def ia_match(a, ia):
try:
loc, rec = get_ia(ia)
except urllib.error.HTTPError:
return False
if rec is None or 'full_title' not in rec:
return False
try:
e1 = build_marc(rec)
except TypeError:
print(rec)
raise
return amazon_merge.attempt_merge(a, e1, threshold, debug=False)
continue
try:
rec = fast_parse.read_edition(marc_data,
accept_electronic=True)
except:
bad_binary = "MARC parse error"
if bad_binary and not formats['xml']:
load_error_mail(ia, bad_binary, 'bad MARC binary, no MARC XML')
continue
if not use_binary and formats['xml']:
if bad_ia_xml(ia) and bad_binary:
load_error_mail(ia, bad_binary,
'bad MARC binary, bad MARC XML')
continue
try:
rec = get_ia(ia)
except (KeyboardInterrupt, NameError):
raise
except NoMARCXML:
write_log(ia, when, "no MARCXML")
continue
except urllib2.HTTPError as error:
write_log(ia, when, "error: HTTPError: " + str(error))
continue
if not use_binary and not formats['xml']:
print('skipping, no MARC')
continue
if not rec:
write_log(ia, when, "error: no rec")
continue
iter = db.query(
"select identifier, updated from metadata where contributor='Cornell University Library' and scanner is not null and noindex is null and mediatype='texts' and (curatestate='approved' or curatestate is null) and scandate is not null order by updated",
{'start': start})
t_start = time()
for row in iter:
ia = row.identifier
print((repr(ia), row.updated))
when = str(row.updated)
if query({'type': '/type/edition', 'ocaid': ia}):
print('already loaded')
continue
if query({'type': '/type/edition', 'source_records': 'ia:' + ia}):
print('already loaded')
continue
try:
loc, rec = get_ia(ia)
except (KeyboardInterrupt, NameError):
raise
except NoMARCXML:
write_log(ia, when, "no MARCXML")
continue
except urllib2.HTTPError as error:
write_log(ia, when, "error: HTTPError: " + str(error))
continue
if loc is None:
write_log(ia, when, "error: no loc ")
if rec is None:
write_log(ia, when, "error: no rec")
continue
print(loc, rec)
if str(marc_data)[6:8] != 'am': # only want books
print 'not a book!'
continue
try:
rec = fast_parse.read_edition(marc_data, accept_electronic = True)
except:
bad_binary = "MARC parse error"
if bad_binary and not formats['xml']:
load_error_mail(ia, bad_binary, 'bad MARC binary, no MARC XML')
continue
if not use_binary and formats['xml']:
if bad_ia_xml(ia) and bad_binary:
load_error_mail(ia, bad_binary, 'bad MARC binary, bad MARC XML')
continue
try:
rec = get_ia(ia)
except (KeyboardInterrupt, NameError):
raise
except NoMARCXML:
write_log(ia, when, "no MARCXML")
continue
except urllib2.HTTPError as error:
write_log(ia, when, "error: HTTPError: " + str(error))
continue
if not use_binary and not formats['xml']:
print 'skipping, no MARC'
continue
if not rec:
write_log(ia, when, "error: no rec")
continue
#iter = db.query("select identifier, updated from metadata where scanner is not null and noindex is null and mediatype='texts' and (curatestate='approved' or curatestate is null) and scandate is not null and updated > $start order by updated", {'start': start})
iter = db.query("select identifier, updated from metadata where contributor='Cornell University Library' and scanner is not null and noindex is null and mediatype='texts' and (curatestate='approved' or curatestate is null) and scandate is not null order by updated", {'start': start})
t_start = time()
for row in iter:
ia = row.identifier
print `ia`, row.updated
when = str(row.updated)
if query({'type': '/type/edition', 'ocaid': ia}):
print 'already loaded'
continue
if query({'type': '/type/edition', 'source_records': 'ia:' + ia}):
print 'already loaded'
continue
try:
loc, rec = get_ia(ia)
except (KeyboardInterrupt, NameError):
raise
except NoMARCXML:
write_log(ia, when, "no MARCXML")
continue
except urllib2.HTTPError as error:
write_log(ia, when, "error: HTTPError: " + str(error))
continue
if loc is None:
write_log(ia, when, "error: no loc ")
if rec is None:
write_log(ia, when, "error: no rec")
continue
print loc, rec
