BibMatch: using logical fields in querystrings

* Added the functionality to allow use of predefined logical fields,
  like 'author' and 'title', in querystrings. I.e -q 'title||author' will
  match with the logical fields instead of physical MARC tags in mind.
  (closes #202)

* Cleaned up and removed some unused variables.

* Updated documentation and added regression test.

modules/bibmatch/doc/admin/bibmatch-admin-guide.webdoc

@@ -57,11 +57,11 @@ To print potential duplicate entries before manual upload, run:
 </pre>
 </blockquote>
 
-To print undecided result from a match on multiple fields:
+To print undecided result from a match on multiple fields, including predefined fields (title, author etc.):
 
 <blockquote>
 <pre>
- $ bibmatch --print-ambiguous --query-string=\"245__a||100__a\" < input.xml > output.xml
+ $ bibmatch --print-ambiguous --query-string=\"245__a||author\" < input.xml > output.xml
 </pre>
 </blockquote>
 
@@ -146,5 +146,13 @@ Command line options:
  -a,  --alter-recid        The recid (controlfield 001) of matched or fuzzy matched records in
                            output will be replaced by the 001 value of the matched record.
                            Useful to prepare files to then be used with BibUpload.
+
+ Common predefined fields used in querystrings: (for Invenio demo site, your fields may vary!)
+
+ 'abstract', 'affiliation', 'anyfield', 'author', 'coden', 'collaboration',
+ 'collection', 'datecreated', 'datemodified', 'division', 'exactauthor',
+ 'experiment', 'fulltext', 'isbn', 'issn', 'journal', 'keyword', 'recid',
+ 'reference', 'reportnumber', 'subject', 'title', 'year'
+
 </pre>
 </blockquote>

modules/bibmatch/lib/bibmatch_engine.py

@@ -99,11 +99,12 @@ def usage():
                            output will be replaced by the 001 value of the matched record.
                            Useful to prepare files to then be used with BibUpload.
 
- Predefined fields:
+ Common predefined fields used in querystrings: (for Invenio demo site, your fields may vary!)
 
- The following predefined field names can be used:
-  "title"
-  "author"
+ 'abstract', 'affiliation', 'anyfield', 'author', 'coden', 'collaboration',
+ 'collection', 'datecreated', 'datemodified', 'division', 'exactauthor',
+ 'experiment', 'fulltext', 'isbn', 'issn', 'journal', 'keyword', 'recid',
+ 'reference', 'reportnumber', 'subject', 'title', 'year'
 
  Examples:
 
@@ -144,7 +145,6 @@ def __init__(self, mode="1"):
         self.format.append([])
         self.format.append([])
         self.format.append([])
-        self.advanced = 0
         return
 
     def from_qrystr(self, qrystr="", search_mode="eee", operator="aa"):
@@ -204,7 +204,6 @@ def default(self):
         self.format.append([])
         self.format.append([])
         self.format.append([])
-        self.advanced = 1
         return
 
     def change_search_mode(self, mode="a"):
@@ -367,6 +366,11 @@ def match_records(records, qrystrs=None, perform_request_search_mode="eee", \
 
             ### get appropriate fields from database
             for field in querystring.field:
+                tags = get_field_tags(field)
+                if len(tags) > 0:
+                    # Fetch value from input record of first tag only
+                    # FIXME: Extracting more then first tag, evaluating each
+                    field = tags[0]
                 ### use expanded tags
                 tag  = field[0:3]
                 ind1 = field[3:4]
@@ -416,7 +420,6 @@ def match_records(records, qrystrs=None, perform_request_search_mode="eee", \
                 p3 = inst[2]
                 f3 = querystring.field[2]
                 m3 = querystring.mode[2]
-                aas = querystring.advanced
 
                 #1st run the basic perform_req_search
                 recID_list = server.search(
@@ -570,8 +573,7 @@ def main():
     batch_output = ""               #print stuff in files
     f_input = ""                    #read from where, if param "i"
     server_url = CFG_SITE_URL       #url to server performing search, local by default
-    modify = 0                      #alter output with matched record indentifiers
-    predefined_fields = ["title", "author"]
+    modify = 0                      #alter output with matched record identifiers
     textmarc_output = 0
 
     for opt, opt_value in opts:
@@ -612,9 +614,8 @@ def main():
         if opt in ["-a", "--alter-recid"]:
             modify = 1
         if opt in ["-f", "--field"]:
-            alternate_querystring = []
-            if opt_value in predefined_fields:
-                alternate_querystring = get_field_tags(opt_value)
+            alternate_querystring = get_field_tags(opt_value)
+            if len(alternate_querystring) > 0:
                 for item in alternate_querystring:
                     qrystrs.append(item)
             else:

modules/bibmatch/lib/bibmatch_regression_tests.py

@@ -791,6 +791,13 @@ def test_check_altered(self):
         [dummy1, matchedrecs, dummy3, dummy4] = match_records(records, modify=1)
         self.assertTrue(record_has_field(matchedrecs[0][0], '001'))
 
+    def test_check_qrystr(self):
+        """bibmatch - check querystrings"""
+        qrystrs = ["author||reportnumber"]
+        records = create_records(self.recxml1)
+        [dummy1, matchedrecs, dummy3, dummy4] = match_records(records, qrystrs=qrystrs)
+        self.assertEqual(1,len(matchedrecs))
+
 TEST_SUITE = make_test_suite(BibMatchTest)
 
 if __name__ == "__main__":