def test_dmdsec_from_csv_parsed_metadata_no_data(self):
"""It should not create dmdSecs with no parsed metadata."""
data = {}
# Test
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(Job("stub", "stub", []), data)
# Verify
assert ret == []
def test_dmdsec_from_csv_parsed_metadata_other_only(self):
"""It should only create an Other dmdSec from parsed metadata."""
data = collections.OrderedDict([
("Title", ["Yamani Weapons"]),
("Contributor", [u"雪 ユキ".encode('utf8')]),
("Long Description", ['This is about how glaives are used in the Yamani Islands'])
])
# Test
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(Job("stub", "stub", []), data)
# Verify
assert ret
assert len(ret) == 1
dmdsec = ret[0]
assert dmdsec.tag == '{http://www.loc.gov/METS/}dmdSec'
assert 'ID' in dmdsec.attrib
mdwrap = dmdsec[0]
assert mdwrap.tag == '{http://www.loc.gov/METS/}mdWrap'
assert 'MDTYPE' in mdwrap.attrib
assert mdwrap.attrib['MDTYPE'] == 'OTHER'
assert 'OTHERMDTYPE' in mdwrap.attrib
assert mdwrap.attrib['OTHERMDTYPE'] == 'CUSTOM'
xmldata = mdwrap[0]
assert xmldata.tag == '{http://www.loc.gov/METS/}xmlData'
# Elements are direct children of xmlData
assert len(xmldata) == 3
assert xmldata[0].tag == 'title'
assert xmldata[0].text == 'Yamani Weapons'
assert xmldata[1].tag == 'contributor'
assert xmldata[1].text == u'雪 ユキ'
assert xmldata[2].tag == 'long_description'
assert xmldata[2].text == 'This is about how glaives are used in the Yamani Islands'
def test_dmdsec_from_csv_parsed_metadata_both(self):
"""It should create a dmdSec for DC and Other parsed metadata."""
data = collections.OrderedDict([
("dc.title", ["Yamani Weapons"]),
("dc.contributor", [u"雪 ユキ".encode('utf8')]),
("dcterms.isPartOf", ["AIC#42"]),
("Title", ["Yamani Weapons"]),
("Contributor", [u"雪 ユキ".encode('utf8')]),
("Long Description", ['This is about how glaives are used in the Yamani Islands'])
])
# Test
state = create_mets_v2.MetsState()
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(Job("stub", "stub", []), data, state)
# Verify
assert ret
assert len(ret) == 2
# Return can be DC or OTHER first, but in this case DC should be first
dc_dmdsec = ret[0]
assert dc_dmdsec.tag == '{http://www.loc.gov/METS/}dmdSec'
assert 'ID' in dc_dmdsec.attrib
mdwrap = dc_dmdsec[0]
assert mdwrap.tag == '{http://www.loc.gov/METS/}mdWrap'
assert 'MDTYPE' in mdwrap.attrib
assert mdwrap.attrib['MDTYPE'] == 'DC'
xmldata = mdwrap[0]
assert xmldata.tag == '{http://www.loc.gov/METS/}xmlData'
dc_elem = xmldata[0]
# Elements are children of dublincore tag
assert dc_elem.tag == '{http://purl.org/dc/terms/}dublincore'
assert len(dc_elem) == 3
assert dc_elem[0].tag == '{http://purl.org/dc/elements/1.1/}title'
assert dc_elem[0].text == 'Yamani Weapons'
assert dc_elem[1].tag == '{http://purl.org/dc/elements/1.1/}contributor'
assert dc_elem[1].text == u'雪 ユキ'
assert dc_elem[2].tag == '{http://purl.org/dc/terms/}isPartOf'
assert dc_elem[2].text == 'AIC#42'
other_dmdsec = ret[1]
assert other_dmdsec.tag == '{http://www.loc.gov/METS/}dmdSec'
assert 'ID' in other_dmdsec.attrib
mdwrap = other_dmdsec[0]
assert mdwrap.tag == '{http://www.loc.gov/METS/}mdWrap'
assert 'MDTYPE' in mdwrap.attrib
assert mdwrap.attrib['MDTYPE'] == 'OTHER'
assert 'OTHERMDTYPE' in mdwrap.attrib
assert mdwrap.attrib['OTHERMDTYPE'] == 'CUSTOM'
xmldata = mdwrap[0]
assert xmldata.tag == '{http://www.loc.gov/METS/}xmlData'
# Elements are direct children of xmlData
assert len(xmldata) == 3
assert xmldata[0].tag == 'title'
assert xmldata[0].text == 'Yamani Weapons'
assert xmldata[1].tag == 'contributor'
assert xmldata[1].text == u'雪 ユキ'
assert xmldata[2].tag == 'long_description'
assert xmldata[2].text == 'This is about how glaives are used in the Yamani Islands'
def test_dmdsec_from_csv_parsed_metadata_repeats(self):
"""It should create multiple elements for repeated input."""
