.pierre@gmail.com> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Comments on last patch.
Note: status are no longer numeric, but strings now.
Bug 10403: Show inactive but unlocked budgets on orderreceive
I have heard from several libraries that at the end of the budget
year even though a budget is inactive for new acquisitions, an
inactive budget should remain modifiable until the books are
closed and the budget is marked inactive. This patch makes it so
that all budgets that are unlocked are available on the order
receipt page (only).
Right now there is no way to change the budget or fund when receiving an
item, which is annoying, particularly at the end of the fiscal year when
every item not already received has to be switched to the following
year's budget. This patch adds the ability to change the budget and fund
when receiving.
To test:
1) Apply patch.
2) Create an order for a vendor, choosing a fund to use for that order.
3) Receive the order, leaving the fund unchanged. Make sure the fund
did not change.
4) Create another order for a vendor, choosing a fund to use for that
order.
5) Receive the order, this time changing the fund. Make sure the fund
is changed.
6) Run the unit test:
> prove t/db_dependent/Acquisition.t
7) Sign off.
(Notes: this patch depends on the Acquisitions.t unit test improvements
in bug 10274; the seemingly-unrelated change in SQLHelper quiets an
irritating warning caused by the NewOrder call in ModReceiveOrder)
Owen Leonard [Mon, 21 Oct 2013 21:54:41 +0000 (14:54 -0700)]
Bug 11097 - Hide duplicate patron button if logged in user lacks CAN_user_borrowers permission
The "duplicate" button on the circ/patron toolbar should not appear if
the logged in user lacks permission to edit patron records. This patch
wraps the button in the required logic.
To test, open a patron record in circulation or patrons and view as a
staff client user who both has and doesn't have CAN_user_borrowers
permission. The button should be shown and not shown accordingly.
Signed-off-by: Ed Veal <ed.veal@bywatersolutions.com> Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Kyle M Hall [Thu, 26 Sep 2013 16:22:39 +0000 (12:22 -0400)]
Bug 10954: ensure that payment receipt displays the payment made
When printing a payment receipt, the payment did not display, only the
total currently owed.
Test Plan:
1) Create a payment for a borrower
2) Print the payment with the "print" button to the right of the payment
3) Note the missing payment line
4) Apply this patch
5) Repeat step 2
6) Note the payment line now displays
Ketan Kulkarni [Wed, 9 Oct 2013 13:27:03 +0000 (18:57 +0530)]
Bug 10785: change 'Dropbox' labels to 'Book drop'
Signed-off-by: KohaCon13 demo <brendan@bywatersolutions.com> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Checked the book drop still works, also grepped for
other appearances of the term.
Fixed capitalization.
Bug 10982: fix link to IFLA UNIMARC documentation in Cataloging
In Cataloging, a ? opens a documentation page on IFLA web site. The current
implementation rely on the tag tab to select the appropriate IFLA
documentation web page. It doesn't work when UNIMARC tag are displayed
in a tab which digit is different for tag first digit.
To reproduce the bug:
- Modify default framework, put all 200 subfields in tab 1
- Add a new biblio record
- Click on ? on 200 field
=> Wrong IFLA page is opened
Apply the patch and retry...
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Simple fix. Work as described. No koha-qa errors
Test on UNIMARC install
1) Switched all 2xx tags to tab 1
2) press any 2xx ?, wrong page
3) patch applied
4) links are now correct
Signed-off-by: Paul Poulain <paul.poulain@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Galen Charlton [Tue, 22 Oct 2013 05:15:07 +0000 (05:15 +0000)]
Bug 11002: (follow-up) use jQuery validator plugin
This patch uses the jQuery validator plugin introduced in
bug 6209 for two reasons:
[1] To supply validation for non-HTML5-compatible web
browsers.
[2] To force use of the validator plugin rather than
built-in validation for HTML5 browsers, which means
that our translations will be used.
Note that a version of the function to prevent double form submission
is used as the submit handler invoked when the validator plugin
is satisified that the form is complete.
To test:
[1] Same as the main test plan, but note that the validation
messages now appear inline rather than as pop-ups.
[2] Verify that the double form submission prevention still
works.
Jonathan Druart [Mon, 7 Oct 2013 08:41:24 +0000 (10:41 +0200)]
Bug 11002: simplify patron mandatory field validation
When the pref BorrowerMandatoryField is set and a librarian forget to
fill fields, a js popup appears with the following message:
"The following fields are mandatory: surname, etc."
The fields are not translatable.
