Bill Erickson [Fri, 3 Feb 2012 21:32:03 +0000 (16:32 -0500)]
TPac; credit card assumes all xacts if none selected
For consistency with the initial payment form and to preserve backwards
compatibility, if no transactions are selected, assume the user wants to
pay for all transactions.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Bill Erickson [Mon, 21 Nov 2011 20:47:37 +0000 (15:47 -0500)]
TPac; credit card payment "Processing..." page
To prevent impatient patrons from re-submitting credit card payment
request forms, insert a Processing... page between the initial form
submission and the actual payment processing.
Steps 3-5 are new or modified.
1. User selects transactions to pay
2. User fills in payment form and Submits
3. Payment form data is cached in memcache with a temporary token
4. Processing page is displayed with message about how processing the
payment can take time, do not refresh, etc.
5. Page is refreshed via http meta refresh to submit the form using
the temp token. The page is not visibly changed, though, since it
takes time for the form submission to begin returning data.
6. User is directed to the payment results page.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Bill Erickson [Mon, 21 Nov 2011 20:44:40 +0000 (15:44 -0500)]
TPac; support context-genereted refresh content
If the tpac context object contains a "refresh" attribute, it will be
used as the content attribute in a <meta http-equiv="refresh" ...>
element in the page head. If present, it will override the default
refresh content for the authentication timeout.
While we're in there, let's use the more correct form of the content
attribute: content="x; url=foo"
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Etheridge [Wed, 11 Jan 2012 18:57:43 +0000 (13:57 -0500)]
LP914821 template selection should be sticky
in Unified Volume/Copy Editor.
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
By default, Dijits and things that inherit from Dijits (at least in
Dojo 1.3.3 that we're still using), have a true value for an attribute
called scrollOnFocus. This makes your browser jump to ensure that the
widget that just received focus is in view.
You don't always want this though. Try expanding the extras on the
record detail page of the TPAC without this commit, but with AutoSuggest
enabled, for example.
This commit sets the scrollOnFocus attribute to false for the
AutoSuggest widget.
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Dan Scott [Sat, 10 Mar 2012 16:27:08 +0000 (11:27 -0500)]
Propagate search location on "Another Search" link
In commit 27b2037d72b I had intentionally prevented "Another Search" on
the results page from propagating the current search location, in the
belief that sites would want to use that as a way of starting a brand
new search session; after all, both the "Advanced search" and "Search
toolbar" offer propagation of the current parameters.
However, a number of people including Kathy Lussier and Ben Shum
indicated that they expected "Another Search" to maintain scope as well,
so we will propagate state here.
Sites may want to customize this link or the top-left logo link if they
want to provide a "Start a new search session" option.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Set whenever we can (including automatically via pcrud)
Clear whenever we disconnect (hopefully) from a location we set from.
Signed-off-by: Thomas Berezansky <tsbere@mvlc.org> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Alter create_auditor_history to include two new columns:
audit_user - To refer to the user who made this change
audit_ws - To refer to the workstation they were logged into
Alter create_auditor_func to include the new columns, as well as make it
stop assuming that the auditor table and the table it is based on use the
same column ordering. In additon, make it use OR REPLACE so it can be run
again later.
Alter create_auditor_lifecycle similarly to create_auditor_func, sans the
OR REPLACE functionality.
Add five new functions:
set_audit_info - To set the user and workstation IDs
get_audit_info - Used by the auditor functions to retrieve said IDs
clear_audit_info - To remove the stored IDs for whatever reason
fix_columns - To add/remove/update columns that have changed since any
auditors were initially created
update_auditors - To run a full pass of dropping all lifecycle views,
update for all column changes, re-create the auditor functions, and
re-create the auditor lifecycle views.