data = collections.OrderedDict(
[
("dc.contributor", ["Yuki", "雪 ユキ".encode("utf8")]),
("Contributor", ["Yuki", "雪 ユキ".encode("utf8")]),
]
)
# Test
state = create_mets_v2.MetsState()
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(
Job("stub", "stub", []), data, state
)
# Verify
assert ret
assert len(ret) == 2
# Return can be DC or OTHER first, but in this case DC should be first
dc_dmdsec = ret[0]
assert dc_dmdsec.tag == "{http://www.loc.gov/METS/}dmdSec"
assert "ID" in dc_dmdsec.attrib
mdwrap = dc_dmdsec[0]
assert mdwrap.tag == "{http://www.loc.gov/METS/}mdWrap"
assert "MDTYPE" in mdwrap.attrib
assert mdwrap.attrib["MDTYPE"] == "DC"
xmldata = mdwrap[0]
assert xmldata.tag == "{http://www.loc.gov/METS/}xmlData"
dc_elem = xmldata[0]
# Elements are children of dublincore tag
assert dc_elem.tag == "{http://purl.org/dc/terms/}dublincore"
assert len(dc_elem) == 2
assert dc_elem[0].tag == "{http://purl.org/dc/elements/1.1/}contributor"
assert dc_elem[0].text == "Yuki"
assert dc_elem[1].tag == "{http://purl.org/dc/elements/1.1/}contributor"
assert dc_elem[1].text == "雪 ユキ"
other_dmdsec = ret[1]
assert other_dmdsec.tag == "{http://www.loc.gov/METS/}dmdSec"
assert "ID" in other_dmdsec.attrib
mdwrap = other_dmdsec[0]
assert mdwrap.tag == "{http://www.loc.gov/METS/}mdWrap"
assert "MDTYPE" in mdwrap.attrib
assert mdwrap.attrib["MDTYPE"] == "OTHER"
assert "OTHERMDTYPE" in mdwrap.attrib
assert mdwrap.attrib["OTHERMDTYPE"] == "CUSTOM"
xmldata = mdwrap[0]
assert xmldata.tag == "{http://www.loc.gov/METS/}xmlData"
# Elements are direct children of xmlData
assert len(xmldata) == 2
assert xmldata[0].tag == "contributor"
assert xmldata[0].text == "Yuki"
assert xmldata[1].tag == "contributor"
assert xmldata[1].text == "雪 ユキ"
def test_dmdsec_from_csv_parsed_metadata_repeats(self):
"""It should create multiple elements for repeated input."""