Test plan:
- fill the pref BorrowerMandatoryField with something like:
title|zipcode|surname|cardnumber|branchcode
- go the members/memberentry.pl?op=add page
- verify you are not allowed to add the patron without filling all
mandatory fields.
Jonathan Druart [Mon, 7 Oct 2013 09:41:22 +0000 (11:41 +0200)]
Bug 11003: fix JS error on the staff cart page
A call to yuiToolbar breaks JavaScript code on the cart page.
Test plan:
- add some biblio to your cart
- open the cart
- a js error occurs (see it with firebug):
ReferenceError: yuiToolbar is not defined
- apply the patch
- retry and verify the error does not appear anymore.
Signed-off-by: Paola Rossi <paola.rossi@cineca.it> Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
Works as described.
Colin Campbell [Wed, 16 Oct 2013 09:09:12 +0000 (10:09 +0100)]
Bug 11058: fix compilation warning in C4::Record
A routine declares two lexical variables named $stylesheet.
This patch renames the second to keep code clearer and
avoid propagating compile time warnings
Signed-off-by: Mark Tompsett <mtompset@hotmail.com> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pl, works as advertised!
Given how easy it is to accidentally receive items from one invoice on
multiple invoices, the ability to merge invoices can be quite handy.
This patch adds that ability to Koha's Acquisitions module.
To test:
1) Apply patch.
2) Run unit test:
> prove t/db_dependent/Acquisition/Invoices.t
3) Create two invoices from the same vendor for merging, and receive at
least one order on each.
4) Do a search on the Invoices page that brings up both the invoices you
created.
5) Check the boxes next to the two invoices.
6) Click "Merge selected invoices."
7) Choose which invoice you want to keep (the default will be the first).
8) Click "Merge."
9) Confirm that the resulting invoice has all the orders you received
listed on it.
10) Sign off.
Signed-off-by: Paola Rossi <paola.rossi@cineca.it> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Merged several invoices sucessfully - with and without received
orders, open and closed. Works nicely.
Jonathan Druart [Fri, 26 Apr 2013 07:30:23 +0000 (09:30 +0200)]
Bug 10724: Pending orders: add basket group name in filters
This patch adds a new filter "basket group name" for pending orders
searches.
Test plan:
Try different filters and check that results are consistent.
Try to filter by basket group name.
Signed-off-by: Pierre Angot <tredok.pierre@gmail.com> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Applied on top of patches for bug 10723.
Passes all tests and QA script.
Note: It's a bit irritating that the basket name is not shown
in the list of pending orders, so there is no way to check the
results are correct without checking from another page.
Jonathan Druart [Wed, 14 Aug 2013 08:43:07 +0000 (10:43 +0200)]
Bug 10723: Merge GetPendingOrders and SearchOrders routines
In the C4::Acquisition module, 2 routines do the same work. This patch
merges these 2 routines.
Test plan:
test the acqui/orderreceive.pl, acqui/uncertainprice.pl
and serials/acqui-search-result.pl, acqui/parcel.pl scripts.
Note: on acqui/parcel the basket filter is a search on basket name (was
on basket id, which was not relevant).
Signed-off-by: Pierre Angot <tredok.pierre@gmail.com> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Passes koha-qa.pm, no adverse bahaviors noted. All sub calls updated.
Owen Leonard [Mon, 21 Oct 2013 00:52:06 +0000 (17:52 -0700)]
Bug 11060: (Bootstrap theme) fix incorrect rendering for right to left languages
The right-to-left CSS copied over from the prog theme does is not
relevant to the redesigned templates. This patch updates the OPAC's
right-to-left CSS and adds a couple of other right-to-left CSS files to
override the default Bootstrap and jQueryUI CSS.
To my eye this looks correct, but I don't speak Arabic.
To test you must have a set of right-to-left templates, like
ar-Arab, installed for the Bootstrap theme. At this time it is necessary
to download the po file to your misc/translator/po directory. Again
using ar-Arab as an example:
Open the OPAC, switch to that right-to-left language, and reload to
refresh the changed CSS. Confirm that it looks like right-to-left is
working.
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
As far as I can tell it looks very well.
No koha-qa errors
1) Downloaded arabic bootstrap PO
2) Installed ar-Arab language, enabled on OPAC
3) Tested as anonymous user and logged in one, all pages look
well: cart, advanced search, user tabs.
I sent an email to Karam Qubsi asking his opinion.
