erickson [Fri, 12 Nov 2010 13:55:58 +0000 (13:55 +0000)]
Server-generated IDL JS
The goal is to reduce use of pre-onload XHR, which is known to cause
problems (dreaded white-screen-of-death) in firefox/xulrunner. Change
allows opac, staff client, and embedded browser interfaces to load
a pre-formatted JS object instead of IDL XML via XHR. In addition to
dropping the XHR, clients no longer parse the XML, which should reduce
page render time. Finally, in the staff interfaces, the full IDL is
once again loaded, so there is no need to specifiy per-page classes.
Per-page classes are still supported and used in the OPAC to reduce the
up-front load time.
Change requires an update to the Evergreen Apache config. See -r18700.
Part of this change included condensing fieldmapper.hash and
fielmapper.dojoData content into fieldmapper.Fieldmapper to avoid
circular dependencies, which was causing problems with IE. Will
eventually want to deprecate .hash and .dojoData, but for now they still
function as before.
phasefx [Thu, 11 Nov 2010 17:41:13 +0000 (17:41 +0000)]
patch from bshum to bring his Search Catalog button in the button bar inline with tsbere's r18431 for new tab vs current tab functionality with the button bar
dbs [Thu, 11 Nov 2010 16:55:52 +0000 (16:55 +0000)]
Avoid munging 035 when a new record is created
If there is no 003 in the record when it is created, then we will not attempt
to generate a 035. If the incoming record contains a 001 and 003, then we will
create a 035.
dbs [Thu, 11 Nov 2010 16:50:10 +0000 (16:50 +0000)]
Do not supply a default value for 003 in new authority records
With cat.maintain_control_numbers enabled by default, we can trust
the database trigger to create the appropriate 003 for us - and by
not supplying a 003 in the new record, we won't create a spurious
035 for a brand new record.
dbs [Thu, 11 Nov 2010 03:49:31 +0000 (03:49 +0000)]
Update the edit dates for authority and MFHD records when they are edited
Addresses the oversight in the original implementation that missed this;
important if we're going to differentiate between creating and editing
a record for triggers.
dbs [Wed, 10 Nov 2010 22:26:42 +0000 (22:26 +0000)]
Enable "maintain control numbers" and "record ID as TCN" behavior by default
Per http://ur1.ca/2bgc4, this behavior hews more closely to the MARC21
specification. Note, however, that duplicate bib detection in the
"Import via Z39.50" interface will be somewhat affected; a more
trustworthy workaround is to include the "Local catalog" in Z39.50 searches to
determine if a matching record already exists in the database.
senator [Wed, 10 Nov 2010 21:51:01 +0000 (21:51 +0000)]
Serials: patch from Dan Wells to address Launchpad bug #670734
Previously, a dialog box to select an org unit for the creation of legacy
MFHD records relied on a feature (window.opener) that has been
removed in newer versions of XULRunner.
This patch has that dialog box use a different, safer approach to accomplish
its task.
senator [Wed, 10 Nov 2010 21:29:51 +0000 (21:29 +0000)]
Booking: fix bug reported by Robert Soulliere wherein the items associated
with picked-up reservations didn't actually get their status set to 'checked
out' like they should have.
Robert reported this bug on Launchpad (#672748) and provided a patch, but
there were some neighboring bugs in the circulation logic anyway, so they've
all been fixed together.
phasefx [Wed, 10 Nov 2010 20:51:23 +0000 (20:51 +0000)]
replace Cookie.js/cookieManager with dojo.cookie. Some things to watch out for as we review this changeset:
In Cookie.js, a -1 passed as the expires parameter indicated a session cookie. We could also specify such things as "+1y" or a specific date. With dojo.cookie, an expires value of -1 deletes a cookie. Not specifying an expires option will create a session cookie. We can enter a date or a number days from today. For the case in the code where we had "+1y", I supplied a value of 365.
In the previous code, we used empty strings for a lot of cookie deletion. The canonical dojo example has us using null (with an expires of -1). There were also some cases where it seemed obvious that we were deleting cookies, but were at the same passing an actual value (see base.js). I changed these to nulls.
