Mike Rylander [Tue, 21 Jan 2014 17:31:13 +0000 (12:31 -0500)]
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>
Ben Shum [Sat, 8 Feb 2014 16:14:23 +0000 (11:14 -0500)]
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>
Dan Scott [Tue, 26 Nov 2013 17:44:51 +0000 (12:44 -0500)]
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>
Kyle Tomita [Thu, 6 Feb 2014 00:23:18 +0000 (16:23 -0800)]
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>
Dan Scott [Fri, 17 Jan 2014 01:59:13 +0000 (20:59 -0500)]
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>
Dan Scott [Tue, 3 Jan 2012 16:53:49 +0000 (11:53 -0500)]
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>
Dan Scott [Wed, 12 Oct 2011 03:53:33 +0000 (23:53 -0400)]
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>
Jeff Godin [Wed, 30 Oct 2013 17:20:28 +0000 (13:20 -0400)]
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>
Dan Scott [Wed, 15 Jan 2014 04:46:41 +0000 (23:46 -0500)]
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>
Ben Shum [Tue, 21 Jan 2014 18:52:49 +0000 (13:52 -0500)]
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>
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>
Ben Shum [Sat, 25 Jan 2014 04:56:45 +0000 (23:56 -0500)]
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>
Dan Scott [Thu, 23 Jan 2014 18:51:30 +0000 (13:51 -0500)]
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>
Dan Scott [Thu, 23 Jan 2014 17:10:50 +0000 (12:10 -0500)]
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>
Dan Scott [Thu, 23 Jan 2014 14:10:59 +0000 (09:10 -0500)]
KPAC: Won't somebody think of the children's record details?
Factor out the library name link logic into a separate template that can be
included by TPAC search results, TPAC record details, and KPAC record details.
Notice along the way that the KPAC is devoid of copy-level structured data,
but save that for later.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Ben Shum <bshum@biblio.org>
Dan Scott [Wed, 22 Jan 2014 15:12:37 +0000 (10:12 -0500)]
TPAC: Link library in results page to library page
Dan Wells noticed that the links on results pages went to a different location
(potentially, depending on the "prefer external library uri" OUS) than the
links on record pages. Make things consistent.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Ben Shum <bshum@biblio.org>
Dan Wells [Fri, 14 Dec 2012 19:02:22 +0000 (14:02 -0500)]
Add new option for max/min lost item pricing
In addition to a 'default' price when price is missing or zero,
we can also accommodate a range of prices by saying it should be
at least 'X' and not more than 'Y'. This also allows you to
effectively set a fixed price for all lost items by setting min
and max to an equal amount.
Signed-off-by: Dan Wells <dbw2@calvin.edu> Signed-off-by: Jeff Godin <jgodin@tadl.org>
Remington Steed [Wed, 27 Nov 2013 14:55:35 +0000 (09:55 -0500)]
LP#842991 Reports needs to error when deleting fails
When attempting to delete a report template, the reports interface
incorrectly shows "Action Succeeded" even when it partly fails. If the
user performing the delete does not own all the dependent reports and
outputs, then the perl function returns 1 to indicate that some were
deleted, but not all. However, the javascript that calls that function
does not check for a return value of 1. This commit adds that check, and
also improves the formatting of the related error messages.
Signed-off-by: Remington Steed <rjs7@calvin.edu> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Serials: improve routing slips per Dan Wells' suggestions on LP #1229349
- Don't repeat the same address for the same user
- Label mailing vs. billing
- Border to separate mailing and billing addresses
- [Also] move some inline styles to a CSS file
Dan Scott [Mon, 6 Jan 2014 19:33:04 +0000 (14:33 -0500)]
Copy details link to Evergreen library page by default
Add a new library setting, 'lib.prefer_external_url', to force library names in
copy details pages to link to their associated 'lib.info_url' setting rather
than the Evergreen-generated page. In the absence of a 'lib.info_url' setting,
the name will fall back to being linked to the Evergreen-generated page.
Per discussion with Ben Shum and Kathy Lussier.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Dan Scott [Mon, 6 Jan 2014 17:22:28 +0000 (12:22 -0500)]
Use the 'format.time' OUS to format library hours
Sparked by a concern that Ben Shum had over the display of library hours in 24
hour format, I eventually figured out that we could use the Template::Toolkit
Date module to format the times according to library preferences. And given
that we already have a format.time OUS, let's use it.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Dan Scott [Mon, 28 Oct 2013 15:26:55 +0000 (11:26 -0400)]
TPAC Library schema info - add some caching
Rather than hitting the database every time, cache our requests for a
period of time. It's unlikely that hours of operation or addresses are
going to change very often...