Signed-off-by: Thomas Berezansky <tsbere@mvlc.org> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Thomas Berezansky <tsbere@mvlc.org> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Michael Peters [Thu, 15 Mar 2012 14:04:05 +0000 (10:04 -0400)]
TTPAC material type icon compression
When cleaning up the images directory for TTPAC, I discovered
that the mattype icons had been saved with a poor compression
scheme and where taking up nearly 3MB, which was just too
high for a small set of icons.
I ran them through Photoshop CS5 and saved us a whopping 2.85MB
of disk space used by the images. The entire lot decreased from
3MB to just 152KB. Average icon size is now just about 1KB.
This should speed up load times when these images are displayed.
Signed-off-by: Michael Peters <mrpeters@library.in.gov> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Michael Peters [Fri, 16 Mar 2012 17:56:57 +0000 (13:56 -0400)]
LP#707167 Make open-ils.actor.usrname.exists not ignore deleted users
Currently, open-ils.actor.usrname.exists filters out deleted users. This
results in a user recieving indication as to why a user update
failed, when the reason is the username they're attempting to use
is in use, but is "deleted" in the database.
This branch makes open-ils.actor.usrname.exists look at deleted users
since they still prevent re-use of the delted user's username.
More information is in the original bug, https://bugs.launchpad.net/evergreen/+bug/707167.
Signed-off-by: Michael Peters <mrpeters@library.in.gov> Signed-off-by: Mike Rylander <mrylander@gmail.com>
New features for telephony currently in use at KCLS
I must ask the community's forgiveness for sitting on this for so long.
This code was developed in fits over time, and it had to be disentangled
from some site-specific configuration and workarounds before it could
be usefully published.
Enhancements over what's already committed to master include:
- a feature that allows an Evergreen system to query the PBX (where
PBX here refers to an Asterisk system plus the eg-pbx-* scripts from
Evergreen - particularly eg-pbx-mediator.pl) for the A/T event IDs
associated with failed notifcations, so that those A/T events can be
"rolled over" into new ones with a different event definition, and
thus a different notification method
- holiday awareness in eg-pbx-allocator (the piece run by cron whose
main responsibility is dripping files from a staging directory into
Asterisk's real spool for call files). There's a script (set_pbx_holidays)
that can be run by cron on an Evergreen system to periodically update the
PBX's table of holidays based on a given org unit's closed_date ranges.
- smart retry. In certain situations, if you put too many callfiles
into Asterisk's spool at once, Asterisk will try to make too many
calls at once, and all such calls just fail. That's what the
allocator is meant to prevent. Smart retry is about moving calls that
have been tried once, and will be retried again later due to resulting
in a busy signal or other problem, out of the spool to make room for
other calls that could be made in the mean time.
- miscellaneous helper scripts and tricks for spreading more
reportable/greppable information among A/T output and cleaning up
callfiles that result in no-op's for Asterisk. Lock file checking for
the allocator. Etc.
I also have tried to throw in some useful documentation. I admit things
look a bit jumbled, but I'll be as available as I can make myself to help
anyone who steps up to clarify documentation or improve code in order to
make general adoption of these features a little easier.
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Use getItemType() instead of getForm() in searchBarInit().
This seems to actually fix the problem with the item type limiter
being lost when going from advanced search to the results page.
Looks like the wrong method was being used to set the form_selector
value in search_bar.js. It was being set to the web form itself and
not to the value of the it (item type) parameter.
Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Liam Whalen [Tue, 1 Nov 2011 18:18:32 +0000 (14:18 -0400)]
When sorting a search results set the isFilterSort variable is passed
to searchBarSubmit. In side searchBarSumbit there is a call to the
opac_utils.js function clearSearchParams function. This removes all
search paramaters when sorting a result set, which is not the desired
effect. I have wraped clearSearchParams in a check to make sure
isFilterSort is not set before calling the clearSearchParams function,
and this check ensures that the current search's paramaters are passed
to sort operations.