But for me it's ok
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
I was testing this at the same time as Bernardo, so I will Pass QA on
it, instead of signing it off
Galen Charlton [Sun, 20 Oct 2013 20:55:06 +0000 (20:55 +0000)]
bug 10016: munge history in SCO to discourage resubmitting user login
This patch uses history.replaceState (introduced in HTML5) to manipulate
the browser history to encourage returning to the SCO patron barcode
form if the back button is used.
Note that a side effect of this patch is that if the user uses
the help link, they will be prompted to enter their barcode
again. It may be better to put the help inline with the rest
of the SCO forms.
To test:
[1] Start a web-based self-check session.
[2] Enter a patron barcode.
[3] Allow the self-check session to time out.
[4] Use the back button. You should get the patron barcode
entry form; you should not be prompted to resubmit form input.
[5] Enter a patron barcode, perform some transactions, then
use the finish button.
[6] Next, use the back button. You should get the patron barcode
entry form.
Galen Charlton [Sun, 20 Oct 2013 17:13:22 +0000 (17:13 +0000)]
Bug 10016: force zero browser-side caching of SCO pages
This patch makes the web-based self-check module pages
specify that no browser (or proxy caching) occur at all.
This prevents a security issue where letting the SCO session time out,
then hitting the back button allowed one to view the previous
patron's session.
This patch adds an optional fifth parameter to output_with_http_headers(),
and output_html_with_http_headers(), a hashref for miscellaneous
options. One key is defined at the moment: force_no_caching, which if
if present and set to a true value, sets HTTP headers to specify no
browser caching of the page at all.
To test:
[1] Start a web-based self-check session and optionally perform
some transactions.
[2] Allow the session to time out (it may be helpful to set
SelfCheckTimeout to a low value such as 10 seconds).
[3] Hit the back button. You should not see the previous patron's
self-check session.
[4] Verify that prove -v t/Output.t passes.
D Ruth Bavousett [Sun, 20 Oct 2013 17:23:44 +0000 (12:23 -0500)]
Bug 8070: improve MARC preview display for RTL languages
The Right Answer, according to the cataloger I spoke to,
is to make it LTR when doing "raw" MARC display.
To test:
1) Install this patch, then run misc/translator/translate install ar-Arab
2) search for a bib in the staff client.
3) Switch to Arabic, and click the link to the left of "MARC Preview"
Signed-off-by: Owen Leonard <oleonard@myacpl.org>
At first glace I was unsure about hard-coding this style in the
template. Further testing indicated that this is required as the
showmarc template doesn't inherit any of the CSS from the containing
page.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Mathieu Saby [Sun, 17 Mar 2013 13:54:12 +0000 (14:54 +0100)]
Bug 9830: Fix some indexes in UNIMARC item indexing
With this combination of sysprefs, and a UNIMARC configuration, it was
impossible to search on location, barcode and ccode indexes :
QueryWeightFields is activated
QueryAutoTruncate only if * is added
But in UNIMARC, location, barcode and ccode (995 $e,$f,h) were indexed
only as "words". They need to be indexed also as "phrase".
Additionnaly, in UNIMARC, information about damaged and withdrawn status
of items is not indexed, while it is done in MARC21.
This patch
- add 2 new indexes for 995$1 (damaged) and 995$3 (withdrawn)
- index location, barcode and ccode as "phrase" as well as "words"
Indexing of items in UNIMARC could be improved later. So this patch also
add comments explaining the origin of Koha 995, I think it could be
useful for further changes.
To test, on a UNIMARC configuration :
A. indexed with GRS-1
1) Set sysprefs QueryWeightFields as "activated" and QueryAutoTruncate
as "only if * is added"
2) Select location index in advanced search and search for a value
existing in your records in 995$e => 0 results
3) Apply patch
4) Rebuild zebra
5) Select location index in advanced search and search for a value
existing in your records in 995$e => x results
6) Mark an item as withdrawn; search "withdrawn:1" => x results, and
among them the biblio to which the item is attached
7) Mark an item as damaged ; search "damaged:1" => x results, and among
them the biblio to which the item is attached
B. indexed with DOM
Do the same operations
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Work as described. No koha-qa errors
Test
Apply the patch
Begin with GRS-1
Full reindex
Search by location, no results
cp files biblio-*-indexdefs.xml and record.abs to destination on etc/zebra
Full reindex
Search by location, got results
Switch to DOM
reset files
Full reindex
Search by location, no results
cp files
Full reindex
Search by location, results !
