def translate(bytes_in, leader_says_marc8=False):
"""
Converts MARC8 to unicode
"""
marc8 = MARC8ToUnicode(quiet=True)
if leader_says_marc8:
data = marc8.translate(mnemonics.read(bytes_in))
else:
data = bytes_in.decode('utf-8')
return normalize('NFC', data)
def translate(bytes_in, leader_says_marc8=False):
marc8 = MARC8ToUnicode(quiet=True)
try:
if leader_says_marc8:
data = marc8.translate(mnemonics.read(bytes_in))
else:
data = bytes_in.decode('utf-8')
return normalize('NFC', data)
except:
print('translate error for:', repr(bytes_in))
print('marc8:', leader_says_marc8)
raise
def translate(bytes_in, leader_says_marc8=False):
"""
Converts a binary MARC field value to unicode str,
from either MARC8 or UTF8 encoded bytes.
:param bytes_in bytes:
:rtype: str
"""
assert isinstance(bytes_in, bytes)
marc8 = MARC8ToUnicode(quiet=True)
if leader_says_marc8:
data = marc8.translate(mnemonics.read(bytes_in))
else:
data = bytes_in.decode('utf-8')
return normalize('NFC', data)
def record_to_xml_node(self, record, quiet=False, namespace=False):
"""
converts a record object to a chunk of xml
# include the marcxml namespace in the root tag (default: False)
record_to_xml(record, namespace=True)
"""
# helper for converting non-unicode data to unicode
# TODO: maybe should set g0 and g1 appropriately using 066 $a and $b?
prefix = 'marc:'
marc8 = MARC8ToUnicode(quiet=quiet)
def translate(data):
if type(data) == six.text_type:
return data
else:
return marc8.translate(data)
root = ET.Element(prefix + 'record')
if namespace:
root.set('xmlns', MARC_XML_NS)
root.set('xmlns:xsi', XSI_NS)
root.set('xsi:schemaLocation', MARC_XML_SCHEMA)
leader = ET.SubElement(root, prefix + 'leader')
leader.text = record.leader
for field in record:
if field.is_control_field():
control_field = ET.SubElement(root, prefix + 'controlfield')
control_field.set('tag', field.tag)
control_field.text = translate(field.data)
else:
data_field = ET.SubElement(root, prefix + 'datafield')
data_field.set('tag', field.tag)
data_field.set('ind1', field.indicators[0])
data_field.set('ind2', field.indicators[1])
for subfield in field:
data_subfield = ET.SubElement(data_field,
prefix + 'subfield')
data_subfield.set('code', subfield[0])
data_subfield.text = translate(subfield[1])
return root
def record_to_xml_node(record, quiet=False, namespace=False):
"""Converts a record object to a chunk of XML.
If you would like to include the marcxml namespace in the root tag set namespace to
True.
"""
# helper for converting non-unicode data to unicode
# TODO: maybe should set g0 and g1 appropriately using 066 $a and $b?
marc8 = MARC8ToUnicode(quiet=quiet)
def translate(data):
if type(data) == str:
return data
else:
return marc8.translate(data)
root = ET.Element("record")
if namespace:
root.set("xmlns", MARC_XML_NS)
root.set("xmlns:xsi", XSI_NS)
root.set("xsi:schemaLocation", MARC_XML_SCHEMA)
leader = ET.SubElement(root, "leader")
leader.text = str(record.leader)
for field in record:
if field.is_control_field():
control_field = ET.SubElement(root, "controlfield")
control_field.set("tag", field.tag)
control_field.text = translate(field.data)
else:
data_field = ET.SubElement(root, "datafield")
data_field.set("ind1", field.indicators[0])
data_field.set("ind2", field.indicators[1])
data_field.set("tag", field.tag)
for subfield in field:
data_subfield = ET.SubElement(data_field, "subfield")
data_subfield.set("code", subfield[0])
data_subfield.text = translate(subfield[1])
return root
from pymarc import MARC8ToUnicode
from unicodedata import normalize
from openlibrary.catalog.marc import mnemonics
from openlibrary.catalog.marc.marc_base import MarcBase, MarcException, BadMARC
marc8 = MARC8ToUnicode(quiet=True)
class BadLength(MarcException):
pass
def handle_wrapped_lines(_iter):
"""
Handles wrapped MARC fields, which appear to be multiple
fields with the same field number ending with ++
Have not found an official spec which describe this.
"""
cur_lines = []
cur_tag = None
maybe_wrap = False
for t, l in _iter:
if len(l) > 500 and l.endswith(b'++\x1e'):
assert not cur_tag or cur_tag == t
cur_tag = t
cur_lines.append(l)
continue
if cur_lines:
yield cur_tag, cur_lines[0][:-3] + b''.join(
i[2:-3] for i in cur_lines[1:]) + l[2:]