Support Stripe payments with some new code and some rearranged code ...

... behind the open-ils.circ.money.payment method.

- Also add Business::Stripe as a CPAN module to the pre-reqs installers.
- If payment processor is Stripe, only show form if Javascript enabled.

Our implementation of payments via Stripe doesn't work without Javascript.
That's part of the point. Using Stripe really limits a site's worries
about PCI compliance because users' credit card number and security
codes are never transmitted to the [Evergreen] server at all.  That data
goes to Stripe instead (using Javascript) and from that we get back a
one-time token from Stripe to give to our server instead.

Thanks to Jason Boyer at the Indiana State Library for recognizing the
value of supporting approaches like that in Evergreen and for starting
the work.

Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com>
Conflicts:
Open-ILS/src/templates/opac/myopac/main_payment_form.tt2

Signed-off-by: Remington Steed <rjs7@calvin.edu>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

DB upgrade script for Stripe payments; seed data; minor template fixes

Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com>
Conflicts:
Open-ILS/src/templates/opac/myopac/main_payment_form.tt2

Signed-off-by: Remington Steed <rjs7@calvin.edu>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

Separate stripe js into stripe.tt2, add seed data for ou settings

Signed-off-by: Remington Steed <rjs7@calvin.edu>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

Rough draft of integrating Stripe support into the opac payment page.

Conflicts:
Open-ILS/src/templates/opac/myopac/main_payment_form.tt2
Open-ILS/src/templates/opac/parts/base.tt2

Signed-off-by: Remington Steed <rjs7@calvin.edu>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

Duplicate 0851 as 0862 for backport clarity

Rather than backport 0851 after 0852 has already been backported,
let's give 0851 a second number.  It probably wouldn't have bitten
anyone, but there's no harm in running this upgrade again, and I
don't think it hurts to be extra clear.

Signed-off-by: Dan Wells <dbw2@calvin.edu>

LP1223903 - Also bump the schema config

Signed-off-by: Ben Shum <bshum@biblio.org>

LP1223903 - Stamping upgrade script for new authority_record_entry indexes

Signed-off-by: Ben Shum <bshum@biblio.org>

LP1223903 - Add release notes for rewrite of marc_export.

Signed-off-by: Jason Stephenson <jstephenson@mvlc.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP1223903 - Rewrite marc_export.in in support-scripts.

This commit rewrites marc_export.in in the support-scripts  src directory.
It will still be transformed into marc_export at build time.  This change
replaces the current marc_export scipt with a faster, DBI based alternative.

The replacement support nearly all of the features of the original, except
for the progress output.  Much of this information is harder to gather
with the new design.

This is a squashed, rebased commit of previous work.

Signed-off-by: Jason Stephenson <jstephenson@mvlc.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP1223903 - Add indexes to authority.record_entry.

We want indexes on edit_date and create_date for authority.record_entry
so that date-based searching is faster for authority record export.  It
likely will prove useful for other features.

Signed-off-by: Jason Stephenson <jstephenson@mvlc.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP1223903 - Add from_bare_hash to Fieldmapper.pm.

Teach the Perl Fieldmapper.pm to build objects from a hashref in a
similar manner to how the JS Fieldmapper already can.  You might use
it if you retrieved a database row as a hashref, like so:

my $hashref = {};
my $class = Fieldmapper::class_for_hint('bre');
my $bre = $class->from_bare_hash($hashref);

We also modify the Fieldmapper's properties method to sort the array of
field names it returns by position.  This makes the return value a lot
more useful for on the fly object construction and other fancy tricks.

Signed-off-by: Jason Stephenson <jstephenson@mvlc.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP#1262722 Fix DB supersedes/deprecates pgtap test

Sort multi-value arrays before passing them off to pgtap for test
comparison. Otherwise, pgtap tests may fail with:

Failed test 5
        have: {AAAC,AAAB}
        want: {AAAB,AAAC}

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

Stamping upgrade script for supsersede/deprecate logic fixes

Signed-off-by: Mike Rylander <mrylander@gmail.com>

DB patch supersede/deprecate logic repairs; unit tests

* Fix some thinkos in the various DB functions for calculating
  supersedes/deprecation logic.

* Added a pgtap test

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Mike Rylander <mrylander@gmail.com>

Stamping upgrade for Granular Staff Initials settings

Signed-off-by: Mike Rylander <mrylander@gmail.com>

Add granular settings for requiring staff initials for notes

Up till now, there has only been one setting for making staff initials required
for all of the various note fields. Instead, let's break this up into separate
org unit settings for each type of note field, giving us more flexiblity in
making use of the append staff initials feature.

