def record_xml_output(rec, tags=None, order_fn=None):
"""Generates the XML for record 'rec' and returns it as a string
@rec: record
@tags: list of tags to be printed"""
if tags is None:
tags = []
if isinstance(tags, str):
tags = [tags]
marcxml = []
# Add the tag 'tag' to each field in rec[tag]
fields = []
if rec is not None:
for tag in rec:
if not tags or tag in tags:
for field in rec[tag]:
if tag == '773':
field = ([subfield for subfield in field[0] if subfield[0] != 'c'], field[1], field[2], field[3], field[4])
fields.append((tag, field))
if order_fn is None:
record_order_fields(fields)
else:
record_order_fields(fields, order_fn)
for field in fields:
marcxml.append(field_xml_output(field[1], field[0]))
return '\n'.join(marcxml)
def marcxml_filter_out_tags(recid, fields):
"""
Returns the fields of record 'recid' that share the same tag and
indicators as those specified in 'fields', but for which the
subfield is different. This is nice to emulate a bibupload -c that
corrects only specific subfields.
Parameters:
recid - *int* the id of the record to process
fields - *list(str)* the list of fields that we want to filter
out. Eg ['909COp', '909COo']
"""
out = ''
record = get_record(recid)
# Delete subfields that we want to replace
for field in fields:
record_delete_subfield(record,
tag=field[0:3],
ind1=field[3:4],
ind2=field[4:5],
subfield_code=field[5:6])
# Select only datafields that share tag + indicators
processed_tags_and_ind = []
for field in fields:
if not field[0:5] in processed_tags_and_ind:
# Ensure that we do not process twice the same datafields
processed_tags_and_ind.append(field[0:5])
for datafield in record.get(field[0:3], []):
if datafield[1] == field[3:4].replace('_', ' ') and \
datafield[2] == field[4:5].replace('_', ' ') and \
datafield[0]:
out += field_xml_output(datafield, field[0:3]) + '\n'
return out
def marcxml_filter_out_tags(recid, fields):
"""
Returns the fields of record 'recid' that share the same tag and
indicators as those specified in 'fields', but for which the
subfield is different. This is nice to emulate a bibupload -c that
corrects only specific subfields.
Parameters:
recid - *int* the id of the record to process
fields - *list(str)* the list of fields that we want to filter
out. Eg ['909COp', '909COo']
"""
out = ''
record = get_record(recid)
# Delete subfields that we want to replace
for field in fields:
record_delete_subfield(record,
tag=field[0:3],
ind1=field[3:4],
ind2=field[4:5],
subfield_code=field[5:6])
# Select only datafields that share tag + indicators
processed_tags_and_ind = []
for field in fields:
if not field[0:5] in processed_tags_and_ind:
# Ensure that we do not process twice the same datafields
processed_tags_and_ind.append(field[0:5])
for datafield in record.get(field[0:3], []):
if datafield[1] == field[3:4].replace('_', ' ') and \
datafield[2] == field[4:5].replace('_', ' ') and \
datafield[0]:
out += field_xml_output(datafield, field[0:3]) + '\n'
return out
def create_MARC(extracted_image_data, tarball, refno):
"""
Take the images and their captions and the name of the associated TeX
file and build a MARCXML record for them.
@param: extracted_image_data ([(string, string, list, list), ...]):
a list of tuples of images matched to labels, captions and contexts from
this document.
