Galen Charlton [Fri, 23 Feb 2018 21:54:51 +0000 (16:54 -0500)]
LP#1676608: inject copy alert dialog in two more places
The copy alert dialog (if an item has relevant copy alerts)
is now injected into the dialog box series when handling
checkouts of items that are in transit or where the patron
record has overridable conditions that would block the checkout.
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Galen Charlton [Thu, 22 Feb 2018 22:02:11 +0000 (17:02 -0500)]
LP#1676608: provide DB update script to convert legacy copy alert messages
Legacy copy alert messages are moved to new normal checkout and
normal checkin copy alerts. This patch also converts the foreign
key relationship from asset.copy_alert to asset.copy to a "fake"
one using a constraint trigger.
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Galen Charlton [Thu, 22 Feb 2018 21:34:15 +0000 (16:34 -0500)]
LP#1676608: tweak defaults for alerts in vol/copy editor
Changed the defaults key for new-style copy alerts to be the same
as what was used for legacy copy alert messages; this means that
users who set editor defaults prior to upgrading are more likely
to have the Copy Alerts button be active out of the box.
Also moved the location of Add/Edit Copy Alerts in the defaults
tab.
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
A handfull of cases have arisen, including but not limited to
the copy alerts feature, where focus-me is set to simply "true"
rather than being bound to to a scope variable. This patch formalizes
this and gets rid of the following console warning:
angular.min.js:119 TypeError: model.assign is not a function
at ui.js:23
at angular.min.js:160
at f (angular.min.js:45)
at angular.min.js:48
(anonymous) @ angular.min.js:119
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Galen Charlton [Wed, 7 Feb 2018 21:41:55 +0000 (16:41 -0500)]
LP#1676608: add some sorting to copy alert dialogs
Copy alert types in drop-downs are now sorted by name
and copy alerts themselves are now consistently sorted
by ID (which will have the affected of putting the most
recent alert at the bottom in the copy alert manager dialog).
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Galen Charlton [Wed, 7 Feb 2018 21:30:04 +0000 (16:30 -0500)]
LP#1676608: conditionally enable manage copy alerts button on item status details
The 'Manage' button for copy alerts on the Item Status Detail view
is now enabled only if there is at least one active copy alert. This
patch also tweaks how the copy alert dialogs sequence record update
and modal-closing so that the caller can be sure that the transaction
has committed.
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
pcrud.apply() is supposed to skip any entries in the list
of CUD actions that don't explicitly mark themselves as
being a creation, update, or deletion, but didn't manage
to fully skip them. As a consequence, the browser console
would note complaints that open-ils.pcrud.apply.$IDLCLASS
methods do not exist.
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
The Copy Alerts feature allows library staff to add customized alert
messages to copies. The copy alerts will appear when a specific event
takes place, such as when the copy is checked in, checked out, or
renewed. Alerts can be temporary or persistent: temporary alerts will be
disabled after the initial alert and acknowledgement from staff, while
persistent alerts will display each time the alert event takes place.
Copy Alerts can be configured to display at the circulating or owning
library only or, alternatively, when the library at which the alert
event takes place is not the circulating or owning library. Copy Alerts
can also be configured to provide options for the next copy status that
should be applied to an item. Library administrators will have the
ability to create and customize Copy Alert Types and to suppress copy
alerts at specific org units.
Copy alerts can be added via the volume/creator and the check in,
check out, and renew pages. Copy alerts can also be managed at the
item status page.
Copy alert types can be managed via the Copy Alert Types page in
Local Administration, and suppression of them can be adminstered
via the Copy Alert Suppression page under Local Administration.
