Implement ISBN lookup in SFX

Check the SFX knowledgebase for a matching ISBN and mark it via the
magical $9 SFX subfield so that we can key off of that for subsequent
queries via the catalogue.

Given that only one library may be running this script for a set of
ebook records, ensure that we're checking the results.

Signed-off-by: Dan Scott <dscott@laurentian.ca>

diff --git a/tools/ebooks/prep_ebook_records.py b/tools/ebooks/prep_ebook_records.py
index 92fd8d1..b18c548 100644 (file)
--- a/tools/ebooks/prep_ebook_records.py
+++ b/tools/ebooks/prep_ebook_records.py
@@ -26,6 +26,7 @@ class Institution():
             "ebrary_code": "algomauca", \
             "proxy": "http://libproxy.auc.ca/login?url=", \
             "link_text": "Available online", \
+            "sfx_url": "http://sfx.scholarsportal.info/algoma", \
             "access_note": "Access restricted to users with a valid Algoma University ID" \
         }
         
@@ -34,6 +35,7 @@ class Institution():
             "ebrary_code": "jndlu", \
             "proxy": "https://librweb.laurentian.ca/login?url=", \
             "link_text": "Available online / disponible en ligne", \
+            "sfx_url": "http://sfx.scholarsportal.info/laurentian", \
             "access_note": "Access restricted to users with a valid Laurentian University ID" \
         }
 
@@ -42,6 +44,7 @@ class Institution():
             "ebrary_code": "oculwindsor", \
             "proxy": "http://ezproxy.uwindsor.ca/login?url=", \
             "link_text": "To view Windsor's electronic resource click here.", \
+            "sfx_url": "http://sfx.scholarsportal.info/windsor", \
             "access_note": "Access restricted to users with a valid University of Windsor ID" \
         }
 
@@ -161,7 +164,7 @@ def check_options(options):
         print "* Missing -c / --consortium argument!"
         _help = True
 
-    if '--restriction' not in options:
+    if '--authorization' not in options:
         print "* Missing -a / --authorization argument!"
         _help = True
 
@@ -229,9 +232,11 @@ def check_libraries(options):
 def parse_opts():
     """Get command-line arguments from the script"""
     try:
-        _short_opts = 'i:o:p:ALWn:s:h'
-        _long_opts = ['input=', 'output=', 'publisher=', 'algoma', \
-                'laurentian', 'windsor', 'note=', 'sample=', 'help']
+        _short_opts = 'i:o:a:c:p:ALWn:s:h'
+        _long_opts = ['input=', 'output=', 'authorization=', 'consortium=', 
+            'publisher=', 'algoma', 'laurentian', 'windsor', 'note=',
+            'sample=', 'help'
+        ]
         opts = getopt.getopt(sys.argv[1:], _short_opts, _long_opts) 
     except getopt.GetoptError, ex:
         print "* %s" % str(ex)
@@ -291,6 +296,11 @@ def process_fields(record, options):
 
     add_publisher(record, new_record, options)
     add_restriction(new_record, options)
+    marked_isbn = mark_isbn_for_sfx(record, options)
+    if not marked_isbn:
+        print("No matching ISBN target found in SFX for %s" %
+            (new_record['856']['u'])
+        )
 
     if 'note' in options:
         note = pymarc.Field(tag = '590',
@@ -340,6 +350,82 @@ def add_publisher(record, new_record, options):
         )
         new_record.add_field(seven_ten)
 
+def mark_isbn_for_sfx(record, options):
+    """
+    Adds a $9 subfield to the 020 (ISBN) field to use for SFX look-ups
+
+    Assumes that the holdings in the SFX knowledgebase have been enabled
+    before the ebook processing script runs, or else we will not find any
+    matches.
+    """
+
+    for isbn in record.get_fields('020'):
+        for isbnval in isbn.get_subfields('a'):
+            isbnval = clean_isbn(isbnval)
+            sfx = 'http://sfx.scholarsportal.info/windsor'
+            # check to see if there are holdings in SFX knowledgebase
+            # use one of the participating libraries
+            if 'windsor' in options['libraries']:
+                sfx = options['settings'].get_settings('windsor')['sfx_url']
+            elif 'laurentian' in options['libraries']:
+                sfx = options['settings'].get_settings('laurentian')['sfx_url']
+            elif 'algoma' in options['libraries']:
+                sfx = options['settings'].get_settings('algoma')['sfx_url']
+            
+            url = "%s?url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&" \
+                "ctx_enc=UTF-8&ctx_ver=Z39.88-2004&rfr_id=info:sid/evergreen&" \
+                "sfx.ignore_date_threshold=1&" \
+                "sfx.response_type=multi_obj_detailed_xml" \
+                "&__service_type=getFullTxt&rft.isbn=%s" % (sfx, isbnval)
+
+            try:
+                req = urllib2.urlopen(url)
+                sfx_res = BeautifulSoup(req.read())
+            except urllib2.HTTPError, ex:
+                print("%s for URL %s" % (ex, url))
+                return None
+            except urllib2.URLError, ex:
+                print("%s for URL %s" % (ex, url))
+                return None
+        
+            # We want a target with a service_type element of 'getFullTxt'
+            targets = sfx_res.ctx_obj.ctx_obj_targets.findAll(
+                'target', recursive=False
+            )
+
+            if len(targets) == 0:
+                # No SFX targets found for this ISBN - next!
+                continue
+
+            for target in targets:
+                if target.service_type.renderContents() == 'getFullTxt':
+                    # Add the $9 subfield to mark this as a good one
+                    isbn.add_subfield('9', 'SFX')
+                    return True
+    return False
+
+def clean_isbn(isbn):
+    """
+    Return a normalized ISBN from a MARC subfield
+
+    Trims whitespace, removes hyphens, and removes trailing descriptions
+    like '(pbk)' and the like so that the ISBN can be reliably used in lookups
+    """
+
+    # Remove whitespace from both sides
+    isbn = isbn.strip()
+
+    # Grab the first string beginning with a digit
+    isbn_match = re.search(r'^[\D]*([\d]+\S+).*?$', isbn) 
+
+    if not isbn_match.group(1):
+        return None
+     
+    # Replace hyphens
+    isbn = isbn_match.group(1).replace('-', '');
+
+    return isbn
+
 def add_restriction(new_record, options):
     """
     Adds a 506 access restriction note per institution
@@ -527,5 +613,5 @@ def get_subfields(field, data):
 
 
 if __name__ == '__main__':
-
+    
     process_records(parse_opts())