data = collections.OrderedDict([
("dc.contributor", ["Yuki", u"雪 ユキ".encode('utf8')]),
("Contributor", ["Yuki", u"雪 ユキ".encode('utf8')]),
])
# Test
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(Job("stub", "stub", []), data)
# Verify
assert ret
assert len(ret) == 2
# Return can be DC or OTHER first, but in this case DC should be first
dc_dmdsec = ret[0]
assert dc_dmdsec.tag == '{http://www.loc.gov/METS/}dmdSec'
assert 'ID' in dc_dmdsec.attrib
mdwrap = dc_dmdsec[0]
assert mdwrap.tag == '{http://www.loc.gov/METS/}mdWrap'
assert 'MDTYPE' in mdwrap.attrib
assert mdwrap.attrib['MDTYPE'] == 'DC'
xmldata = mdwrap[0]
assert xmldata.tag == '{http://www.loc.gov/METS/}xmlData'
dc_elem = xmldata[0]
# Elements are children of dublincore tag
assert dc_elem.tag == '{http://purl.org/dc/terms/}dublincore'
assert len(dc_elem) == 2
assert dc_elem[0].tag == '{http://purl.org/dc/elements/1.1/}contributor'
assert dc_elem[0].text == 'Yuki'
assert dc_elem[1].tag == '{http://purl.org/dc/elements/1.1/}contributor'
assert dc_elem[1].text == u'雪 ユキ'
other_dmdsec = ret[1]
assert other_dmdsec.tag == '{http://www.loc.gov/METS/}dmdSec'
assert 'ID' in other_dmdsec.attrib
mdwrap = other_dmdsec[0]
assert mdwrap.tag == '{http://www.loc.gov/METS/}mdWrap'
assert 'MDTYPE' in mdwrap.attrib
assert mdwrap.attrib['MDTYPE'] == 'OTHER'
assert 'OTHERMDTYPE' in mdwrap.attrib
assert mdwrap.attrib['OTHERMDTYPE'] == 'CUSTOM'
xmldata = mdwrap[0]
assert xmldata.tag == '{http://www.loc.gov/METS/}xmlData'
# Elements are direct children of xmlData
assert len(xmldata) == 2
assert xmldata[0].tag == 'contributor'
assert xmldata[0].text == 'Yuki'
assert xmldata[1].tag == 'contributor'
assert xmldata[1].text == u'雪 ユキ'
def update_metadata_csv(job, mets, metadata_csv, sip_uuid, sip_dir, state):
job.pyprint('Parse new metadata.csv')
full_path = metadata_csv.currentlocation.replace('%SIPDirectory%', sip_dir,
1)
csvmetadata = createmetscsv.parseMetadataCSV(job, full_path)
# FIXME This doesn't support having both DC and non-DC metadata in dmdSecs
# If createDmdSecsFromCSVParsedMetadata returns more than 1 dmdSec, behaviour is undefined
for f, md in csvmetadata.items():
# Verify file is in AIP
job.pyprint('Looking for', f, 'from metadata.csv in SIP')
# Find File with original or current locationg matching metadata.csv
# Prepend % to match the end of %SIPDirectory% or %transferDirectory%
try:
file_obj = models.File.objects.get(sip_id=sip_uuid,
originallocation__endswith='%' +
f)
except models.File.DoesNotExist:
try:
file_obj = models.File.objects.get(
sip_id=sip_uuid, currentlocation__endswith='%' + f)
except models.File.DoesNotExist:
job.pyprint(f, 'not found in database')
continue
job.pyprint(f, 'found in database')
fsentry = mets.get_file(file_uuid=file_obj.uuid)
job.pyprint(f, 'was associated with', fsentry.dmdids)
# Create dmdSec
new_dmdsecs = createmets2.createDmdSecsFromCSVParsedMetadata(
job, md, state)
# Add both
for new_dmdsec in new_dmdsecs:
# need to strip new_d to just the DC part
new_dc = new_dmdsec.find('.//dcterms:dublincore',
namespaces=ns.NSMAP)
new_metsrw_dmdsec = fsentry.add_dublin_core(new_dc)
if len(fsentry.dmdsecs) > 1:
fsentry.dmdsecs[-2].replace_with(new_metsrw_dmdsec)
job.pyprint(f, 'now associated with', fsentry.dmdids)
return mets
def test_dmdsec_from_csv_parsed_metadata_other_only(self):
"""It should only create an Other dmdSec from parsed metadata."""