Co-authored-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Mike Rylander [Wed, 8 Nov 2017 21:26:10 +0000 (16:26 -0500)]
LP#1733695: Pull record attribute values from multiple sources
Previously, a multi-valued record attribute could only be extracted using
one strategy (XPath, tag+subfield, fixed field, etc). For multi-valued
attributes, though, there's no reason to restrict this as long as the values
from various sources would, after normalization, have the same shape. So
now we allow that, and we update the item_lang definition in order to allow
searching for records with multiple languages in various boolean ways.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Jason Boyer [Thu, 7 Dec 2017 19:45:21 +0000 (14:45 -0500)]
LP1737016: Ease Constraints on afs.code
acq.funding_source.code currently has a UNIQUE
constraint applied, which can be confusing to
users. This branch only requires that a code be
unique to the funding source's owner, simplifying
the management of funding sources.
Signed-off-by: Jason Boyer <JBoyer@library.in.gov> Signed-off-by: Remington Steed <rjs7@calvin.edu> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Display a message in the web staff bib record display indicating when a
bib record has been merged to another record. The message includes the
merge date and target record and is rendered as a link to the target
record.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Remington Steed <rjs7@calvin.edu> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Adds "merge_date" and "merged_to" fields for bib records. Merge date
is applied to all records involved in a record merge at merge time.
merged_to is applied to all subordinate (merged) records and points to
the target record.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Remington Steed <rjs7@calvin.edu> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Jason Stephenson [Sun, 29 Oct 2017 19:14:41 +0000 (15:14 -0400)]
Lp 1694058: Fix Issue With Place Holds Reported in Testing
When staff did have to override in order to place the hold, and the
"Place another hold for this title" link was subsequnetly used, the
inputs for the patron barcodes and some other fields on the place
holds page were duplicated. To avoid this, we now use uniq from the
List::MoreUtils library when retrieving the hold targets list from the
CGI parameters.
Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
LP 1694058: Final OPAC Changes for Multiple Hold Placement
Add JavaScript code to disable the number of copies selection when a
part is chosen for a record that has both parts and non parts holds
available. If a part is chosen, the number of copies select should be
disabled and reset to 1. It should be enabled again if the All Parts
option is chosen. It does not appear if the hold requires a part.
Test Plan for the OPAC changes:
1. Set circ.holds.max_duplicate_holds to 5 for BR1.
2. Login to the OPAC as any patron from BR1.
3. Perform any search.
4. Select a title and click Place Hold.
5. See that the selection box for placing multiple holds does not appear on the place holds screen.
6. Place the hold to see that hold placement still works as expected.
8. Go to the patron’s list of holds in My Opac to see that the hold appears.
9. Cancel the hold so it doesn’t interfere with later testing.
10. Logout.
11. Login to the OPAC as any patron from BR2 or any branch other than BR1.
12. Perform any search.
13. Select a title and click Place Hold.
14. See that the selection box for placing multiple holds does not appear on the place holds screen.
15. Place the hold to see that hold placement still works as expected.
16. Go to the patron’s list of holds in My Opac to see that the hold appears.
17. Cancel the hold so it doesn’t interfere with later testing.
18. Logout.
19. Login to the web staff client as a circulator for BR1 at BR1.
20. Select a patron barcode for any BR1 patron.
21. Perform any search.
22. Choose a title without parts and click place holds.
23. See that the selection box for placing multiple holds appears with 1 as the default.
24. Click the selection box to see that the values go from 1 to 5.
25. Place any number of holds for the patron whose barcode was chosen in step 15.
26. Open the patron in the staff client and check the patron’s holds list to see that they have the correct number of holds for this title listed.
27. Repeat steps 21 – 26 with metarecord holds on one of the metarecord titles.
28. Perform a search and select a title.
29. Choose an item to place a copy hold for the same patron.
30. Notice that the multiple hold number selection box does not appear.
31. Place the hold as normal (if you like) to test that copy hold placement still works.
32. Search for a title with parts.
33. Select a part in the parts drop down or radio button.
34. Notice that the number of copies select box is reset to 1 and disabled when a part is selected.
35. Repeat the above steps for a BR2 circulation account logged in at BR2 (or any other branch staff).
36. Notice that the selection box never appears when placing holds where the org. unit setting does not apply.
37. Clear the org. unit setting for BR1 and repeat the above steps, if desired.
The web staff client tests should work in the XUL staff client as well.
Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Jason Stephenson [Sat, 30 Sep 2017 19:30:22 +0000 (15:30 -0400)]
LP 1694058: Add confirmation dialog for multiple title holds.
Add a dialog to confirm that the user really wants to place the
requested number of title or metarecord holds to the
validateHoldForm() function.
Along the way, we add a format() function to the JS String prototype
in the i18n_strings.tt2 so that we can have translated strings with
placeholders in JavaScript.
Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
LP 1694058: Add backend code to allow multiple hold placement.
We add a constant for the circ.holds.max_duplicate_holds setting.
We modify Holds.pm to check if we're placing a title or metarecord
hold, that we have the CREATE_DUPLICATE_HOLDS permission, and that we
haven't placed more than the maximum allowed number of duplicate holds
before returning the HOLD_EXISTS event.
Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
* Remove grunt devDependencies
* Add Webpack devDepenencies
* Copy and minify operations are now handled by Webpack via 'npm run
build' for dev builds and 'npm run build-prod' for
production/minified builds.
* Running 'npm run build-watch' executes webpack in --watch mode to
watch for affected file changes and automatically rebuild. Useful for
development.
* Karma unit tests are now invoked directly from node via 'npm run
test'.
* Docs and release installer updated to match.
* Removed long-outdated inline installer readme.
Webpack is configured to create bundles from sets of JS files. As it
stands, there are 2 sets: core.bundle.js and vendor.bundle.js. Core has
all of the EG core services that are loaded on every page. Vendor
contains all of the 3rd-party dependencies (angular, etc.). These 2
bundles are loaded on every web staff page (via base_js.tt2).
All other <script/> tags remain as they were.
When building in dev mode ('npm run build'), core service JS files are
bunded un-minified with an inline source map for debugging. Vendor
files (angularjs, etc.) are always minified because they are quite
large.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org>
Bill Erickson [Fri, 22 Dec 2017 17:19:43 +0000 (12:19 -0500)]
LP#1739803 Quiet some repetitive console messages for unit tests
These console logs repeat during unit test runs and add clutter (ditto
the browser console log). I commented them out instead of removing them
since they may still be useful for debugging in the future.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org>
Rebuild package.json starting with Angular 1.6. This resulted in a
number of dependency updates and small changes required to keep the
dependencies happy.
* Teach print.js to use promise-style $http responses, instead of
.success() and .error()
* Remove unused '$http' dependency in hatch.js
* Fix egDate custom test matcher to work with newer version of
karma-jasmine.
* Build IDL for test scripts using Grunt/exec "cwd" since Grunt fails to
understand "&&" commands, like 'cd foo && bar'.
* For correctness, move dev-only depenencies into the package.json
devDependencies section.
See also
https://docs.angularjs.org/guide/migration#migrating-from-1-5-to-1-6
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org>
Since the display is not refreshed until the payment finishes, it
is possible to fire off a second payment by double-clicking.
This commit is a simple way to disable the button while a payment is
in process, then enable it again after the display refreshes.
Since the 'sendPayment()' function is wrapped, we may need to move
this logic up, but this being a single promise is simpler, and it
solved the problem in testing.
Signed-off-by: Dan Wells <dbw2@calvin.edu> Signed-off-by: Jason Stephenson <jason@sigio.com>
Mike Rylander [Wed, 31 Jan 2018 22:07:42 +0000 (17:07 -0500)]
LP#1746584: Adjust bib visibility test for patron browse
With updates to address visibility testing issues for LURIs, a change was made
to allow the default bib tests to supply the most appropriate query_int
operator with which to join those to non-default tests. The browse code,
however, is all database-level and was not adjusted with the Perl code that
implements general search. This commit addresses that issue by acknowledging
that the bib vis testing code to provide its own operator, either | or &, as
appropriate to the actual default test values.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Jason Stephenson <jason@sigio.com>
Fetch copies in a series instead of in parallel when loading copy
barcodes from a file in the Item Status interface. This helps avoid
excessive pcrud process count.