@param: refno (string): the name for the record number field, or None
@output: a MARCXML record detailing all the arguments as appropriate
at tarball.insert.xml and a duplicate one at tarball.correct.xml
@return: the path to the MARCXML record, None if no plots
"""
root_dir = os.path.dirname(tarball) + os.sep + os.path.basename(tarball) + \
'_plots' + os.sep
# For building result MARCXML
marcxml = ['<record>']
# Datafield := (subfields, ind1, ind2, controlfield)
# Subfield := (code, value)
#FIXME: Determine what to do without refno
if refno and refno.isdigit():
field = (None, ' ', ' ', refno)
marcxml.append(field_xml_output(field, '001'))
index = 0
for (image_location, caption, dummy, contexts) in extracted_image_data:
if image_location == '':
# we don't know the image, but the captions are for separate things
for cap in caption.split(' : '):
# Add DUMMY-PLOT MARCXML per loose captions
subfields = []
subfields.append(('a', DUMMY_IMAGE_TMP))
subfields.append(('t', "PlotMisc"))
subfields.append(('d', "%05d %s" % (index, cap)))
subfields.append(('n', "fig%05d" % (index,)))
subfields.append(('o', "HIDDEN"))
marcxml.append(field_xml_output((subfields, ' ', ' ', None), "FFT"))
index = index + 1
else:
# Merge subfolder into docname, until root directory
relative_image_path = image_location.replace(root_dir, '')
docname = "_".join(relative_image_path.split('.')[:-1]).replace('/', '_').replace(';', '').replace(':', '')
if len(caption) < 3:
subfields = []
subfields.append(('a', image_location))
subfields.append(('t', "PlotMisc"))
subfields.append(('d', "%05d %s" % (index, caption.replace(' : ', ''))))
subfields.append(('n', docname))
subfields.append(('o', "HIDDEN"))
marcxml.append(field_xml_output((subfields, ' ', ' ', None), "FFT"))
else:
# Add PLOT MARCXML
subfields = []
subfields.append(('a', image_location))
subfields.append(('t', "Plot"))
subfields.append(('d', "%05d %s" % (index, caption.replace(' : ', ''))))
subfields.append(('n', docname))
marcxml.append(field_xml_output((subfields, ' ', ' ', None), "FFT"))
if contexts:
# Add CONTEXT MARCXML
subfields = []
subfields.append(('a', "%s.context" % (image_location,)))
subfields.append(('t', "Plot"))
subfields.append(('f', ".png;context"))
subfields.append(('n', docname))
subfields.append(('o', "HIDDEN"))
marcxml.append(field_xml_output((subfields, ' ', ' ', None), "FFT"))
index = index + 1
marcxml.append('</record>')
return '\n'.join(marcxml)
def create_MARC(extracted_image_data, tarball, refno):
"""
Take the images and their captions and the name of the associated TeX
file and build a MARCXML record for them.
@param: extracted_image_data ([(string, string, list, list), ...]):
a list of tuples of images matched to labels, captions and contexts from
this document.
@param: refno (string): the name for the record number field, or None
@output: a MARCXML record detailing all the arguments as appropriate
at tarball.insert.xml and a duplicate one at tarball.correct.xml
@return: the path to the MARCXML record, None if no plots
"""
root_dir = os.path.dirname(tarball) + os.sep + os.path.basename(tarball) + \
'_plots' + os.sep
marcxml_fft = []
index = 0
for (image_location, caption, dummy, contexts) in extracted_image_data:
if len(image_location) < 3:
# If not useful URL -> move on to next
continue
# Merge subfolder into docname, until root directory
relative_image_path = image_location.replace(root_dir, '')
docname = "_".join(relative_image_path.split('.')[:-1]).replace(
'/', '_').replace(';', '').replace(':', '')
if type(caption) == list:
caption = " ".join(caption)
if len(caption) < 3:
subfields = []
subfields.append(('a', image_location))
subfields.append(('t', "PlotMisc"))
subfields.append(
('d', "%05d %s" % (index, caption.replace(' : ', ''))))
subfields.append(('n', docname))
subfields.append(('o', "HIDDEN"))
marcxml_fft.append(
field_xml_output((subfields, ' ', ' ', None), "FFT"))
else:
# Add PLOT MARCXML
subfields = []
subfields.append(('a', image_location))
subfields.append(('t', "Plot"))
subfields.append(
('d', "%05d %s" % (index, caption.replace(' : ', ''))))
subfields.append(('n', docname))
marcxml_fft.append(
field_xml_output((subfields, ' ', ' ', None), "FFT"))
if contexts:
# Add CONTEXT MARCXML
subfields = []
subfields.append(('a', "%s.context" % (image_location, )))
subfields.append(('t', "Plot"))
subfields.append(('f', ".png;context"))
subfields.append(('n', docname))
subfields.append(('o', "HIDDEN"))
marcxml_fft.append(
field_xml_output((subfields, ' ', ' ', None), "FFT"))
index += 1
if marcxml_fft:
# For building result MARCXML
marcxml_header = ['<record>']
# Datafield := (subfields, ind1, ind2, controlfield)
# Subfield := (code, value)
#FIXME: Determine what to do without refno
if refno and refno.isdigit():
field = (None, ' ', ' ', refno)
marcxml_header.append(field_xml_output(field, '001'))
marcxml = marcxml_header + marcxml_fft
marcxml.append('</record>')
return '\n'.join(marcxml)
return ""
def update_record(record_id, authors):
"""Update authors in CDS record.