In opac_utils.js, grabFleshedUser(), I fixed a dead code path where a cookie deletion attempt came after a return statement.
I ran through the opac within and outside of the staff client, tried all of the acquisition UI's from the Acq menu, and most of the interfaces under Admin, and didn't find any auth cookie errors.
miker [Wed, 10 Nov 2010 19:53:37 +0000 (19:53 +0000)]
Improved permission checking speed for PCRUD mode.
When there are more than some number of rows (5, by default) we use usr_has_perm_at_all to get the list of locations and cache them. This cache is flushed per-session.
gmc [Wed, 10 Nov 2010 13:18:29 +0000 (13:18 +0000)]
do not use TRUNCATE when refreshing reporter.materialized_simple_record
Previous behavior would break Slony replication after doing a
bib load. Since a deletion is slower than a truncate, if you're
not using Slony replication, you may prefer to truncate rmsr
prior to calling reporter.enable_materialized_simple_record_trigger.
dbs [Mon, 8 Nov 2010 21:21:31 +0000 (21:21 +0000)]
Add the OpenURL resolver caching service as an optional component
Currently limited to supporting the Ex Libris SFX OpenURL resolver,
this caches the results of looking up full-text availability of
resources by ISBN or ISSN.
For an example of incorporating this service into your OPAC, see
http://svn.open-ils.org/trac/ILS-Contrib/browser/conifer/branches/rel_1_6_1/web/opac/skin/lul/xml/result/result_table.xml#L102
for returning the full-text availability of a given resource by ISSN.
senator [Mon, 8 Nov 2010 20:44:06 +0000 (20:44 +0000)]
Acq: in acq search, avoid trying to create a dropdown from every bib ID in
the system, providing a normal text input field instead (or search by file
of terms).
erickson [Mon, 8 Nov 2010 16:21:15 +0000 (16:21 +0000)]
holds retrieval API call sorting and cleanup; sort non-cancelled holds by ready-for pickup, then active, then suspended; use json_query to fetch IDs first, so id_list calls can avoid fetching the full hold objects in the ML. sort fleshed transit by send time to pick up the latest transit
dbs [Mon, 8 Nov 2010 00:34:00 +0000 (00:34 +0000)]
Improve Fedora prerequisite installer
* Explicitly install wget, which isn't installed in a minimal install
* Hack JavaScript-SpiderMonkey Makefile.PL for 32-bit Fedora
* Provide a fedora14 target
* Change "fedora-13" to "fedora13" to match OpenSRF prereq installer
dbs [Fri, 5 Nov 2010 17:50:05 +0000 (17:50 +0000)]
Keep the hierarchy in line for O:A:S:Publisher::action.pm
Oddly enough, this didn't seem to affect hold-targeting functionality, but
testing was focused on placing a hold via the OPAC, and there's potential
for other things to go amiss.
dbs [Fri, 5 Nov 2010 17:35:38 +0000 (17:35 +0000)]
Initial support for recall of items triggered by placing a hold
In academic libraries, it is common for groups like faculty and graduate
students to have extended loan periods (for example, 120 days), while
others have more common loan periods such as 3 weeks. In these environments,
it is desirable to have a hold placed on an item that has been loaned out
for an extended period to trigger a "recall": truncating the loan period,
setting available renewals to 0, optionally changing the fines associated
with overdues for the new due date, and notifying the current patron of
the recall.
New actor.org_unit_setting entries and hold targeting logic enable libraries to
control whether a hold will trigger a recall of a circulating item as follows:
* "Recalls: Circulation duration that triggers a recall" (recall threshold)
is specified as an interval (for example, "21 days"); any items with a
loan duration of less that this interval will not be considered for a
recall
* "Recalls: Truncated loan period" (return interval) is specified as an
interval (for example, "7 days"), such that the user who currently
has the item checked out will get the greater of either the recall
threshold, or the return interval.