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Recent changes to LDAP broke auth_by_bind in many situations. This bug
resets the behaviour to what it used to be, however also allows the new
behaviour by adding the 'anonymous_bind' parameter to the LDAP config.
Testing:
1) Find an LDAP configuration that was broken recently that uses
auth_by_bind
2) Apply this patch
3) See if it works again.
Additionally, testing the original path in the case of 'anonymous_bind'
being set should probably be done too, but I have no idea about the LDAP
server config for that.
Fridolyn SOMERS [Thu, 28 Feb 2013 17:21:18 +0000 (18:21 +0100)]
Bug 9739: more options for serialsUpdate.pl
Script serialsUpdate.pl must be set in crontab to run daily.
It checks serials that are late, modifies status and adds a note.
This patch adds some options :
--note : defined note, a defaut one is used if not defined
--no-note : disable automatic note
--verbose : used to control output
This patch also implements confirmation option "-c" who whas present but not used.
Test plan :
Run script on a database with late serials :
- serialsUpdate.pl -h
=> you get help text in output
- serialsUpdate.pl --man
=> you get full help
- serialsUpdate.pl -v
=> you get output of changed serials but database has not changed
- serialsUpdate.pl -c -v
=> you get output of changed serials and database has changed, with note "Automatically set to late"
- serialsUpdate.pl -c
=> database has changed without anything in output
- serialsUpdate.pl -c --note "LATE"
=> database has changed, with note "LATE"
- serialsUpdate.pl -c --no-note
=> database has changed with no note
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Works as described. No koha-qa errors.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
New options behave as described.
There are some existing problems with this script that I have noted
directly on the bug report.
Katrin Fischer [Sun, 29 Sep 2013 16:45:06 +0000 (18:45 +0200)]
Bug 10969: Fix sample itemtypes for translated installers
To test:
- Install the specific language and run the web installer.
Or:
- Truncate or delete all from the itemtypes table in your database.
- Try to run the complete SQL in the itemtypes sample files.
Verify the table is filled correctly.
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Works for all languages. No koha-qa errors.
Tested removing itemtypes, then loading new file,
and checking on staff.
For de, es, nb, and pl no problems. For ru I disable foreign
keys constraint before TRUNCATE and reenable it afterwards.
Signed-off-by: Paul Poulain <paul.poulain@biblibre.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Bug 8368: fix email lists from OPAC when using non-English templates
This patch ports to list sending by email, the technique used in sending
cart, i.e. (1) format email in HTML, and (2) transform it into Text with
TT filter html2text.
Signed-off-by: Owen Leonard <oleonard@myacpl.org> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Fixes a very annoying translation bug successfully.
Also passes all tests and QA script.
To test:
- Add a few records to a shelf, ideally use some with diacritics.
- Send shelf from English templates, verify email is ok
- Send shelf from translated templates (de-DE or similar) - verify
email content is broken.
- Apply patches, update po files and reinstall the language.
- Send shelf from English templates again, verify there is no
regression.
- Send shelf form translates templates - verify this email is now
also working correctly.
Patch also changes the name of the file attachement from shelf.iso2709
to list.iso2709.
Jonathan Druart [Mon, 9 Sep 2013 12:12:05 +0000 (14:12 +0200)]
Bug 10843: fix crash that can occur when confirming hold if ReservesMaxPickUpDelay is undefined
Test plan:
1) set an empty string for the ReservesMaxPickUpDelay pref
2) place a hold on an item
3) check in the item
4) click on "Print and confirm"
5) an error occurs
> The 'days' parameter (undef) to DateTime::Duration::new was an 'undef'
6) apply the patch
7) repeat steps 1 to 4
8) the error does not occur anymore.
Signed-off-by: Srdjan <srdjan@catalyst.net.nz> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
An empty string didn't do it for me, I had to set the
variable for the systempreference to NULL. I am not sure
if this can happen when editing from the interface, but
this change should not have any ill side effects and it has
unit tests!
Owen Leonard [Wed, 11 Sep 2013 15:51:54 +0000 (11:51 -0400)]
Bug 10850 - give quick spine label template page its own ID
Out of three conditional cases for display of the <body> tag one lacked
a unique ID. This patch copies the unique ID from the other cases to the
one lacking.
This patch also corrects the case of an "onload" attribute which should
follow XHTML style rules.
To test, the patch for Bug 9618 must be applied. Set the
SpineLabelAutoPrint system preference to "[don't] automatically pop up a
print dialog." Submit a barcode on the quick spine label creator and
view source. The <body> tag should have an ID attribute.
