def build_record(counts, fields, recid=None, status_code=0):
"""Given a series of MARC XML-ized reference lines and a record-id, write a
MARC XML record to the stdout stream. Include in the record some stats
for the extraction job.
The printed MARC XML record will essentially take the following
structure:
<record>
<controlfield tag="001">1</controlfield>
<datafield tag="999" ind1="C" ind2="5">
[...]
</datafield>
[...]
<datafield tag="999" ind1="C" ind2="6">
<subfield code="a">
Invenio/X.XX.X refextract/X.XX.X-timestamp-err-repnum-title-URL-misc
</subfield>
</datafield>
</record>
Timestamp, error(code), reportnum, title, URL, and misc will are of
course take the relevant values.
@param status_code: (integer)the status of reference-extraction for the
given record: was there an error or not? 0 = no error; 1 = error.
@param count_reportnum: (integer) - the number of institutional
report-number citations found in the document's reference lines.
@param count_title: (integer) - the number of journal title citations
found in the document's reference lines.
@param count_url: (integer) - the number of URL citations found in the
document's reference lines.
@param count_misc: (integer) - the number of sections of miscellaneous
text (i.e. 999C5$m) from the document's reference lines.
@param count_auth_group: (integer) - the total number of author groups
identified ($h)
@param recid: (string) - the record-id of the given document. (put into
001 field.)
@param xml_lines: (list) of strings. Each string in the list contains a
group of MARC XML 999C5 datafields, making up a single reference line.
These reference lines will make up the document body.
@return: The entire MARC XML textual output, plus recognition statistics.
"""
record = BibRecord(recid=recid)
record['999'] = fields
field = record.add_field(CFG_REFEXTRACT_TAG_ID_EXTRACTION_STATS)
stats_str = "%(status)s-%(reportnum)s-%(title)s-%(author)s-%(url)s-%(doi)s-%(misc)s" % {
'status' : status_code,
'reportnum' : counts['reportnum'],
'title' : counts['title'],
'author' : counts['auth_group'],
'url' : counts['url'],
'doi' : counts['doi'],
'misc' : counts['misc'],
}
field.add_subfield(CFG_REFEXTRACT_SUBFIELD_EXTRACTION_STATS,
stats_str)
field.add_subfield(CFG_REFEXTRACT_SUBFIELD_EXTRACTION_TIME,
datetime.now().strftime("%Y-%m-%d %H:%M:%S"))
field.add_subfield(CFG_REFEXTRACT_SUBFIELD_EXTRACTION_VERSION,
CFG_REFEXTRACT_VERSION)
return record
def test_simple(self):
from invenio.legacy.docextract.record import BibRecord
from invenio.legacy.docextract.convert_journals import convert_journals
record = BibRecord()
record.add_subfield('100__a', 'Test Journal Name')
record.add_subfield('773__p', 'Test Journal Name')
record.add_subfield('999C5s', 'Test Journal Name,100,10')
converted_record = convert_journals(self.kb, record)
expected_record = BibRecord()
expected_record.add_subfield('100__a', 'Test Journal Name')
expected_record.add_subfield('773__p', 'Converted')
expected_record.add_subfield('999C5s', 'Converted,100,10')
self.assertEqual(expected_record, converted_record)
def test_simple(self):
from invenio.legacy.docextract.record import BibRecord
from invenio.legacy.docextract.convert_journals import convert_journals
record = BibRecord()
record.add_subfield('100__a', 'Test Journal Name')
record.add_subfield('773__p', 'Test Journal Name')
record.add_subfield('999C5s', 'Test Journal Name,100,10')
converted_record = convert_journals(self.kb, record)
expected_record = BibRecord()
expected_record.add_subfield('100__a', 'Test Journal Name')
expected_record.add_subfield('773__p', 'Converted')
expected_record.add_subfield('999C5s', 'Converted,100,10')
self.assertEqual(expected_record, converted_record)