These three types of notes will now have settings to add initials for each:

- patron penalties/messages
- patron information notes
- copy (item) notes

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Mike Rylander <mrylander@gmail.com>

LP#1272074 Physical Characteristics Wizard for the MARC Editor (Part II)

These files should have been in the other commit with the above title,
but I lost them in a rebase effort.

Signed-off-by: Dan Wells <dbw2@calvin.edu>

Stamping 0858: Fixed field enhancements

Also, move test file to a possibly better location.

Signed-off-by: Dan Wells <dbw2@calvin.edu>

LP#1272074 Release notes (with pictures!) of fixed fields MARC editor enhancements

Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

LP#1272074 Fix faulty physical characteristics seed data

For 007 (00/a=f), "subfield" e (Level of contraction) starts at position
5, not 4.

Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

LP#1272074 Physical Characteristics Wizard for the MARC Editor

Right-click in an 007 field to find a new entry in the context menu that
launches this wizard.  The prompts and suggested values come from data
already found in the database.

Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

LP#1272074 Context menus suggesting values for marc fixed field editor

Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com>
Signed-off-by: Dan Wells <dbw2@calvin.edu>

LP#1187035 Remove OpenILS::Utils::Editor part 4.

Make sure all CStoreEditors in ClosedDates run updates in transactions,
roll back transactions on error, and commit the transactions on success.

Added a live test for closed data updating.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

LP#1187035 Remove OpenILS::Utils::Editor part 3.

Remove obsolete lib/OpenILS/Utils/Editor.pm from MANIFEST.

Signed-off-by: Jeff Godin <jgodin@tadl.org>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

LP#1187035 Remove OpenILS::Utils::Editor part 2.

* Address 2 places where we were calling $e->request with a
 CStoreEditor for a storage method.  Since CStoreEditor assumes all
 calls go to open-ils.cstore, these would have failed.

* While we're in a cleaning mood, remove some obsolete commented code.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

LP#1187035 Remove OpenILS::Utils::Editor

OpenILS::Utils::Editor is obsolete, yet still referenced in a number
of places. Let's remove it.

Signed-off-by: Jeff Godin <jgodin@tadl.org>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

LP#1272575 Use getElementById over querySelector

Favor document.getElementById over document.querySelector for simple
by-id element referencing.

Signed-off-by: Jason Etheridge <jason@esilibrary.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

LP#1184885 ACQ: i18n for search abbreviations

Allow translating the Acq Search dropdown class name abbreviations.

Signed-off-by: Pasi Kallinen <pasi.kallinen@pttk.fi>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

LP#1187029 release notes typo repair for ingest cleanup

Signed-off-by: Bill Erickson <berick@esilibrary.com>

LP#1187029 Remove unused open-ils.ingest service

Remove open-ils.ingest service and related scripts and code.

In addition to removing OpenILS::Application::Ingest, this commit
removes code that was commented out at the time of open-ils.ingest's
replacement with in-db ingest, example configuration elements, a
use_ok test, and the following obsolete scripts which made use of or
were otherwise related to open-ils.ingest:

* Open-ILS/src/extras/import/direct_ingest.pl
* Open-ILS/src/extras/import/generate-srfsh-indexer.pl
* Open-ILS/src/extras/import/importer.sh

The included release notes recommend removal of the open-ils.ingest service
from configuration files when upgrading, and suggest removal of Ingest.pm from
its location in @INC.

Signed-off-by: Jeff Godin <jgodin@tadl.org>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

LP#1187034 Remove obsolete collections code

Remove obsolete collections code

Specifically:

open-ils.collections.users_owing_money.retrieve
open-ils.storage.money.collections.users_owing_money

Signed-off-by: Jeff Godin <jgodin@tadl.org>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

LP1125471 - Use label instead of id in recording in-house use prompt

When recording multiple in-house use with the library settings for threshold
prompts, the prompt would bring up the ID as the variable instead of the name
of the non-cataloged material. This caused confusion among staff members who
might see a message like: "Are you sure you want to mark 9 as having been used
21 times?" (9 being the id, not the name of the non-cataloged material)

This commit leverages existing code which had been commented out and under
the ID to use the label name for the non-cataloged material instead. Also,
minor bugfix with a missing semi-colon at the end of the function being used.

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

Stamping 0857 for located uri visibility changes

Stamping 0857 for located uri visibility changes, also adjusting
upgrade script to use ARRAY_AGG over ARRAY_ACCUM (which recently
landed in master).