Signed-off-by: Liam Whalen <lwhalen@evergreen-dev.catalogue.nrcan.gc.ca> Signed-off-by: Jason Stephenson <jason@sigio.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Dan Scott [Wed, 18 Jan 2012 05:27:05 +0000 (00:27 -0500)]
LP918020: Fix regression in biblio.extract_located_uris
The 2.0-2.1 upgrade script contained an old version of
biblio.extract_located_uris that functioned poorly. Fix the 2.0-2.1
upgrade script and ensure that sites upgrading to 2.1.2 get the fix as
well.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Etheridge [Fri, 24 Feb 2012 19:36:16 +0000 (14:36 -0500)]
lp928937, null date sorting
Changes util.date.db_date2Date so that nulls are not converted into the epoch.
This fixes date sorting in XUL-based lists in the staff client such that unset
dates are listed before actual dates.
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Avoid JSPAC hold placement interface problems when no holdable copies
This addresses a bug brought to my attention by John Jones.
When you're trying to place certain holds (at least title level holds) on
records for which there are no holdable formats (no descriptors as returned by
open-ils.search.metabib.record_to_descriptors), then
holdArgs.recordDescriptors in holds.js can be an empty array. There is
a line of code that assumes this will never happen.
The end user-visible result of the javascript failing here is an alert()
dialog about phone number format, but this is a red herring. That only
happens because the problem code is in one big function that has too
many responsibilites.
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Michael Peters <mrpeters@library.in.gov> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Bill Erickson [Tue, 7 Feb 2012 19:40:19 +0000 (14:40 -0500)]
Tpac (etc): improve payment history sorting/paging
Payments within a transaction will have the same timestamp. With a
large enough set of such payments, sorting by timestamp alone when
retrieving payments is not enough to guarantee unique pages of payments.
This adds a secondary sort on payment ID as a tie-breaker.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Apparently they ditched a containing folder in the zip...lets autodetect
that from now on.
Signed-off-by: Thomas Berezansky <tsbere@mvlc.org> Signed-off-by: Michael Peters <mrpeters@library.in.gov> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Etheridge [Mon, 23 Jan 2012 15:05:19 +0000 (10:05 -0500)]
patron search result order sorting
The default patron search "order by" on the server side is last name, first
name, and actor.usr.id.
The staff client intended to request last name, first name, middle name, and
date of birth (when actor.usr.id to be tacked on in the middle layer), but a bug
prevented this.
This fixes that.
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Bill Erickson [Wed, 1 Feb 2012 18:53:55 +0000 (13:53 -0500)]
Paging repairs for Event definition admin UI
If there are multiple event definitions with the same name that lie along
the page boundaries, it's possible to page through event definitions
seeing some def's multiple times and others never. This is a result of
sorting solely on event def. name when fetching event def's for display.
This patch adds secondary (and tertiary, and quaternary, and quinary)
sort columns as tie breakers to guarantee a unique set of defs across
pages.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Bill Erickson [Thu, 5 Jan 2012 18:18:29 +0000 (13:18 -0500)]
Do not exclude "identifier" facets by default
No longer ignore "identifier" fields when collecting and caching
facet data. This change causes all config.metabib_field's that are
marked as facet_fields to be taken into consideration.
Apart from marking a field as facet_field=false, it's also possible to
control which fields are visible within the jspac and tpac (pending LP
911908)
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Dan Scott [Mon, 19 Mar 2012 20:04:49 +0000 (16:04 -0400)]
TPAC: Prevent OU-invisible copies from being displayed to public
Per LP# 959577, "We've found that copies from OU's with OPAC visible set
to false are displaying on the record details page of tpac."
This fixes that by including the pertinent clause in mk_copy_query()'s
version of what in-db unapi does. Some day we will unite under the same
logic and be stronger!
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org>
Staff client: add a back-to-search-results button in the chrome that ...
... wraps the OPAC (whether JSPAC or TPAC).
You don't really need this when in OPAC view, because the OPAC page
itself will have a link that does the same thing. However, if you're
browsing through records in a search result set one at a time in some
other view, say the MARC view, then this button can come in handy.