* "Recalls: An array of fine amount, fine interval, and maximum fine"
is an optional setting that applied the new fine rules to the current
circulation period.
When a hold is placed and no available copies are found by the hold targeter,
the recall logic checks to see if the recall threshold and return interval
settings are set; if so, then the hold targeter checks the currently
checked-out copies to determine if any of the currently circulating items at
the designated pickup library have a loan duration longer than the recall
threshold. If so, then the eligible item with the nearest due date will be
recalled.
gmc [Fri, 5 Nov 2010 16:45:45 +0000 (16:45 +0000)]
fix user password reset request time column def
Needs to be a timestamp with time zone; fixes a bug
where it was interpreted as a UTC time, throwing off
the calculation of the expiration of the password reset
request.
phasefx [Fri, 5 Nov 2010 13:22:15 +0000 (13:22 +0000)]
Patch from Benjamin Shum for the staff client button bar. Thanks!
---
In our consortium, we have the Button Bar for our Evergreen staff
clients enabled all the time. As a result, we have had lots of feedback
from users that there is a need to include a "Search Catalog" option on
the Button Bar.
This is our first attempt at writing something against trunk and
submitting it as a minor patch. Melissa Lefebvre created the icon for
the button by adapting existing icon sets. I set out to create the
requisite .diff file against a recent copy of trunk. I was unsure of
how to apply our changes to other locales' lang.dtd files, the one I
found most immediately was for en-US (our default).
If the community approves, we would like to see this become part of
future releases of Evergreen 1.6 and 2.0.
Thank you for your consideration.
-- Ben
--
Benjamin Shum
Open Source Software Coordinator
Bibliomation, Inc.
32 Crest Road
Middlebury, CT 06762
203-577-4070, ext. 113
senator [Thu, 4 Nov 2010 21:19:12 +0000 (21:19 +0000)]
Net::uFTP isn't giving us needed functionality
Since the OpenILS::Utils::RemoteAccount already uses Net::SSH2 directly
(Net::uFTP was supposed to be an abstraction over SSH /and/ FTP), just use
Net::FTP (installed by default with perl on most systems) for FTP
gmc [Thu, 4 Nov 2010 01:03:42 +0000 (01:03 +0000)]
add TT helper to squeeze strings into the UNOB straitjacket
The force_jedi_unob helper strips non-ASCII characters from
template output but does so nicely by converting the string
to NFD first. This is meant to be used by EDI templates, although
could be useful for A/T output that goes to receipt
printers that cannot print diacritics.
senator [Tue, 2 Nov 2010 16:27:23 +0000 (16:27 +0000)]
Show money spent from each fund in an invoice voucher/print-out
In the (near) future, we really ought to use a library like Money::Currency
and refactor any Acq code that uses floating-point math. It's conceivable
that numeric type fields in the acq schema need reviewed (should some of them be
something like numeric(8,2) ?)
erickson [Tue, 2 Nov 2010 03:00:57 +0000 (03:00 +0000)]
AutoFieldWidget linked class autoloaders, etc.
* leave linked selector disabled until entries are fetched;
* added support for applying default linked class object loaders;
* added default loader for acq providers, since large sets of
providers can take an excessive amount of time w/ direct pcrud call.
erickson [Tue, 2 Nov 2010 02:55:21 +0000 (02:55 +0000)]
in SIP patron retrieval, only fetch non-archived penalties and penalties
that that matter (fines, overdues, blocking penalties). pair down the
penalty comparisons to avoid fleshing the penalty type, potentially
numerous times for a given type, by using the constant identifiers.
dbs [Tue, 2 Nov 2010 02:46:51 +0000 (02:46 +0000)]
Enable merge of authority records to do the right thing
The target and source authority record was flipped, causing
the update to fail. In reconsidering this function, it is not
necessary to change the contents of the source authority record
just to propagate the content of the target authority record
to any linked bibliographic records.