Since this causes the action to take a little longer, the commit also
includes a progress dialog indicating copy retrieve progress.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Jason Stephenson <jason@sigio.com>
Bill Erickson [Thu, 25 Jan 2018 22:57:32 +0000 (17:57 -0500)]
LP#1745499 Patron bucket from file query consolidation
Replace one-pcrud-call-per-barcode with a single (streaming) pcrud
search call to fetch patron cards when using the barcode file upload
option in the web staff pending patron bucket UI. This avoids spawning
high number of pcrud processes.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Jason Stephenson <jason@sigio.com>
Jason Stephenson [Wed, 14 Feb 2018 23:32:46 +0000 (18:32 -0500)]
Lp 1749593: Fix TypeError: Cannot set property 'grid' of undefined
Commit b35f0c0 added a reference to the grid in the grid.dataProvider
in services/grid.js in the web staff client. This is great, but the
dataProvider could be undefined where it was being set. This leads to
busted grid views and a TypeError in the console.
This commit moves the code down until after a fallback dataProvider is
used when the dataProvider is undefined and thereby avoids both
problems of adding the grid to an undefined dataProvider.
Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Jason Boyer [Mon, 29 Jan 2018 22:02:01 +0000 (17:02 -0500)]
LP1737540: Add Patron Information to Receipts
Adds the following fields to a patron object on print_data:
first_given_name
second_given_name
family_name
suffix
barcode
money_summary.balance_owed
money_summary.total_paid
money_summary.total_owed
expire_date
alias
has_email
has_phone
On the following receipts: Checkout, Items Out, and Bill Payment.
(money_summary left out on bill payment because it will always be out of date.)
Signed-off-by: Jason Boyer <jboyer@library.in.gov> Signed-off-by: Jason Stephenson <jason@sigio.com>
Cesar Velez [Mon, 8 Jan 2018 21:55:56 +0000 (16:55 -0500)]
LP#1739648 - fix grid nextPage button by setting grid.totalCount
This wires up the grid.totalCount in the implemented DataProvider's
get method, which allows us to properly disable the next page button
for when we know there's no more data.
Signed-off by: Cesar Velez <cesar.velez@equinoxinitiative.org>
Signed-off-by: Jason Etheridge <jason@EquinoxInitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Cesar Velez [Thu, 4 Jan 2018 22:36:02 +0000 (17:36 -0500)]
LP#1739648 - fix item duplication in itemsout and holds grids
Since these grids have caching, if the offset was greater
the cached data length, the grid would try to fetch data again, even if
it's already there, pushing duplicate items into the cache. This should
prevent that.
Signed-off by: Cesar Velez <cesar.velez@equinoxinitiative.org>
Signed-off-by: Jason Etheridge <jason@EquinoxInitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
If you click "Make last visible" on the column that is already the last
visible, it bumps it down one position when it should not move. This
commit adjusts the logic so that columns that are before or already in
the last-visible slot (thus, '<' or '=') will end up in the last-visible
slot, and columns only go after the last-visible slot if they are
already after it.
Signed-off-by: Remington Steed <rjs7@calvin.edu> Signed-off-by: Bill Erickson <berickxx@gmail.com>
Adds a new "Manage Columns" grid action which opens a batch grid column
configuration dialog. In the dialog, users can show/hide multiple
columns at once. The column position actions
(left/right/far-left/far-right) and sort priority settings have also
been moved to this new UI.
The existing 'Configure Columns' action has been renamed to 'Managed
Column Widths' and now only supports making columns wider and narrower.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Remington Steed [Thu, 8 Feb 2018 15:46:31 +0000 (10:46 -0500)]
Docs: Add missing section to root.adoc
This section was separated into its own file a few months ago, and was
added to root_command_line_admin.adoc, but needs to also be added to the
main root.adoc. This commit does that.
Cesar Velez [Fri, 5 Jan 2018 19:23:01 +0000 (14:23 -0500)]
LP#1739669 - add PaymentType and persistkey bill payment hist grid
This adds "circ.patron.billhistory_payments" as the persist-key for that
grid and also adds a Payment Type column. Note that the Payment Type is
that of the individual payment, while the Last Payment Type is that
of the most recent payment on the billing in question.