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Michael Peters <mrpeters@library.in.gov> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Mike Rylander [Mon, 19 Mar 2012 15:15:24 +0000 (11:15 -0400)]
Vandelay-based record matching and import for Acquisitions
From the Lauchpad description:
In its current form, Acquisitions record importing is inflexible and
unforgiving. The user is given practically no control over how inbound
records match against existing records nor how to deal with record
collisions. This proposal attempts to resolve this problem by leveraging
Vandelay for matching, merging, overlaying, and importing acquisitions
records.
Suggested work flow:
1. Any time an ACQ record should be loaded into the catalog or simply
linked to an existing catalog record, the user is presented with an
additional set of Vandelay upload options to control the matching and
import behavior. This would include the vendor file upload UI, PO
activation, and PO "Load Bibs and Items". The Vandelay options would
include the full set of Match Sets, Merge Profiles, import options, etc.
the user normally has.
2. Once options are selected and, when applicable, the MARC file is uploaded,
the acquisitions records are added to a new type of ACQ Vandelay queue and
processed like any other Vandelay records.
3. Vandelay will be taught to recognize ACQ records and update the lineitems
to reflect any imported/linked catalog records.
4. Any records that fail to import will be available for manual inspection
within the Vandelay interface an can be re-imported from there.
Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Etheridge [Wed, 14 Mar 2012 06:52:49 +0000 (02:52 -0400)]
add Line# column to autogrid
hidden by default, but we need to make it show up in autogrid's column picker
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Etheridge [Wed, 14 Mar 2012 04:03:07 +0000 (00:03 -0400)]
make ordinal column work in Holdings Maintenance
brute force kludge
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Etheridge [Mon, 12 Mar 2012 20:06:51 +0000 (16:06 -0400)]
adds a Line# column to all xul lists
displays the ordinal position of the row
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jason Etheridge [Mon, 12 Mar 2012 19:39:20 +0000 (15:39 -0400)]
row_node vs my_node in util.list
Both of these should be holding the same xul treeitems, and I don't recall
wanting different names here, so I'm renaming both to treeitem_node.
Signed-off-by: Jason Etheridge <jason@esilibrary.com> Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Bill Erickson [Fri, 16 Mar 2012 17:38:13 +0000 (13:38 -0400)]
pcrud fleshing bug fixes
1. Avoid perm checks on org tree root lookup.
2. Test for child objects to flesh into has_a / might_fields differently
to avoid segfault when kids array is empty
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Fix a bug spotted by Jason Stephenson and Thomas Berezansky
I was leaving local variables called 'cache' high and dry in many
functions calling the initSessionCache() function that was introduced to
reduce code repetition.
This seems to fix that
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Bill Erickson <berick@esilibrary.com>
Mike Rylander [Wed, 14 Mar 2012 17:26:17 +0000 (13:26 -0400)]
Adjust controller-requiring logic for pcrud perm-verify mode
In cstore we always require the controller. However, in pcrud, we need to be
able to skip the controller requirement when we're verifying permissions. This
allows that.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Bill Erickson <berick@esilibrary.com>
This allows fleshing in pcrud queries if
a) the objects you would flesh have a pcrud controller in the IDL,
b) you have permissions for all the objects you would flesh just as
you would have to have them if you asked for them directly.
This is harder to read than it might be because of efforts to avoid
redundant perm checks for every fleshed object of the same class, and
efforts to keep results streaming out quickly.
Tests pretty well for me, but more rigor appreciated.
I received lots of help in making this happen from Mike Rylander and
Bill Erickson.
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Bill Erickson <berick@esilibrary.com>
Ben Shum [Sat, 17 Mar 2012 02:55:07 +0000 (22:55 -0400)]
Tpac - add additional subfields for title in search results
Originally, titles in Tpac only showed based on tag 245 subfield a.