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Dan Scott [Tue, 17 Dec 2013 21:29:35 +0000 (16:29 -0500)]
Link from copies to library info page
In the absence of a specific lib.info_url org unit setting, link to the library
info page.
If we _do_ have a lib.info_url org unit setting, link from the library info
page to to the external web site.
Rather than wrapping everything in the table cell as the seller, focus
on the <a> and give it an explicit Library typeof attribute; then we can
offer up the name of the library as the name attribute, should the
crawler be too lazy to follow the link.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Dan Scott [Sat, 26 Oct 2013 21:32:22 +0000 (17:32 -0400)]
First inklings of schema.org support for library pages
Adds a TPAC page where the information for a given library can be publicly
exposed; in addition, that information is marked up with schema.org structured
data so that, for example, a search engine can offer up the hours of operation
and contact information for a given branch right on the search results page.
We give the content-wrapper a main-content for indentation, to be consistent
with other pages in the TPAC.
We also add the cargo-cult common-full-pad div for a bottom margin.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Steven Callender [Thu, 10 Oct 2013 20:09:06 +0000 (16:09 -0400)]
Fixed title/author display at checkout for non-pre-cat items.
Fixed a display problem on checkout where items that were formely pre-cats
were still displaying the pre-cat data. Now it will make sure the item is
still a pre-cat before using the dummy data.
Signed-off-by: Steven Callender <stevecallender@esilibrary.com> Signed-off-by: Ben Shum <bshum@biblio.org>
Dan Scott [Thu, 16 Jan 2014 03:33:25 +0000 (22:33 -0500)]
TPAC: Display authors using inline-block
Rather than letting author credits wrap willy-nilly, use inline-block mode so
that each line will begin with a new author. Devices with constrained screens
in particular benefit from this, as the list of authors is much easier to scan,
and wider screens can still fit multiple authors on a single line.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Ben Shum <bshum@biblio.org>
Dan Scott [Wed, 15 Jan 2014 20:25:02 +0000 (15:25 -0500)]
TPAC: Use indexed subfields only in author search links
Addresses LP# 1267231 in which we found that the titles of works in the added
author field (such as subfield t) were showing up in the link ahead of the
author's birth and death date (if applicable). Now we reserve the link for
only the indexed author subfields (depends on whether the name is personal,
corporate, or conference, but generally subfields a/b/c/d/e/n/q), then the
extra subfields go after the name + dates, then we finally put the relationship
into the parentheses after everything else.
We also simplify the markup so that each name is contained in a single <span
class="rdetail_author_div"> element to make it easier to control the layout.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Ben Shum <bshum@biblio.org>
Serials: Batch Receive interface Copy Location widget was inert
Choosing a value from the Copy Location widget wasn't working. You'd
get a default copy location (or the one supplied by the Receive Unit
Copy Template) no matter what you set that control too.
This is fixed. One does not read from XUL menulist element by using its
'value' property.
LP#882586: In Serial Control View, show copy templates owned by ancestors
When creating/editing distributions, dropdowns for Receive Unit Template
and Bind Unit Template now show copy templates owned not only at the
distribution's holding lib, but at any of the holding lib's ancestors.
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Dan Scott [Mon, 20 Jan 2014 03:24:07 +0000 (22:24 -0500)]
Create 008 - take only the first field/subfield combo
In the event that there were multiple $a subfields in, say, 044, the create 008
function could spew raw XML into the 008 field. Avoid that by taking the primary
instance.
Pasi Kallinen [Thu, 18 Jul 2013 18:00:31 +0000 (21:00 +0300)]
Allow generating fixed data in control field 008 quickly
Language and country codes come from 041a and 044a,
Date1 comes from 260c, and the record date is set to current day.