:param int record_id: record to update author datafields
Example:
record_id = 2150939
:param dict authors: dictionary where keys are author full names and
values the CDS profile ids to be updated in the given record
Example:
authors = {'Ellis, John': '2108556'}
:return: string representing the record XML element containing
author (`100`) and/or co-author (`700`) datafields. Empty string if
nothing to update
Example:
'<record>
<controlfield tag="001">2150939</controlfield>
<datafield tag="100" ind1=" " ind2=" ">
<subfield code="a">Ellis, John</subfield>
<subfield code="u">King's Coll. London</subfield>
<subfield code="u">CERN</subfield>
<subfield code="0">AUTHOR|(CDS)2108556</subfield>
<subfield code="9">#BEARD#</subfield>
</datafield>
</record>'
"""
record = get_record(record_id)
record_author = record_get_field_instances(record, "100")
record_coauthors = record_get_field_instances(record, "700")
if len(record_author) > 1:
print ("Oops: several '100' (main author) fields have been found in "
"record '{0}'".format(record_id))
return ""
datafields = ""
author = False
for author_field in record_author:
try:
author_name = field_get_subfield_values(author_field, 'a')[0]
try:
cds_id = authors[author_name]
if extend_author_field(author_field, cds_id):
datafields += field_xml_output(author_field, "100")
author = True
except KeyError:
pass
except IndexError:
# Author field (`100`) does not have subfield `a`
pass
if len(authors) > 1 or not author:
for coauthor_field in record_coauthors:
try:
coauthor_name = field_get_subfield_values(
coauthor_field, 'a')[0]
try:
cds_id = authors[coauthor_name]
if extend_author_field(coauthor_field, cds_id):
author = True
except KeyError:
pass
except IndexError:
# Co-author field (`700`) does not have subfield `a`
pass
datafields += field_xml_output(coauthor_field, "700")
# Nothing to update
if not author:
# print "No authors to update in record '{0}'".format(record_id)
return ""
record = ('<record><controlfield tag="001">{0}</controlfield>{1}'
'</record>'.format(record_id, datafields))
return record
def create_MARC(extracted_image_data, tarball, refno):
"""
Take the images and their captions and the name of the associated TeX
file and build a MARCXML record for them.
@param: extracted_image_data ([(string, string, list, list), ...]):
a list of tuples of images matched to labels, captions and contexts from
this document.
@param: refno (string): the name for the record number field, or None
@output: a MARCXML record detailing all the arguments as appropriate
at tarball.insert.xml and a duplicate one at tarball.correct.xml
@return: the path to the MARCXML record, None if no plots
"""
root_dir = os.path.dirname(tarball) + os.sep + os.path.basename(tarball) + "_plots" + os.sep
marcxml_fft = []
index = 0
for (image_location, caption, dummy, contexts) in extracted_image_data:
if len(image_location) < 3:
# If not useful URL -> move on to next
continue
# Merge subfolder into docname, until root directory
relative_image_path = image_location.replace(root_dir, "")
docname = "_".join(relative_image_path.split(".")[:-1]).replace("/", "_").replace(";", "").replace(":", "")
if type(caption) == list:
caption = " ".join(caption)
if len(caption) < 3:
subfields = []
subfields.append(("a", image_location))
subfields.append(("t", "PlotMisc"))
subfields.append(("d", "%05d %s" % (index, caption.replace(" : ", ""))))
subfields.append(("n", docname))
subfields.append(("o", "HIDDEN"))
marcxml_fft.append(field_xml_output((subfields, " ", " ", None), "FFT"))
else:
# Add PLOT MARCXML
subfields = []
subfields.append(("a", image_location))
subfields.append(("t", "Plot"))
subfields.append(("d", "%05d %s" % (index, caption.replace(" : ", ""))))
subfields.append(("n", docname))
marcxml_fft.append(field_xml_output((subfields, " ", " ", None), "FFT"))
if contexts:
# Add CONTEXT MARCXML
subfields = []
subfields.append(("a", "%s.context" % (image_location,)))
subfields.append(("t", "Plot"))
subfields.append(("f", ".png;context"))
subfields.append(("n", docname))
subfields.append(("o", "HIDDEN"))
marcxml_fft.append(field_xml_output((subfields, " ", " ", None), "FFT"))
index += 1
if marcxml_fft:
# For building result MARCXML
marcxml_header = ["<record>"]
# Datafield := (subfields, ind1, ind2, controlfield)
# Subfield := (code, value)
# FIXME: Determine what to do without refno
if refno and refno.isdigit():
field = (None, " ", " ", refno)
marcxml_header.append(field_xml_output(field, "001"))
marcxml = marcxml_header + marcxml_fft
marcxml.append("</record>")
return "\n".join(marcxml)
return ""