data = collections.OrderedDict(
[
("Title", ["Yamani Weapons"]),
("Contributor", ["雪 ユキ".encode("utf8")]),
(
"Long Description",
["This is about how glaives are used in the Yamani Islands"],
),
]
)
# Test
state = create_mets_v2.MetsState()
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(
Job("stub", "stub", []), data, state
)
# Verify
assert ret
assert len(ret) == 1
dmdsec = ret[0]
assert dmdsec.tag == "{http://www.loc.gov/METS/}dmdSec"
assert "ID" in dmdsec.attrib
mdwrap = dmdsec[0]
assert mdwrap.tag == "{http://www.loc.gov/METS/}mdWrap"
assert "MDTYPE" in mdwrap.attrib
assert mdwrap.attrib["MDTYPE"] == "OTHER"
assert "OTHERMDTYPE" in mdwrap.attrib
assert mdwrap.attrib["OTHERMDTYPE"] == "CUSTOM"
xmldata = mdwrap[0]
assert xmldata.tag == "{http://www.loc.gov/METS/}xmlData"
# Elements are direct children of xmlData
assert len(xmldata) == 3
assert xmldata[0].tag == "title"
assert xmldata[0].text == "Yamani Weapons"
assert xmldata[1].tag == "contributor"
assert xmldata[1].text == "雪 ユキ"
assert xmldata[2].tag == "long_description"
assert (
xmldata[2].text
== "This is about how glaives are used in the Yamani Islands"
)
def test_dmdsec_from_csv_parsed_metadata_dc_only(self):
"""It should only create a DC dmdSec from parsed metadata."""
data = collections.OrderedDict([
("dc.title", ["Yamani Weapons"]),
("dc.creator", ["Keladry of Mindelan"]),
("dc.subject", ["Glaives"]),
("dc.description", ["Glaives are cool"]),
("dc.publisher", ["Tortall Press"]),
("dc.contributor", [u"雪 ユキ".encode('utf8')]),
("dc.date", ["2015"]),
("dc.type", ["Archival Information Package"]),
("dc.format", ["parchement"]),
("dc.identifier", ["42/1"]),
("dc.source", ["Numair's library"]),
("dc.relation", ["None"]),
("dc.language", ["en"]),
("dc.rights", ["Public Domain"]),
("dcterms.isPartOf", ["AIC#42"]),
])
# Test
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(Job("stub", "stub", []), data)
# Verify
assert ret
assert len(ret) == 1
dmdsec = ret[0]
assert dmdsec.tag == '{http://www.loc.gov/METS/}dmdSec'
assert 'ID' in dmdsec.attrib
mdwrap = dmdsec[0]
assert mdwrap.tag == '{http://www.loc.gov/METS/}mdWrap'
assert 'MDTYPE' in mdwrap.attrib
assert mdwrap.attrib['MDTYPE'] == 'DC'
xmldata = mdwrap[0]
assert xmldata.tag == '{http://www.loc.gov/METS/}xmlData'
# Elements are children of dublincore tag
dc_elem = xmldata[0]
assert dc_elem.tag == '{http://purl.org/dc/terms/}dublincore'
assert len(dc_elem) == 15
assert dc_elem[0].tag == '{http://purl.org/dc/elements/1.1/}title'
assert dc_elem[0].text == 'Yamani Weapons'
assert dc_elem[1].tag == '{http://purl.org/dc/elements/1.1/}creator'
assert dc_elem[1].text == 'Keladry of Mindelan'
assert dc_elem[2].tag == '{http://purl.org/dc/elements/1.1/}subject'
assert dc_elem[2].text == 'Glaives'
assert dc_elem[3].tag == '{http://purl.org/dc/elements/1.1/}description'
assert dc_elem[3].text == 'Glaives are cool'
assert dc_elem[4].tag == '{http://purl.org/dc/elements/1.1/}publisher'
assert dc_elem[4].text == 'Tortall Press'
assert dc_elem[5].tag == '{http://purl.org/dc/elements/1.1/}contributor'
assert dc_elem[5].text == u'雪 ユキ'
assert dc_elem[6].tag == '{http://purl.org/dc/elements/1.1/}date'
assert dc_elem[6].text == '2015'
assert dc_elem[7].tag == '{http://purl.org/dc/elements/1.1/}type'