Signed-off-by: Pasi Kallinen <pasi.kallinen@pttk.fi> Signed-off-by: Dan Scott <dscott@laurentian.ca>
Conflicts:
Open-ILS/xul/staff_client/server/cat/marcedit.js
Galen Charlton [Fri, 4 Oct 2013 21:44:56 +0000 (14:44 -0700)]
LP#1235474: fix failure of authority browse context menu to render
This patch fixes a situation where if an authority record that would
be brought up by the authority browse context menu in the MARC editor
lacks an 003 field, the context menu would not be rendered all, meaning
that the default context menu (e.g., cut/copy/paste/etc.) would be
displayed instead.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Dan Scott <dscott@laurentian.ca>
Elliot Voris [Mon, 9 Dec 2013 18:57:20 +0000 (12:57 -0600)]
LP104785: Selfcheck needs to be run with HTTPS
Redirect attempts to connect to the self-check via HTTP to the more secure
HTTPS instead. Addresses this bug reported in launchpad:
https://bugs.launchpad.net/evergreen/+bug/1047485
Based on work for Apache 2.2 by Michael Peters <mrpeters@library.in.gov>,
updated and applied to 2.4 by Elliot Voris.
Signed-off-by: Elliot Voris <evoris@slcconline.edu> Signed-off-by: Dan Scott <dscott@laurentian.ca>
Dan Scott [Sun, 19 Jan 2014 16:30:57 +0000 (11:30 -0500)]
Fedora: Net::SSLeay 1.55 is too new for Net::HTTPS::Any
Although Net::HTTPS:Any wants Net:SSLeay 1.30 or later, 1.55 is apparently too
late, or something. The alternative is Crypt::SSLeay, which seems to work, so
let's go with that.
Pasi Kallinen [Fri, 28 Jun 2013 05:47:11 +0000 (08:47 +0300)]
Replace deprecated javascript escape() with encodeURIComponent()
Signed-off-by: Pasi Kallinen <pasi.kallinen@pttk.fi> Signed-off-by: Dan Scott <dscott@laurentian.ca>
Conflicts:
Open-ILS/xul/staff_client/chrome/content/auth/controller.js
Open-ILS/xul/staff_client/chrome/content/util/error.js
Open-ILS/xul/staff_client/chrome/content/util/print.js
Open-ILS/xul/staff_client/server/admin/offline_manage_xacts.js
Open-ILS/xul/staff_client/server/circ/print_list_template_editor.js
Mike Rylander [Thu, 16 Jan 2014 18:58:12 +0000 (13:58 -0500)]
Fix proximity adjustment calculator
The canonical use case for proximity adjustment is to use it broadly
to effect the local proximity of items within a particular area, in
a like manner, across many parts of the org tree. However, in cases
where it is lightly used, or used with non-overlapping criteria columns,
there are cases where the adjustment will spread beyond its intended
range of influence.
The changes here create a much stricter and correct test for rule
matching, which will end in the correct choice of rules in all cases.
Signed-off-by: Mike Rylander <mrylander@gmail.com> Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
In prereq installer, don't try to chown extracted files to original UID/GID
As this Makefile is designed to be used by a make process running as
root, this will avoid problems on some systems when tar extracts
files and tries to give them the same UID/GID as they have in their
archives, but those accounts don't exist on the local system.
e.g, libdbi has files set 1000/1000. On most systems, there's a user
and a group with these IDs, but there can be systems where nothing
uses those IDs, and where tar fails with e.g. 'tar: libdbi-0.8.3:
Cannot change ownership to uid 1000, gid 1000: Invalid argument'
Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Galen Charlton [Tue, 14 Jan 2014 18:30:48 +0000 (13:30 -0500)]
LP#1234201: fix menu item to display patron requests (if summary is horizontal)
This patch adds a missing command element to the horizontal
patron summary display so that the 'Acquisition Requests' menu
item will work.
To test:
[1] Set the Patron circulation summary is horizontal library setting
to true.
[2] Retrieve a patron and try using the Acquisition Requests menu
item. It will do nothing.
[3] Apply the patch and clear cache in the staff client.
[4] Repeat step 2. This time, the Acquisitions Request menu item
should bring up the list of the patron's requests.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Srey Seng <sreyseng@gmail.com>
Dan Scott [Fri, 25 Oct 2013 17:54:35 +0000 (13:54 -0400)]
Add pgTAP test for normalized MARC records
Specifically, we're concerned about ensuring that any changes related to
the fix for Encode (requiring the removal of the effectively useless but
now erroring decode_utf8()) do not inadvertently change the content of
the MARC records on ingest. Because we wouldn't want to have to reingest
everything to normalize search, etc.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Dan Wells <dbw2@calvin.edu>
Dan Scott [Mon, 21 Oct 2013 15:58:06 +0000 (11:58 -0400)]
Encode.pm change to the UTF8 flag
In the past, decode_utf8() silently ignored attempts to decode data
with the UTF8 flag on. However, Encode.pm changed around 2.53 and
resulted in wide-character errors preventing the likes of
--load-all-sample from working.
