jeff

Normalizes values such as:

" 9999999999"
"(999) 999-9999"
"(999) 9999999"
"(999)999-9999"
"9-(999)-999-9999"
"9-999-999-9999"
"999 999 9999"
"999 999-9999"

jeff

%%%
%%%     Debian ejabberd configuration file
%%%     This config must be in UTF-8 encoding
%%%
%%% The parameters used in this configuration file are explained in more detail
%%% in the ejabberd Installation and Operation Guide.
%%% Please consult the Guide in case of doubts, it is available at
%%% /usr/share/doc/ejabberd/guide.html

%%% This configuration file contains Erlang terms.

jeff

#!/usr/bin/env python

"""
This is a scraper for the LoC website that will parse the human
readable HTML documentation for MARC tags, and generate machine
readable (XML) documentation.

You'll need to have python, elementtree [1] and elementtidy [2] installed
before running this script.

jeff

-- queue bre records for a batch re-ingest
-- select bres with a title containing the string "'s "
-- which do not have a keyword index_vector
-- matching 's'
INSERT INTO tadl_batch.bre (bre,batch_name)
WITH apos_recs AS (select distinct mkfe.source id, mkfe.value title
from metabib.keyword_field_entry mkfe
join asset.call_number acn on (acn.record = mkfe.source and not acn.deleted)
join asset.copy acp on (acp.call_number = acn.id and not acp.deleted)
where mkfe.field = 122

jeff

-- This will show counts grouped by weeks, but will
-- likely include a partial week at the start
-- (check the "days_in_sample" column)
SELECT date_trunc('week',xact_start)::date week,
circ_lib,
count(distinct xact_start::date) days_in_sample,
count(*) circ_count
FROM action.circulation
WHERE xact_start > '2012-12-01'
GROUP BY 1,2

jeff

# Generated by iptables-save v1.4.2 on Fri Jan  4 13:40:19 2013
*filter
:INPUT ACCEPT [0:0]
:FORWARD ACCEPT [0:0]
:OUTPUT ACCEPT [0:0]
:techs - [0:0]
-A INPUT -i lo -j ACCEPT 
-A INPUT -p tcp -m tcp --dport 22 -m state --state RELATED -j ACCEPT
-A INPUT -p tcp -m tcp --dport 22 -j techs 
-A INPUT -p tcp -m tcp --dport 22 -j LOG 

jeff

-- any way to write this so that i have 'c: donation' in only one place?

BEGIN;
UPDATE value.items
SET is_donated = true
WHERE id IN (SELECT copy FROM value.copynotes WHERE content = ': donation');
DELETE FROM value.copynotes
WHERE content = 'c: donation';
COMMIT;

jeff

-- revised to use a sub-select and to report only those records with BOTH a
-- matching 856$3 and 856$9 -- this sub-select may not be the most efficient way
-- to do this. also eliminated tag/subfield/value from the output
SELECT au.usrname, bre.tcn_value, bre.id, bre.create_date, 
bre.edit_date
FROM metabib.real_full_rec as mrfr
LEFT JOIN biblio.record_entry as bre on bre.id = mrfr.record
LEFT JOIN actor.usr as au on au.id = bre.editor
WHERE bre.deleted = 'f'
AND mrfr.tag = '856'

jeff

#!/usr/bin/perl

use strict; use warnings;

while (<>) {
    chomp;
    my $bc = $_;
    $bc =~ s/"//g;
    next if (length($bc) != 14);
    next if ($bc !~ m/^1\d+$/);

jeff

Enabling OpenILS::WWW::AddedContent to return cover images based on more than just ISBN

Supporting legacy without making the New Awesome have to bend over backward

* make AddedContent superclass look for a use-time argument to enable "legacy" lookup-by-isbn support

* change apache handler to use legacy option
* create new apache handler at new url to use new default of lookup-by-record-id
for records: AC_HOST/opac/extras/ac/jacket/small/r/RECORD_ID
for metarecords: AC_HOST/opac/extras/ac/jacket/small/m/METARECORD_ID