Signed-off-by: Paola Rossi <paola.rossi@cineca.it> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Jonathan Druart [Wed, 17 Jul 2013 13:02:43 +0000 (15:02 +0200)]
Bug 10038: fix doubled 'default' entry when creating a new authority type
Test plan:
- Create a new authority type
- Click MARC structure
- Verify the pull down has only 1 entry for Default
- Go on the authority type home (admin/authtypes.pl)
- Click on the "MARC structure" link for the default type
- Verify the pull down has only 1 entry for Default
This patch adds a sort (on the authtypecode) for these 2 lists.
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Comment: Works as described. No koha-qa errors.
When creating a new framework it would be better to have Default
on top, but one is way better than two :)
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Agreed, one is better than two :)
All tests and QA script pass.
Jonathan Druart [Mon, 23 Sep 2013 15:07:09 +0000 (17:07 +0200)]
Bug 10935: fix breakage of CSV acq exports when using a non-English translation
To reproduce:
- cd misc/translator
- ./translate update LANG
- ./translate install LANG
- go to the Koha mainpage and change the language.
- go to acqui/basketgroup.pl?booksellerid=XX and try to export a
basketgroup.
The headers is followed by the first basketgroup information. There is
no carriage return.
It looks like it is caused by a routine used by the translator script
(TmplTokenizer::string_canon).
To test this patch:
- apply it
- cd misc/translator
- ./translate -f update LANG
- translate headers in your po file
- ./translate -f install LANG
- go to acqui/basketgroup.pl?booksellerid=XX and try to export a
basketgroup.
- verify that the csv looks good now.
- same thing for basket.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Good idea and seems to work - just fixing a small glitch
with the first entry of the list in a follow-up.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz> Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Jonathan Druart [Fri, 13 Sep 2013 14:28:39 +0000 (16:28 +0200)]
Bug 10884: circulation table: fix odd behavior on checkboxes
Test plan:
Play with renew, check in and export checkboxes.
The expected behavior is: the renew and check in cb should work as radio
button, without affecting the export cb.
Signed-off-by: Owen Leonard <oleonard@myacpl.org> Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Template changes only.
Amended test plan:
* Create a CSV profile under Tools > CSV profiles
* Add CSV profile to system preference ExportWithCsvProfile
1) Check export checkbox
2) Check check in for same item
3) Switch to Renew
The export checkbox should not get unchecked when switching
between renew and check in.
Owen Leonard [Mon, 7 Oct 2013 19:12:03 +0000 (15:12 -0400)]
Bug 10309: (follow-up) Add changes from Bug 8712, "and" labels in advanced search
This patch adds the changes from Bug 8712 to the new theme. Some HTML
markup errors have also been corrected. From Bug 8712:
"The "and" label not only improves the accessibility but also gives the
user an idea that the operation between multiple search fields is a
boolean "AND" by default."
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
"and" looks well on advanced search. No errors
Owen Leonard [Mon, 7 Oct 2013 18:45:06 +0000 (14:45 -0400)]
Bug 10309: (follow-up) Update for Bug 10856, improve shelf browser
This patch updates this theme with the changes made in Bug 10856.
Because of the way JavaScript is loaded in the new theme, the JS in the
shelfbrowser include had to be moved to the opac-detail template.
Test plan from that bug:
- On a detail biblio page, click on a "Browse shelf" link.
- Play with the next and previous links.
- Deactivate Javascript (using NoScript for example) and check that you
get the same behavior (but the page is reloaded).
- Launch the unit tests: prove t/db_dependent/ShelfBrowser.t
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Works well. No errors
Disabled JavaScript on Firefox (about:config, javascript.enable -> false)
Browse shelf works by reloading.
Unit test report success.
Owen Leonard [Mon, 7 Oct 2013 14:39:19 +0000 (10:39 -0400)]
Bug 10309: (Follow-up) Applying changes from Bug 10349
This patch applies the changes from Bug 10349, Don't show empty
Descriptions/Title notes tabs in OPAC and staff.
"The descriptions/title notes tab appears on the detail page in both
staff client and OPAC even if there are no notes. This is probably a
relic of the pre-T:T days when it wasn't possible to use || in an IF.
This patch adds a check for the various variables which might trigger
the display of the tab.
To test, apply the patch and view records in the OPAC and staff client
which do and do not have title notes attached (whether that be in the
MARC record or in the biblio.notes column). In the OPAC Syndetics
content should also be tested if possible. The descriptions/title notes
tab should only appear if there is content."