This sometimes led to difficulty identifying specific records from
each other if it was part of a set or had additional title information
in other subfields.
Modified default title function to include the following:
subfield b = remainder of title
subfield n = number of part/section of a work
subfield p = name of part/section of a work
Signed-off-by: Ben Shum <bshum@biblio.org> Signed-off-by: Dan Scott <dan@coffeecode.net>
Michael Peters [Thu, 15 Mar 2012 13:13:14 +0000 (09:13 -0400)]
TTPAC icon color tweaks
In TTPAC, there are a few tiny icons used that were still in a blue
hue. This adds a green hue to the icons, so they match the skin, plus
boasts a 50% size decrease for the *_sign.png images, for
a tiny bit of extra performance.
Signed-off-by: Michael Peters <mrpeters@library.in.gov> Signed-off-by: Dan Scott <dan@coffeecode.net>
Dan Scott [Sat, 10 Mar 2012 19:34:11 +0000 (14:34 -0500)]
Concerto: increase copies, assign statuses
Add more call numbers and create multiple copies per branch.
Include BR3 and the Bookmobile for diversity and to provide
some easier testing of search scopes at different depths.
Switch to functions for the purpose of populating call numbers and
copies, so we can make changes in one central place rather than
repeating ourselves like crazy.
For a better facsimile of the real-world, assign Reshelving / Checked
out / Lost statuses to a subset of the copies. (When / if we create
transactional data, we could throw this bogus data away.)
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Bill Erickson <berick@esilibrary.com>
Bill Erickson [Wed, 7 Mar 2012 15:39:44 +0000 (10:39 -0500)]
ACQ PO state defaults to "pending" from upload
When uploading a batch of records to create a PO, give the PO the
initial state of "pending" instead of "on-order". If the PO is meant to
be activated, its state will updated to "on-order" during the activation
phase.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: James Fournie <jfournie@sitka.bclibraries.ca>
Enable notify checking for holds in A/T Validators
HoldIsAvailable and HoldIsCanceled get new checks, plus add HoldNotifyCheck
validator.
All three check for "check_$type_notify" parameters evaling to true, where
$type is email, sms, or phone. If true the hold's appropriate value is
checked and, if not suitable, the validator returns false.
For email the email_notify value must pass the is_true test.
For sms/phone the appropriate notify value must pass perl's truth test.
Upgrade script modifies default event definitions, with the exception of
the sms hold available event definition. That has no static id in the seed
values.
Signed-off-by: Thomas Berezansky <tsbere@mvlc.org> Signed-off-by: Bill Erickson <berick@esilibrary.com>
Bill Erickson [Wed, 29 Feb 2012 23:06:21 +0000 (18:06 -0500)]
Acq+Vandelay : consistent post-upload link names
For consistency with POs and selection lists, label the newly created
Vandelay queue link as "View Upload Queue" instead of showing the queue
name. While we're in there, let's i18n-ize those 3 labels.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Wed, 29 Feb 2012 23:03:08 +0000 (18:03 -0500)]
Acq+Vandelay: post-upload links open in embedded browser
This change causes the picklist, PO, and queue links which appear after
a vendor file upload to open their respective pages with the xul browser
wrapper. Without this, the existing page CSS is ignored and everything,
including href's, are grey and washed out.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Wed, 29 Feb 2012 22:34:14 +0000 (17:34 -0500)]
Acq+Vandelay : skip unneeded vandeley forms
If all lineitems in question are linked to catalog records, avoid
showing the unnecessary vandelay form. This was already true from the
PO activation action, but this patch adds the same check to the
create-po action from a selection list and PO 'load bibs and items'.
Minor label change for consistency in li_table (Queue -> queue).