To test
-------
[1] Set up a patron bill that has multiple payments on it off
different types (e.g., cash, credit, etc.)
[2] Apply the patch.
[3] In the payments tab of the Bill History page in the web staff
client, note that there is now a Payment Type column available.
Verify that it matches the types of the payments entered in
step 1.
[4] Verify that column settings for that payment history grid can be
saved.
Signed-off by: Cesar Velez <cesar.velez@equinoxinitiative.org> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Jason Etheridge <jason@EquinoxInitiative.org>
Galen Charlton [Tue, 9 Jan 2018 21:51:54 +0000 (16:51 -0500)]
LP#1724052: move stat-cat cache initialization to patron search service
When the patron search was split out of the patron app per bug 1701001,
the startup code that copied the current patron stat-cat list into
egCore.env was not moved over as well and remained in the patron app. As
a consequence, the patron search-to-hold dialog would fail to display
any patrons in the results list that had statistical categories assigned.
This patch moves that bit of initialization into the service.
To test
-------
[1] Ensure that some patron records have statistical categories assigned.
[2] In the web staff client, perform a catalog search, place a hold,
then use the 'Patron Search' button. Perform a search; note that
patrons with statistical categories are not displayed and that
errors are present in the browser debug console.
[3] Apply the patch and repeat step #2. This time, patrons that have
stat cats should be included in the results.
[4] Perform a patron search in the main patron search interface;
verify that patrons with statistical categories are retrieved.
[5] Directly go to the page listing stat cats for a patron (e.g.,
/eg/staff/circ/patron/$ID/stat_cats). Verify that the name
of the OU that owns the stat cat is displayed.
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Cesar Velez <cesar.velez@equinoxinitiative.org> Signed-off-by: Bill Erickson <berickxx@gmail.com>
Chris Sharp [Sun, 14 Jan 2018 13:39:53 +0000 (08:39 -0500)]
LP#1743220: Update script to accommodate Ubuntu 16.04.
- renamed the script to "install-ubuntu.sh"
- changed the hard-coded Ruby version to a variable
that depends on the Ubuntu release (per the lsb_release
command)
- saw the script work on my 16.04 server
Signed-off-by: Chris Sharp <csharp@georgialibraries.org> Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org>
Galen Charlton [Tue, 6 Feb 2018 20:58:57 +0000 (15:58 -0500)]
LP#1743262: (follow-up) disable submit button if approval code not supplied
This patch causes the submit button to be disabled on the CC payment
dialog if the user specifies recording an externally-process CC payment
but has not yet supplied an approval code.
1. Indicate in the CC form that the approval_code field is required,
styled consistently with the patron edit app.
2. Prevent the form dialog from closing via Submit when no approval code
is set.
3. Handle payment failures better by preventing the post-API code from
running (i.e. receipt printing) when the API fails.
To test
-------
[1] In the web staff client, attempt to record an externally-supplied
credit card payment without specifying an approval code. Note that
an alert message is presented.
[2] Apply the patch and repeat step 1. This time, the approval code is
displayed as a required field and the dialog cannot be submitted
until one is specified.
Works around an issue where users cannot log into the web client using iOS
because Safari, which all browsers running on iOS use to some extent, does not
support BroadcastChannel.
Many thanks to Galen Charlton for providing a diff with this solution!
Galen Charlton [Thu, 25 Jan 2018 21:58:33 +0000 (16:58 -0500)]
LP#1745486: avoid retrieving user by id::numeric during auth init
During open-ils.auth.authenticate.init, if looking up the user by barcode,
the actor.usr row is subsequently fetched by ID. However, the ID was
turned into a float, meaning that the query as sent to the database
was equivalent to
SELECT * FROM actor.usr WHERE id = 123.0000;
While PostgreSQL will accept this, it ends up doing a sequential
scan of the actor.usr table rather than an index lookup, making the
retrieval up to two orders of magnitude slower than it needs to be
and adding a couple percent unecessary I/O load on large, heavily
used database. This patch fixes that.