LP 1271630 Allow Located URIs to supply copy-like visibility to bibs
https://bugs.launchpad.net/evergreen/+bug/1271630

LP 874296  Replace ARRAY_ACCUM with ARRAY_AGG (and STRING_AGG)
https://bugs.launchpad.net/evergreen/+bug/874296

Signed-off-by: Jeff Godin <jgodin@tadl.org>

Release notes

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

Teach evergreen.located_uris() about the new URI visiblity flag

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

Upgrade script

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

Global flag for setting Located URI behavioral mode

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

Check behavioral setting for Located URIs

When the opac.located_uri.act_as_copy Global Flag is enabled, treat
Located URIs like copies for visibility testing purposes.  We use
actor.org_unit_full_path instead of actor.org_unit_descendants so
that we do not lose the desirable "licensed here" modeling that
Located URIs provide.

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

evergreen.located_uris() returns a rank, and we should use it

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

Vandelay: increase queued item creation timeout

Allow up to 10 minutes to create each queued vandelay record to avoid
timeouts in item creation caused by pathologicaly long-running record
match set queries.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Ben Shum <bshum@biblio.org>

Keep standard INNER joins above possible OUTERs

We always join metabib.record_attr, and we always use an INNER join
to the core table, so keep that join first.  Likewise with the join
to biblio.record_entry.  Dynamic joins may be OUTER, though, and while
join reordering in Postgres is smart, let's not give it the chance to
fail.

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Ben Shum <bshum@biblio.org>

Assume that future checkins are junk input

This can happen with SIP2 clients with badly set clocks.

Signed-off-by: Thomas Berezansky <tsbere@mvlc.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

Change "Alias" to "Holds Alias" in Patron Search.

This same field is referred to as "OPAC/Staff Client Holds Alias" in
the Register Patron interface.  This change is for consistency in labeling.

Signed-off-by: Chris Sharp <csharp@georgialibraries.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

Release notes for "Update Expire Date" button

Add release notes for "Update Expire Date" button in user editor.

Signed-off-by: Jeff Godin <jgodin@tadl.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

Add button to update expire date field

When updating/renewing a patron (expired or not), it would be
convenient to have a means of populating the expire date field with
a value of perm_interval + today. Currently, staff are required to
manually enter a date, or could toggle the profile from one profile
then back to the desired profile, but this is not ideal.

This commit adds a new button next to the expire date, which when
activated will re-calculate the date based on the current profile's
perm_interval and today's date.

This is similar to how the expire date is populated when creating a
new user, or when changing the profile group. A button is used here
so that the updating of the expire date is an intentional process,
not one that happens upon any edit.

Signed-off-by: Jeff Godin <jgodin@tadl.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP1046026 - Fix duplicate rows in holds pull list

If there are multiple CAPTURE-blocking penalties we could see dupes on the
pull list.  This avoids that.

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Ben Shum <bshum@biblio.org>

Prevent odd TPAC base URL redirect

As it stands, in a default install, if you browse to

http://yourserver.com/eg/opac

you are redirected to a login page, and if you login, you then get
to the 404 page. This behavior is caused by the fact that *any*
URLs which do not match a few select cases are assumed to be
"secure" resources, when in fact they might not even exist.

This commit catches any attempt to browse this 'base' and simply
redirects to the "home" page instead.

This addresses the most obvious and likely 404 of this type, but
we might consider catching some other cases (e.g. eg/opac/record/).
In addition (or as an alternative), we could also create a function
call which simply checks the validity of a URL before we ask the
user to login to see it.

Signed-off-by: Dan Wells <dbw2@calvin.edu>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

Fix some javascript errors, trying to access nonexistent widgets

Signed-off-by: Pasi Kallinen <pasi.kallinen@pttk.fi>
Signed-off-by: Melissa Ceraso <mceraso@biblio.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

Stamping upgrade for STRING_AGG fix for metabib.staged_browse()

Signed-off-by: Ben Shum <bshum@biblio.org>

STRING_AGG() fix for metabib.staged_browse()

We need to supply TEXT values to STRING_AGG(), so cast the input INTEGER
accordingly.

Also, I LIKE ALL-CAPS.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

Release notes for WCAG compliance for the TPAC

Super basic, but gives a pointer to the WCAG overview for
more information.

Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 add expressive facet link titles

Let the user know exactly what clicking on a facet will do.

Thanks to Dan Scott for the suggestion.