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Wed, 29 Feb 2012 21:06:55 +0000 (16:06 -0500)]
Acq + Vandelay : exit early for 'load bibs and items' in picklist
The ACQ UI does not support loading bibs and items directly from the
picklist interface. If the user selects "load bibs and items" from the
picklist page they will be presented with the vandelay options, but the
actual upload/merge/etc will not do anything. This change prevents the
vandelay options from showing in this case.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Tue, 21 Feb 2012 15:56:38 +0000 (10:56 -0500)]
Acq+Vandelay : more aggressive streamed response throttling
With the addition of Vandelay processing, the server now wants to
respond with status information a lot more often when uploading /
process ACQ records. This patch applies the level of throttling to ACQ
processing as Vandelay uses natively, which starts verbose and quickly
cuts back on chatter as processing continues up to (currently) a
throttle window of 1 delivered response per 256 responses. Apart from
saving bandwith, this helps alleviate a known problem in the staff
client where long-running, dense streams of multipart responses
eventually stop bubbling up from the browser to the application.
Making the responses more sparse seems to help this a lot.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Tue, 14 Feb 2012 16:29:30 +0000 (11:29 -0500)]
Acq+Vandelay avoid premature asset creation
The presence of Vandelay parameters in API calls does not mean the user
necessarily wants to create PO assets (real copies/callnumbers). That's
controlled by a different, existing API parameter. This adds support
for processing bib data only in the presence of Vandelay parameters when
the user has not explicitly chosen to create assets.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Fri, 10 Feb 2012 15:42:24 +0000 (10:42 -0500)]
Acq+Vandelay form re-use new queue
In the Vandelay agent form, if a user creates a new destination queue
for the uploaded records, allow the user to re-use the new queue without
having to refresh the interface. This is primarily useful in the ACQ
upload UI, where re-using the pre-filled form is possible.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Fri, 10 Feb 2012 15:40:17 +0000 (10:40 -0500)]
AutoFieldWidget gets new useWriteStore option
If the useWriteStore attribute is set to true, the underlying data store
for dynamically create data will live in an ItemFileWriteStore instead
of a ReadStore, which allows the caller to modify the data after the
widget is rendered. It's useful for in-place additions of newly created
items.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Fri, 10 Feb 2012 15:37:52 +0000 (10:37 -0500)]
Acq upload UI re-use new picklist
In the ACQ file upload UI, if the user loads the records into a new
selection list, this change allows the user to continue using the new
list without having to refresh the page.
Without this change, re-using the new list would result in the system
trying to create a new list with the same name, which fails.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Wed, 8 Feb 2012 19:37:44 +0000 (14:37 -0500)]
ACQ+Vandelay open picklist/PO/queue in new tab
In the staff client, after a file upload, if a picklist, queue, or PO
are created, make the links open their respective pages in new tabs.
When not running inside the staff client, allow the links be regular
href's.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Wed, 1 Feb 2012 14:12:23 +0000 (09:12 -0500)]
ACQ+Vandelay detect existing queue with same name
Within the acq/vandelay agent, detect when a user has entered the name
of one of their existing queues. When detected, update the queue
selector to match the found queue (which updates the linked match set
and clears the queue selector input). Otherwise, users may not realize
they are adding items to an existing queue (with a pre-linked match
set).
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Tue, 31 Jan 2012 17:17:20 +0000 (12:17 -0500)]
ACQ+Vandelay disallow match_set change on selected queue
Consistent with Vandelay, if a queue is selected, show the linked
match_set and disable the match_set selector. Existing queues are
already linked to a match set, so allowing the user to select one has no
effect, which is confusing.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Tue, 31 Jan 2012 15:49:14 +0000 (10:49 -0500)]
ACQ+Vandelay API support for queue-and-match only from ACQ
Allow users to load ACQ records into Vandelay queues and apply a
match_set at queue creation time, without requiring the user to
import/merge the ACQ records.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Tue, 31 Jan 2012 15:01:01 +0000 (10:01 -0500)]
ACQ+Vandelay ACQ Records in vandelay queue selector
Adds a new type of record to select from the Vandelay queue selector.