To test
-------
[1] Turn on PostgreSQL statment logging and log in as a user
by barcode. Note that there's a retrieval of actor.usr by
a float form of the ID.
[2] Apply the patch and repeat step 1. This time, the query
retrieves the row by the integer form of the ID.
Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Bill Erickson <berickxx@gmail.com>
When replacing a patron's barcode in the patron editor, discard any
previous replacement attempts done in the current edit session. This
prevents cases where secondary, etc. replacement cards are accidentally
created or NULL cards that result in ugly error messages.
Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Chris Sharp <csharp@georgialibraries.org>
Mike Rylander [Tue, 16 Jan 2018 23:54:47 +0000 (18:54 -0500)]
LP#1743650: Bib vis testing needs different handling
For bib-level visibility testing, we can only use the source helper for
positive inclusion, and have to deal with LURIs on a case-by-case basis.
This is because, unlike the copy visibility cache, the LURI cache is pre-
composed in a single value for all LURIs on a record, not separate ones
for each. This is fine, as we just need to find the effectively visible
org units and test for that subset of the relevant orgs. This commit provides
a sub to test for that, and handles LURI test composition at the perl
level rather than in the database.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Mike Rylander [Tue, 16 Jan 2018 21:26:31 +0000 (16:26 -0500)]
LP#1743639: Test location as proxy for location group
Location groups are a table remove from copies, and can cause very large
updates to the copy vis cache. So, we will expand location groups to
locations, in situ, to avoid this problem and the need to supply a trigger
to update th copy vis cache for groups. This will allow us to reclaim
the location group bit in the visibility cache namespace if necessary, at
a later date.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Mike Rylander [Tue, 16 Jan 2018 19:39:08 +0000 (14:39 -0500)]
LP#1743639: opac_visible on acplg is not what it seems
The asset.copy_location_group table has a column called opac_visible.
However, unlike other tables with a column of that name, it controls the
visibility of the location group itself, not the visibility of the copies
within the group. This is regretable because the 3.0+ search code thinks
that it should hide copies.
This commit adjusts the default location group filter applied to patron
searches so that they act as expected.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Kyle Huckins [Thu, 11 Jan 2018 18:10:51 +0000 (18:10 +0000)]
lp1742194 - Print Current Bills only printing circ bills
- Check xact for Circulation before assigning circulation data to
xact for printing.
To test
-------
[1] Set up a patron with at least one "grocery" bill.
[2] In the web staff client, go to
Patron Account > Bills > Actions > Print Bills
[3] Note that the current bills summary is not printed
and that there an error in the browser console.
[4] Apply the patch and repeat step #2. This time, the
bills summary should be printed.
Dan Wells [Wed, 27 Dec 2017 13:36:23 +0000 (08:36 -0500)]
LP#1730470 Restore XUL serial receive compatibility
The older XUL interface does not have quite the same fleshing as the
new web-staff, so adding the 'move previous' code into the unitize
function breaks the old interface.
Since we already are looking up the distribution to get the setting,
let's just tack it on when it is missing.
To test
-------
[1] Set up a serial subscription with predicted issuances, then from
the XUL serials control view attempt to receive an issue. Note
that there will be a crash.
[2] Apply the patch and repeat step #1; this time, the issuance
receive should work.
[3] Verify that receiving issues via the web staff interface continues
to work.
Signed-off-by: Dan Wells <dbw2@calvin.edu> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Boyer [Fri, 5 Jan 2018 21:55:15 +0000 (16:55 -0500)]
LP1741072: Fix JS test for template conversion
Changing 2 fields from "strings" to numbers does
cause such a test to fail, as one might expect.
Now the test is updated to meet the current
expectations of the template conversion.