assert dc_elem[7].text == 'Archival Information Package'
assert dc_elem[8].tag == '{http://purl.org/dc/elements/1.1/}format'
assert dc_elem[8].text == 'parchement'
assert dc_elem[9].tag == '{http://purl.org/dc/elements/1.1/}identifier'
assert dc_elem[9].text == '42/1'
assert dc_elem[10].tag == '{http://purl.org/dc/elements/1.1/}source'
assert dc_elem[10].text == "Numair's library"
assert dc_elem[11].tag == '{http://purl.org/dc/elements/1.1/}relation'
assert dc_elem[11].text == 'None'
assert dc_elem[12].tag == '{http://purl.org/dc/elements/1.1/}language'
assert dc_elem[12].text == 'en'
assert dc_elem[13].tag == '{http://purl.org/dc/elements/1.1/}rights'
assert dc_elem[13].text == 'Public Domain'
assert dc_elem[14].tag == '{http://purl.org/dc/terms/}isPartOf'
assert dc_elem[14].text == 'AIC#42'
def test_dmdsec_from_csv_parsed_metadata_both(self):
"""It should create a dmdSec for DC and Other parsed metadata."""
data = collections.OrderedDict(
[
("dc.title", ["Yamani Weapons"]),
("dc.contributor", ["雪 ユキ".encode("utf8")]),
("dcterms.isPartOf", ["AIC#42"]),
("Title", ["Yamani Weapons"]),
("Contributor", ["雪 ユキ".encode("utf8")]),
(
"Long Description",
["This is about how glaives are used in the Yamani Islands"],
),
]
)
# Test
state = create_mets_v2.MetsState()
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(
Job("stub", "stub", []), data, state
)
# Verify
assert ret
assert len(ret) == 2
# Return can be DC or OTHER first, but in this case DC should be first
dc_dmdsec = ret[0]
assert dc_dmdsec.tag == "{http://www.loc.gov/METS/}dmdSec"
assert "ID" in dc_dmdsec.attrib
mdwrap = dc_dmdsec[0]
assert mdwrap.tag == "{http://www.loc.gov/METS/}mdWrap"
assert "MDTYPE" in mdwrap.attrib
assert mdwrap.attrib["MDTYPE"] == "DC"
xmldata = mdwrap[0]
assert xmldata.tag == "{http://www.loc.gov/METS/}xmlData"
dc_elem = xmldata[0]
# Elements are children of dublincore tag
assert dc_elem.tag == "{http://purl.org/dc/terms/}dublincore"
assert len(dc_elem) == 3
assert dc_elem[0].tag == "{http://purl.org/dc/elements/1.1/}title"
assert dc_elem[0].text == "Yamani Weapons"
assert dc_elem[1].tag == "{http://purl.org/dc/elements/1.1/}contributor"
assert dc_elem[1].text == "雪 ユキ"
assert dc_elem[2].tag == "{http://purl.org/dc/terms/}isPartOf"
assert dc_elem[2].text == "AIC#42"
other_dmdsec = ret[1]
assert other_dmdsec.tag == "{http://www.loc.gov/METS/}dmdSec"
assert "ID" in other_dmdsec.attrib
mdwrap = other_dmdsec[0]
assert mdwrap.tag == "{http://www.loc.gov/METS/}mdWrap"
assert "MDTYPE" in mdwrap.attrib
assert mdwrap.attrib["MDTYPE"] == "OTHER"
assert "OTHERMDTYPE" in mdwrap.attrib
assert mdwrap.attrib["OTHERMDTYPE"] == "CUSTOM"
xmldata = mdwrap[0]
assert xmldata.tag == "{http://www.loc.gov/METS/}xmlData"
# Elements are direct children of xmlData
assert len(xmldata) == 3
assert xmldata[0].tag == "title"
assert xmldata[0].text == "Yamani Weapons"
assert xmldata[1].tag == "contributor"
assert xmldata[1].text == "雪 ユキ"
assert xmldata[2].tag == "long_description"
assert (
xmldata[2].text
== "This is about how glaives are used in the Yamani Islands"
)
def test_dmdsec_from_csv_parsed_metadata_dc_only(self):
"""It should only create a DC dmdSec from parsed metadata."""