Selecting "Acquisitions Records" will show queues with type 'acq'.
Selecting the existing "Bibliographic Records" selector will only show
queues with queue_type 'bib' (i.e. non-acq queues).
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Thu, 19 Jan 2012 19:02:32 +0000 (14:02 -0500)]
ACQ+Vandelay permission improvements
Be more vigilant about enforcing permissions for various ACQ and
Vandelay actions.
1. Using vandelay to create new bib records now requres the IMPORT_MARC
permission (same as open-ils.cat.biblio.record.xml.import). If the
permission fails, the queued record will fail import and be stamped with
a new "import.record.perm_failure" vandelay import error.
2. Added suport for testing additional permissions before a new record
is created via vandelay. This allows interfaces leveraging vandelay
(e.g. ACQ) to create a higher barrier to entry. Added an ACQ perm
IMPORT_ACQ_LINEITEM_BIB_RECORD_UPLOAD that prevents users from creating
new bib records directly from the ACQ vendor MARC file upload interface.
The secondary permission is not meant as a security enhancement, per se,
since the user is required to have the IMPORT_MARC permission to get this
far in the process. It's more of a feature-specific precautionary
permission to prevent unintended record creation within certain
permission groups.
3. Checking the CREATE_PURCHASE_ORDER permission during the ACQ vendor
file upload process. It's absence appears to have been an oversight.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Fri, 6 Jan 2012 21:31:22 +0000 (16:31 -0500)]
ACQ+Vandelay link to vandelay queue from lineitem list UI
In the general purpose lineitem interface (li_table), there is now a
link to the Vandelay queue which contains the queued bib record used to
load/merge/link the lineitem. Queue opens in a new staff client tab.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Tue, 6 Dec 2011 18:50:29 +0000 (13:50 -0500)]
ACQ+Vandelay import/merge user interface components
Acquisitions support for leveraging Vandelay import/merge controls
for loading ACQ records and/or linking ACQ records to existing
catalog records.
Dev includes:
* General purpose Vandelay input form and merge progress indicator (VLAgent)
* VLAgent is now shown in the following interfaces when a bib load/merge/link action is required:
* Vendor record upload interface
* PO creation
* PO "load bibs and items" action
* PO activation
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Wed, 30 Nov 2011 20:42:30 +0000 (15:42 -0500)]
ACQ+Vandelay API to support Vandelay-mediated imports
Adds support for merging, overlaying, importing, and linking (a no-op
merge) acquisitions lineitems to catalog records via Vandelay.
All non-imported lineitems will now be processed through Vandelay during
record upload and asset creation (load bibs and items), based on the
Vandelay import parameters provided by the caller. Each processed
lineitem will also be linked to its queued Vandelay record. In the case
of a failed import, such records can be subsequently imported manually
from the Vandelay interface using the manual record overlay options.
ACQ API calls affected now report on the progress of Vandelay record
import in addition to lineitem creation, copy creation, etc.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Bill Erickson [Wed, 30 Nov 2011 19:54:41 +0000 (14:54 -0500)]
ACQ+Vandelay schema and IDL changes w/ upgrade script
Adds a queued_record link from acq lineitems to vandelay queued record.
Ditto acq_lineitem_history. This will be used to link vandelay-mediated
acq lineitems to the vandelay record used for their import.
For ease of future changes, port the queue_type column from bib and
authority queues to ENUMs and drop the constraint, which is implicit in
the ENUM.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Ben Shum [Thu, 8 Mar 2012 07:00:26 +0000 (02:00 -0500)]
Add patron profile filter to horizontal search
LP 715837 added the ability to filter by a specific patron
profile during a patron search, but only in the default vertical
search interface. For sites using horizontal search, this new
filter option had never been included.