Signed-off-by: Jason Boyer <JBoyer@library.in.gov> Signed-off-by: Bill Erickson <berickxx@gmail.com> Signed-off-by: Galen Charlton <gmc@equinoxinitiative.org>
Mike Rylander [Mon, 11 Dec 2017 21:21:39 +0000 (16:21 -0500)]
LP#1736763: Wait until the org tree is there before using it
In offline mode, we need to wait for the org tree to be fetched or recreated
before we set the working location dropdown. This commit adds a function
to loop and wait for that to happen.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Cesar Velez <cesar.velez@equinoxinitiative.org>
Mike Rylander [Wed, 6 Dec 2017 23:18:29 +0000 (18:18 -0500)]
LP#1730758: Track record visibility on all Located URI DML
This commit simplifies the handling of Located URI call numbers by separating
them from normal call numbers. This will allow proper updating of bib-level
visibility attributes for all actions that may happen to a Located URI.
We also insure that we don't return a NULL INT[] as an attribute set for
either copies or bibs. This was always impossible for copies, but was
possible for bibs. Now both are future-proof.
The upgrade script contains a post-transaction command to forcibly update the
visibility attributes of all bibs that make use of Located URIs or bib sources.
It may take a while to run on large datasets. If it it running too long, it
can be canceled and the following psql commands will create and run a script
that will perform the same action serially over time without blocking writes to
bibs:
\t
\o /tmp/luri_visibility_update.sql
SELECT 'UPDATE biblio.record_entry SET ' ||
'vis_attr_vector = biblio.calculate_bib_visibility_attribute_set(id) ' ||
'WHERE id = ' || id || '; SELECT ' || id || ';'
FROM biblio.record_entry
WHERE id IN (
SELECT DISTINCT cn.record
FROM asset.call_number cn
WHERE NOT cn.deleted
AND cn.label = '##URI##'
AND EXISTS (
SELECT 1
FROM asset.uri_call_number_map m
WHERE m.call_number = cn.id
)
UNION
SELECT id FROM biblio.record_entry WHERE source IS NOT NULL
);
\o
\t
\i /tmp/luri_visibility_update.sql
It will output the id of each updated bib so that the script can be killed
and then edited to remove completed bibs. The remainder can be run at a
later time.
NOTE: When the internal flag 'ingest.reingest.force_on_same_marc' is enabled,
we do NOT update the bib's visibility attributes, as doing so causes a loop
and an eventual trigger stack violation. This flag should ONLY be used when
forcing reingest of record attributes (NOT visibility attributes), search,
facet, and display fields, so if using this flag under normal operation,
proceed at your own risk and know that Located URI and bib source changes
will not be reflected in the visibility attributes of the record.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Either the source or the existance of a Located URI will trigger record
visibility, so we will OR them together.
NOTE: This branch is dependant on the branch supplied on LP#1730758. Without
that fix, the bib-level visibility attributes are not updated properly when
Located URIs are added or changed.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
The site() filter and #staff modifier are used to decide when and how to
include certain query filters, such as circ_lib or luri_org. Unfortunately,
site() is sometimes excluded (whole-tree search) and the test for staff-
iness was not specific enough, leading to incorrect queries in those cases
where information was lacking. Now, we treat site() specially, forcing a
default of "top-of-tree", and we check for the #staff modifier directly
where necessary.
Note: this commit also addresses LP#1736992 which is about staff searches
using the limit-to-available modifier. As a side effect of the special
site() treatment, that issue is resolved.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Kathy Lussier <klussier@masslnc.org>
Cesar Velez [Thu, 30 Nov 2017 20:36:35 +0000 (15:36 -0500)]
LP#1710405 - remove Modify + Use Edits buttons in z3950 overlay
The workflow for using a z3950 record to overlay a marked bib
was confusing since it requires first saving the z3950 marc edits
via the Modify button at the top, and only then hitting Use Edits,
at the bottom of the modals. User often miss the click Modify part,
thus this simplifies that, having just a "Save" button and a Cancel.
Signed-off by: Cesar Velez <cesar.velez@equinoxinitiative.org>
Signed-off-by: Jason Etheridge <jason@EquinoxInitiative.org>
projects / working / Evergreen.git / log