Since our data should already be UTF8 when it hits these functions, we
should not be trying to decode it in the first place.
Signed-off-by: Dan Wells <dbw2@calvin.edu> Signed-off-by: Dan Scott <dscott@laurentian.ca>
Galen Charlton [Tue, 14 Jan 2014 16:00:38 +0000 (11:00 -0500)]
LP#1269042: prevent acq seach from building dropdown of every copy ID in the DB
This patch ensures that if a user sets a search field in the
acquisitions general search form to the line item details copy ID
(labeled "LID - Evergreen Copy ID"), the form presents a normal
input widget rather than trying to build a drop-down containing
every copy ID in the database. Except for very small databases,
the previous behavior would mean that an open-ils.pcrud backend
would eventually consume a great deal of memory trying to service
a request for all copies in the system.
Signed-off-by: Galen Charlton <gmc@esilibrary.com> Signed-off-by: Mike Rylander <mrylander@gmail.com>
Jeff Godin [Wed, 30 Oct 2013 19:29:24 +0000 (15:29 -0400)]
Treat empty username as invalid in user editor
There seems to be undesired interaction between the required
attribute and the isValid method on a dojo/dijit ValidationTextBox.
If both are set, isValid needs to check for an empty value,
otherwise the field is considered valid the moment it gains focus.
We don't want an empty username to be considered valid, because we
know that we will be unable to save the user.
We can work around this by teaching the isValid function to consider
an empty value as invalid.
Signed-off-by: Jeff Godin <jgodin@tadl.org> Signed-off-by: Melissa Ceraso <mceraso@biblio.org> Signed-off-by: Ben Shum <bshum@biblio.org>
Dan Scott [Thu, 9 Jan 2014 17:19:38 +0000 (12:19 -0500)]
Fedora needs Locale::Maketext::Lexicon too
Per commit 66c45b647, install Locale::Maketext::Lexicon on Fedora to satisfy
that dependency. How did _this_ one slip by so long! Thank heavens for
tests.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Chris Sharp <csharp@georgialibraries.org>
Dan Scott [Thu, 9 Jan 2014 03:39:57 +0000 (22:39 -0500)]
Repair the live_t regression tests
When we added more realistic hours of operation for some of the branches,
several of the live_t regression tests failed because they depended on the
default hours of operation for BR4. Rather than change the regression test,
just switch BR4 back to the default hours of operation and use BR1 as the
closed-on-weekends branch instead.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Bill Erickson <berick@esilibrary.com>
Srey Seng [Tue, 7 Jan 2014 22:30:49 +0000 (14:30 -0800)]
LP#1266937: Fix missing/incorrect regex for authority in Vandelay.pm
Add in missing "~" in function import_record_list_impl when
type is authority records. Also replace "auth" with "authority"
so that the generated values match up with functions in the
database.
Signed-off-by: Srey Seng <sreyseng@gmail.com> Signed-off-by: Bill Erickson <berick@esilibrary.com>
Chris Sharp [Wed, 8 Jan 2014 19:51:59 +0000 (14:51 -0500)]
fix pub date data type in reporter.classic_item_list
As noticed by Ben Shum in IRC, reporter.classic_item_list
had pub_date, a free text field, classified as "int". This
changes that to "text" to prevent errors.
Signed-off-by: Chris Sharp <csharp@georgialibraries.org> Signed-off-by: Ben Shum <bshum@biblio.org>
Dan Scott [Mon, 30 Dec 2013 06:47:19 +0000 (01:47 -0500)]
TPAC: Display public hold notes in user holds list
If a given hold has a note marked as public, and not marked for staff, then
display the note in the My Account -> Holds list. This enhancement also
includes sample notes for each hold, to include data that ensures that the TPAC
does not display any hold notes that are not marked public, or which are marked
for staff.
Signed-off-by: Dan Scott <dscott@laurentian.ca> Signed-off-by: Lebbeous Fogle-Weekley <lebbeous@esilibrary.com>
projects / working / Evergreen.git / log