Signed-off-by: Ben Shum <bshum@biblio.org> Signed-off-by: Jason Stephenson <jstephenson@mvlc.org> Signed-off-by: Jason Etheridge <jason@esilibrary.com>
Dan Scott [Sun, 11 Mar 2012 00:53:47 +0000 (19:53 -0500)]
Add styles to org selector for loc groups
Looks like user agents don't let us do a lot with styling OPTION
elements (text colour, background colour) but this gives sites that want
to further distinguish copy groups from libraries in the org selector
something to work with.
Bill Erickson [Thu, 1 Mar 2012 18:51:53 +0000 (13:51 -0500)]
Copy Location Groups : sort to top option
Adds a 'top' flag to copy_location_groups which, when enabled, will
cause the location group to sort above the child org units in the org
unit selector in the tpac.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Dan Scott <dan@coffeecode.net>
This adds a new CGI param "copy_depth" which the record detail page uses
to determine which copies to display in the copy grid and what serials
holdings data to show.
The goal is to separate search depth from copy display depth for 2
main reasons:
1. When present, the search ou is set by the "locg" parameter (org +
copy location group). The "Show" links in the record detail page use
"loc" to determine the copy location org (and subsequently the depth),
which is ignored in the presence of "locg". In other words, we need
a different way to communicate which range of copies to display.
2. Separating copy depth and search depth allows us to display
search location-related summary information while at the same time
displaying a broader set of copy information. For example, searching
BR1, we can see copy summary info for BR1, SYS1, and CONS even when
explicitly viewing copy information for CONS. In other words, viewing a
broader set of copies for a record does not change the search/context
org unit, it only extends the set of copies to display.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Dan Scott <dan@coffeecode.net>
Bill Erickson [Wed, 22 Feb 2012 15:31:37 +0000 (10:31 -0500)]
Copy Location Search Groups : Release Notes
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Dan Scott <dan@coffeecode.net>
Bill Erickson [Fri, 17 Feb 2012 16:57:19 +0000 (11:57 -0500)]
Copy Location Search Groups : TPac org unit selector
Adds support for viewing and searching on copy location groups in the
tpac. Groups appear within the org unit selector, when the selector is
used in a search context. Groups display below the owning org unit
similar to a child org unit. Groups are displayed for all org units
that meet the following criteria: search org unit, physical location,
patron home org unit, plus ancestors and descendents of each.
To support this, TPac gets a new "locg" CGI parameter, which contains
the org unit and copy location group. It takes the form
org_id:group_id. The TPac mod_perl code will extract this value and
popuplate the search_ou accordingly. For consistency, we also use
ctx.search_ou instead of directly checking CGI.param('loc') within the
template environment.
This also includes a rewrite of the org_selector.tt2 template. It
changes it from a recursive routine to a depth-first while loop. I did
this mainly because the recursive approach was suffering from global
variable clobbering in the template environment. In theory, this new
approach should be faster as well.
Signed-off-by: Bill Erickson <berick@esilibrary.com> Signed-off-by: Dan Scott <dan@coffeecode.net>
Dan Scott [Sat, 10 Mar 2012 17:36:08 +0000 (12:36 -0500)]
TPAC: Teach mkurl() to discard POST params
Unless we explicitly ask for POST params with the "params" argument to
mkurl(), prevent them from propagating. (Can't think of situations when
we would want to propagate a POST param, but we err on the side of
flexibility so...)
Thanks to Thomas Berezansky for suggesting the CGI url_param() method of
filtering POST params.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Thomas Berezansky <tsbere@mvlc.org>
Dan Scott [Fri, 10 Feb 2012 02:55:51 +0000 (21:55 -0500)]
Sync AuthProxy config values with opensrf.xml.example
The open-ils.auth/app_settings/auth_limits section of
opensrf.xml.example offers different settings than O:A:AuthProxy looks
up, which would typically lead to the default fallbacks. Also,
O:A:AuthProxy appears to mistakenly use seed_timeout instead of
block_timeout.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Dan Wells <dbw2@calvin.edu>
projects / working / Evergreen.git / log