* Note that accessibility validators may dislike the repetitive titles,
  but clearly adding them provides more information (for everyone).
  Plus, as facets pile up, there's no reasonable way to differentiate
  between all titles, anyway, without title attributes becoming too
  long and unreadable.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

Remove useless image-based fines border

The images ended up looking like displaced blocks, but
aren't necessary anyway. If one really wants a border,
use CSS.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

Remove extraneous ] in holds TPAC page

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Bill Erickson <berick@esilibrary.com>

LP#1268636 Add helpful empty search term message

When a search is submitted with no query value, instead of silently
redirecting the user the home page, direct the user to the results page
and show a helpful message (within the low-hits template).

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 improve fines warning color contrast

Use the text_attention CSS color instead of the text_badnews for the my
account fines label, since the badnews color does not contrast highly
engouh with the dark green background for WCAG compliance.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 remove a few invalid attrs / unused tags

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 additional myopac nav headers

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 TPAC avoid <b> tags in pref lib msg

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 TPAC more nav headers for search results

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 TPAC initial navigation headers

Added a screen-reader only CSS class ("sr-only") for adding content
which should only be read by screen readers and is otherwise invisible.

Added hidden generic <h1> and page-specific <h2> entries to improve
web accessibility navigation.

Additional content-specific headers (h3 and down) are recommended for
content grouping, but not included in this commit.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 My account various repairs

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 avoid empty <ul>'s in browse

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 advanced search repairs

* avoid dupe IDs
* removes extraneous <div>'s
* clean up attrs w/ preceding spaces

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 alt tags and cleanup continued

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 more expressive <a> titles in search results

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 HTML validation repairs

remove invalid span@nowrap attribute

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 additional label repairs

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 HTML table attributes validation repairs

Replace old-school inline cellpadding, cellspacing, border attributes
with CSS.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 facets get more specific more/less titles

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

LP#1268636 home page label / alt repairs

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

Include calculated 0-prox locations in the stalling interval for op capture

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Ben Shum <bshum@biblio.org>

Use copy owning library as the context for finding best hold sort order setting

In the olden days before best hold sort order and proximity adjustment, there was
no need to consider any context org other than "here", because the choices were,
essentially, FIFO or Traditional applied within the hard-boundary range.  However,
now we need to decided whose rule to follow, and the most correct arbiter of that
is the owning library of the call number of the copy.  We avoid copy circ lib
because that can change with floating copies.

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Ben Shum <bshum@biblio.org>

Fix copy_info variables one last time for library_name_link purposes

In cd056eba75038c6a1fabd2a83ba97d81a444bc75, I fixed record by inadvertently
passing around the wrong variables. In fact, we do not need the value set
there, and we broke results view as a result of the swap around.

Put things right what once went wrong.

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

Address some remaining hardcoded install prefixes

Ran a clean install of Evergreen on Ubuntu Precise and most things were
more or less working using simple "./configure; make; make install" with
the major exception of the TPAC. It turned out that there is a simple
variable that can be tweaked in the Apache config; a better way to
handle this might be to preprocess the source file at make time, but for
now at least we set the variable.

In addition, clean up two remaining scripts that have hardcoded paths.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP1155769: Items repeated in tpac display, others not visible

This was an issue with the query and ordering by a column that was
sometimes null (copy_number).  Added barcode to be a sort rather than
replacing copy_number, per bshum's suggestion.  Not replacing allows
libraries that use copy_number to continue to use them as well.

Signed-off-by: Kyle Tomita <tomitakyle@gmail.com>
Signed-off-by: Ben Shum <bshum@biblio.org>

Sign off on upgrade script for (STRING|ARRAY)_AGG

Signed-off-by: Dan Scott <dscott@laurentian.ca>

Tweak STRING_AGG() arguments (expects TEXT or BYTEA)

Signed-off-by: Dan Scott <dscott@laurentian.ca>

Add upgrade script for changeover to array_agg() and string_agg()

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

Found a few last functions that need changing to native SQL

These new uses probably went in more recently as authority fixes. Updated them to
use the native functions as well.

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

Keep dropping and creating array_accum()

We want to keep array_accum() around for now to support any custom SQL that
sites might have generated. Plus, we don't want it to show up in our search
path ahead of the native array_agg()!

We can always rebase this as a fix into the previous commit to squash history a
bit. "Nope, that never happened!"

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

More native SQL functions

Found four more places that this needed to be changed.