Instead, take the approach of updating the ID of the controlled
field in the bib record, then temporarily set "reingest on same
MARC" to TRUE and update the target authority record by setting
deleted = FALSE (which propagates the "changes" to the linked
bib records), then set "reingest on same MARC" flag back to its
original value.
gmc [Mon, 1 Nov 2010 23:22:53 +0000 (23:22 +0000)]
yet another replication race condition fix
Fixes problems that can occur when creating a lineitem
from an existing bib in the catalog; adds an authoritative
version of open-ils.acq.lineitem.retrieve.
dbs [Mon, 1 Nov 2010 20:41:14 +0000 (20:41 +0000)]
Replace hard-coded '(CONS)' for MARC control number identifier in authority records
We created an actor.org_unit_setting, 'cat.marc_control_number_identifier', for
specifying the preferred MARC control number identifier, but when we create a
new authority record from the MARC editor, the hardcoded value of 'CONS' is being
used.
This teaches the staff client how to pull the appropriate value from the AOUS
when invoking the MARC Editor.
dbs [Mon, 1 Nov 2010 19:44:16 +0000 (19:44 +0000)]
Ensure that changes to authority records propagate to linked bibliographic records
Per https://bugs.launchpad.net/evergreen/+bug/669596, updated authority records
weren't being reflected in bibliographic records with fields that link to those
authority records. We were missing the call to authority.propaagate_changes()
withing the ingest trigger on authority.record_entry.
miker [Mon, 1 Nov 2010 17:18:56 +0000 (17:18 +0000)]
Improve hold permit testing
* Discard matchpoints where the OU in a particular slot is not ancestor-or-self of the target object's value
* Discard matchpoints where strict_ou_match is true and the OU is not strictly matched
phasefx [Mon, 1 Nov 2010 17:13:42 +0000 (17:13 +0000)]
this wraps the input elements used for opac login in an actual form element, with the goal of allowing browsers to save credentials for the site. So far this appears to work for Firefox 3.6.12, but not for Chrome 7.0.517.41, Opera 10.63, nor Safari 5.0.2. IE 8.0.6001.18702 will prompt for saving credentials, but won't actually populate the form on a revisit. Some of this may be due to me using an invalid/expired SSL-cert on my dev box, but I suspect it's really because another page is not being loaded after form submission. Any ideas/suggestions?
gmc [Mon, 1 Nov 2010 15:39:39 +0000 (15:39 +0000)]
let libraries allow funds to be overcommitted
The check constraints on the balance warning and
stop limits were too strict - a library can validly
choose to permit a fund to have a commitment greater
than 100% of its allocation
gmc [Fri, 29 Oct 2010 01:25:01 +0000 (01:25 +0000)]
adjust quoting of phrase searches
This has three main effects:
* This formalizes the current behavior where a
phrase search like
title:"^Harry Potter"
acts as a left-anchored search and
title:"Harry Potter$"
acts as a right-anchored search.
In particular, this can be useful for constructing
searches of bibliographic call numbers.
* Other regex metacharacters in phrase searches are
now escaped.
* Phrase searches like "C++" will no longer crash; in fact,
this makes a phrase search currently the only way to
accurately retrieve all C++ titles with the usual
normalization rules
phasefx [Thu, 28 Oct 2010 21:19:48 +0000 (21:19 +0000)]
Put a placeholder row into the checkin list to better handle the async checkin option (so that items show up in the same order in which they were scanned; it also feels more responsive). This also forces us into a better way of handling errors: we keep the rows for such scans in the checkin list, but shove the error condition into the alert message field. These rows can't be acted upon with the Actions for Selected Items menu.
senator [Thu, 28 Oct 2010 18:14:06 +0000 (18:14 +0000)]
Alternate Serial Control UI misc bugfixes, minor tweaks and improvements.
Still fighting an autogrid that sometimes renders in an imcomplete, cut-off
sort of way in the distributions tab of the subscription details interface.
Add distribution count to summary tab.
The batch receive interface now prepopulates the circ mod, copy location, and
price fields based on each item's stream's distribution's copy template, when it
can. When users had gone through the steps to create a copy template, it was
confusing not to see that work reflected in the batch receive interface.
projects / working / Evergreen.git / log