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Works well. No errors
Tested on OPAC, records without notes don't show title notes tab.
Owen Leonard [Mon, 7 Oct 2013 14:33:03 +0000 (10:33 -0400)]
Bug 10309: (follow-up) Updates for Bug 6594, Schema.org structured data
This patch adds the changes from Bug 6594:
"To support schema.org processors, such as Google, Bing, and Yandex,
structure our data so that it has machine-readable attributes. This pass
declares the CreativeWork sub-types as well as Product for the main
bibliographic record details, and uses the Offer type for holdings
information per the W3C Schema Bib Extend community group discussions."
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Tested against linter.structured-data.org, microdata present
No errors.
Owen Leonard [Wed, 22 May 2013 13:37:31 +0000 (09:37 -0400)]
Bug 10309 - New OPAC theme based on Bootstrap
The goal of this theme is to provide a fully-responsive OPAC which
offers a high level of functionality across multiple devices with varied
viewport sizes. Its style is based on the CCSR theme, with elements of
the Bootstrap framework providing default styling of buttons, menus,
modals, etc.
The Bootstrap grid is used everywhere, but Bootstrap's default
responsive breakpoints have been expanded to allow for better
flexibility for our needs.
All non-translation-depended files are in the root directory of this new
theme:
css, images, itemtypeimg, js, less, and lib. Languages.pm has been
modified to ignore the new directories when parsing the theme language
directories.
This theme introduces the use of LESS (http://lesscss.org/) to build
CSS. Three LESS files can be found in the "less" directory: mixins.less,
opac.less, and responsive.less. These three files are compiled into one
CSS file for production: opac.css. "Base" theme styles are found in
opac.less. A few "mixins" (http://lesscss.org/#-mixins) are found in
mixins.less. Any CSS which is conditional on specific media queries is
found in responsive.less.
At the template level some general sturctural changes have been made.
For the most part JavaScript is now at the end of each template as is
recommended for performance reasons. JavaScript formerly in
doc-head-close.inc is now in opac-bottom.inc.
In order to be able to maintain this structure and accommodate
page-specific scripts at the same time the use of BLOCK and PROCESS are
added. By default opac-bottom.inc will PROCESS a "jsinclude" block:
[% PROCESS jsinclude %]
Each page template in the theme must contain this block, even if it is
empty:
[% BLOCK jsinclude %][% END %]
Pages which require that page-specific JavaScript be inserted can add it
to the jsinclude block and it will appear correctly at the bottom of the
rendered page.
The same is true for page-specific CSS. Each page contains a cssinclude
block:
[% BLOCK cssinclude %][% END %]
...which is processed in doc-head-close.inc:
[% PROCESS cssinclude %]
Using these methods helps us maintain a strict separation of CSS links
and blocks (at the top of each page) and JavaScript (at the bottom). A
few exceptions are made for some JavaScript which must be processed
sooner: respond.js (https://github.com/scottjehl/Respond, conditionally
applied to Internet Explorer versions < 9 to allow for layout
responsiveness), the _() function required for JS translatability, and
Modernizr (http://modernizr.com/, a script which detects browser
features and allows us to conditionally load JavaScript based on
available features--or lack thereof).
Another new JavaScript dependency in this theme is enquire.js
(http://wicky.nillia.ms/enquire.js/), which lets us trigger JavaScript
events based on viewport size.
I have made an effort to re-indent the templates in a sane way,
eliminating trailing spaces and tabs. However, I have not wrapped lines
at a specific line length. In order to improve template legibility I
have also tried to insert comments indicating the origin of closing tags
like <div> or template directives like [% END %]:
</div> <!-- / .container-fluid -->
[% END # / IF ( OpacBrowseResults && busc ) %]
TESTING
Proper testing of this theme is no easy task: Every template has been
touched. Each page should work reasonable well at a variety of screen
dimensions. Pages should be tested under many conditions which are
controlled by toggling OPAC system preferences on and off. A variety of
devices, platforms, and browsers should be tested.
This corrects a failing test and follows a recommendation
by the maintainer of DBIx::Class. This patch also
adds a couple new directories for t/00-testcritic.t to
check.
Galen Charlton [Mon, 14 Oct 2013 22:31:37 +0000 (22:31 +0000)]
Bug 10565: (follow-up) add new user permission for patron list management
This patch adds a new user permission for patron list management,
tools => manage_patron_lists.
This closes a security issue with the original patch series where
patron lists and their contents could be retrieved and modified
without requiring authentication of any sort.
projects / koha.git / log