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

Prevent {NULL} values from blocking bib saving

With both the old ARRAY_ACCUM and ARRAY_AGG, and with array_nulls on,
PostgreSQL 9.0 on Debian Squeeze was generating values of '{NULL}'
rather than actual nulls. This would in turn prevent bibs without ISBNs
or ISSNs from being saved successfully. A CASE statement protects
against that problem and returns a clean NULL.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

ARRAY_ACCUM() -> ARRAY_AGG() for new authority nfi functions

Fix the new places where array_accum() has been able to sneak into the
schema.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

Teach asset.merge_record_assets about STRING_AGG()

Now that asset.merge_record_assets() is working, cut over to
STRING_AGG() so it can iterate over those unnecessary nodes faster :)

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

One more STRING_AGG() conversion

Teach biblio.indexing_ingest_or_delete() to use STRING_AGG() where
appropriate.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

Native SQL functions for Vandelay

Use STRING_AGG() where appropriate instead of
ARRAY_TO_STRING(ARRAY_ACCUM()), and ARRAY_AGG() instead of
ARRAY_ACCUM().

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

Replace ARRAY_ACCUM() custom function with ARRAY_AGG()

Tested each replacement to ensure that the return of NULL instead of
'{}' empty array doesn't break anything - although some functions didn't
seem to work with either function (for example,
asset.merge_record_assets() doesn't seem to actually move an 856 from
the source to the target record).

Also replace ARRAY_TO_STRING(ARRAY_AGG()) with STRING_AGG() in a couple
of places; see
http://www.depesz.com/index.php/2010/02/17/waiting-for-9-0-string_agg/
for one reason why (likely performance gain).

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

Skip duplicate username check when username unchanged

Changing a username, then changing back to the user's current
username results in a duplicate username detection -- a false
positive due to the API call finding a user with that username --
the current user. This prevents you from saving any other changes
you may have made to the user.

This can be prevented by skipping the duplicate username check if
the user is not new and the username in the field matches the
username stored in the in-memory Javascript representation of the
user.

There's still a potential corner case which could be avoided by
using a new/modified API call to either accept an "except this
au.id" argument or to return the au.id of the found user, but I
don't see a need for that at this point.

Signed-off-by: Jeff Godin <jgodin@tadl.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

OpenSearch release note

Add a short release note about the return of OpenSearch support.

Signed-off-by: Ben Shum <bshum@biblio.org>

Restore OpenSearch support and use TPAC search

This enables users to easily add the Evergreen search engine to their browser's
built-in set of search engines, and most importantly directs searches to the
TPAC instead of the weird old Supercat interface.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP1271218 - clarify how opt-in user notifications are shown in my account

As reported in LP127128, end users may not need to be shown the message that
"No notification preferences are configured" when in fact this message is only
displayed when the user opt-in notification settings are not configured.

So, alter this to only show the table and div when there are settings to be
configured and otherwise show nothing.

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Chris Sharp <csharp@georgialibraries.org>
Signed-off-by: Ben Shum <bshum@biblio.org>

LP#1164720 Prevent empty names in TPAC lists

Form submission for list creation with an empty name value now prevents
list creation and redirects the user to a new error page explaining why
list creation failed.

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Ben Shum <bshum@biblio.org>

Bumping base schema version to match latest upgrade

Signed-off-by: Bill Erickson <berick@esilibrary.com>
Signed-off-by: Jeff Godin <jgodin@tadl.org>

Revert "Eliminate an annoying and useless warning in the JavaScript Console."

This reverts commit bb4a797010201d6f26156c9e921d78690f265e35.

My fault for picking this in without more testing. This broke things like noted
in LP1275074.

Include RFID docs with full path

Ben Shum noticed that the RFID docs weren't showing up in the documentation; we
need to include the full relative path to the RFID docs.

Signed-off-by: Dan Scott <dscott@laurentian.ca>

Fix copy_info variables for result and record view

Because we yanked out the code that generates the link into a separate INCLUDE
file, the function was looking for something for copy_info. The one for result
was missing entirely and the one for record was backwards.

Signed-off-by: Ben Shum <bshum@biblio.org>
Signed-off-by: Dan Scott <dscott@laurentian.ca>

KPAC library page: give the kids some style

Just the bare minimum: some padding around the body, and a font colour and size
that's reasonably consistent with the rest of the look of the KPAC for h2 and
h3 elements.

Yes, I know there is text-transform:uppercase and h2 is sized at 48px but that
seems to be used for "You saved it!" and "Get it!" banner things. On the
library info page, we're using headings for headings.

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

Clean up the breadcrumb trail in the KPAC

Links to the library pages don't maintain GET context, so we can't feed search
or "Get It" links in the breadcrumb trail, currently. We can revisit this,
of course.

Wasn't there some fairy tale about kids and a breadcrumb trail... once upon a
time?

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>

KPAC root, OPAC root: we all live under one root

Signed-off-by: Dan Scott <dscott@laurentian.ca>
Signed-off-by: Ben Shum <bshum@biblio.org>