From 20a0e8388235730542b66e62c10d5f0ab95f7f42 Mon Sep 17 00:00:00 2001
From: dbs <dbs@6d9bc8c9-1ec2-4278-b937-99fde70a366f>
Date: Thu, 25 Nov 2010 04:22:46 +0000
Subject: [PATCH] And now we handle ebrary records per our spec (no proxy,
 substitute channel)

git-svn-id: svn://svn.open-ils.org/ILS-Contrib/conifer/branches/rel_1_6_1@1086 6d9bc8c9-1ec2-4278-b937-99fde70a366f
---
 tools/ebooks/prep_ebook_records.py | 40 +++++++++++++++++++++++++++++++-------
 1 file changed, 33 insertions(+), 7 deletions(-)

diff --git a/tools/ebooks/prep_ebook_records.py b/tools/ebooks/prep_ebook_records.py
index 5fe7c1b846..7f630cb12d 100644
--- a/tools/ebooks/prep_ebook_records.py
+++ b/tools/ebooks/prep_ebook_records.py
@@ -13,7 +13,7 @@ requirements that would be the same for each record and therefore can
 be accommodated in batch load.
 """
 
-import os, os.path, sys, getopt, pymarc, pymarc.marc8
+import os, os.path, sys, getopt, pymarc, pymarc.marc8, re
 
 class Institution():
     """Defines standard settings for each Conifer institution"""
@@ -262,21 +262,47 @@ def process_urls(field, options):
         print "* No subfield 'u' found in this 856"
         return None
 
+
     for lib in options['libraries']:
         data = options['settings'].get_settings(lib)
+        subs = get_subfields(field, data)
         eight_five_six = pymarc.Field(tag = '856',
             indicators = ['4', '0'],
-            subfields = [
-                'u', data['proxy'] + field['u'],
-                'y', data['link_text'],
-                'z', data['public_note'],
-                '9', data['code']
-            ]
+            subfields = subs 
         )
         new_fields.append(eight_five_six)
 
     return new_fields
 
+def get_subfields(field, data):
+    """Creates 856 subfields required by Conifer"""
+
+    subs = []
+    url = field['u']
+
+    # Is this an ebrary URL?
+    ebrary = False
+    if url.find('.ebrary.com') > -1:
+        ebrary = True
+        
+    # ebrary URLs look like: http://site.ebrary.com/lib/<channel>/Doc?id=2001019
+    # we need to replace <channel> with the library-specific channel
+    if ebrary:
+        ebrary_url = re.search(r'^(.+?/lib/).+?(/.+?)$', url) 
+        url = ebrary_url.group(1) + data['ebrary_code'] + ebrary_url.group(2)
+        subs.extend(['u', url])
+    else:
+        subs.extend(['u', data['proxy'] + field['u']])
+
+    subs.extend([
+            'y', data['link_text'],
+            'z', data['public_note'],
+            '9', data['code']
+    ])
+
+    return subs
+
+
 if __name__ == '__main__':
 
     process_records(parse_opts())
-- 
2.11.0