data = collections.OrderedDict(
[
("dc.title", ["Yamani Weapons"]),
("dc.creator", ["Keladry of Mindelan"]),
("dc.subject", ["Glaives"]),
("dc.description", ["Glaives are cool"]),
("dc.publisher", ["Tortall Press"]),
("dc.contributor", ["雪 ユキ".encode("utf8")]),
("dc.date", ["2015"]),
("dc.type", ["Archival Information Package"]),
("dc.format", ["parchement"]),
("dc.identifier", ["42/1"]),
("dc.source", ["Numair's library"]),
("dc.relation", ["None"]),
("dc.language", ["en"]),
("dc.rights", ["Public Domain"]),
("dcterms.isPartOf", ["AIC#42"]),
]
)
# Test
state = create_mets_v2.MetsState()
ret = create_mets_v2.createDmdSecsFromCSVParsedMetadata(
Job("stub", "stub", []), data, state
)
# Verify
assert ret
assert len(ret) == 1
dmdsec = ret[0]
assert dmdsec.tag == "{http://www.loc.gov/METS/}dmdSec"
assert "ID" in dmdsec.attrib
mdwrap = dmdsec[0]
assert mdwrap.tag == "{http://www.loc.gov/METS/}mdWrap"
assert "MDTYPE" in mdwrap.attrib
assert mdwrap.attrib["MDTYPE"] == "DC"
xmldata = mdwrap[0]
assert xmldata.tag == "{http://www.loc.gov/METS/}xmlData"
# Elements are children of dublincore tag
dc_elem = xmldata[0]
assert dc_elem.tag == "{http://purl.org/dc/terms/}dublincore"
assert len(dc_elem) == 15
assert dc_elem[0].tag == "{http://purl.org/dc/elements/1.1/}title"
assert dc_elem[0].text == "Yamani Weapons"
assert dc_elem[1].tag == "{http://purl.org/dc/elements/1.1/}creator"
assert dc_elem[1].text == "Keladry of Mindelan"
assert dc_elem[2].tag == "{http://purl.org/dc/elements/1.1/}subject"
assert dc_elem[2].text == "Glaives"
assert dc_elem[3].tag == "{http://purl.org/dc/elements/1.1/}description"
assert dc_elem[3].text == "Glaives are cool"
assert dc_elem[4].tag == "{http://purl.org/dc/elements/1.1/}publisher"
assert dc_elem[4].text == "Tortall Press"
assert dc_elem[5].tag == "{http://purl.org/dc/elements/1.1/}contributor"
assert dc_elem[5].text == "雪 ユキ"
assert dc_elem[6].tag == "{http://purl.org/dc/elements/1.1/}date"
assert dc_elem[6].text == "2015"
assert dc_elem[7].tag == "{http://purl.org/dc/elements/1.1/}type"
assert dc_elem[7].text == "Archival Information Package"
assert dc_elem[8].tag == "{http://purl.org/dc/elements/1.1/}format"
assert dc_elem[8].text == "parchement"
assert dc_elem[9].tag == "{http://purl.org/dc/elements/1.1/}identifier"
assert dc_elem[9].text == "42/1"
assert dc_elem[10].tag == "{http://purl.org/dc/elements/1.1/}source"
assert dc_elem[10].text == "Numair's library"
assert dc_elem[11].tag == "{http://purl.org/dc/elements/1.1/}relation"
assert dc_elem[11].text == "None"
assert dc_elem[12].tag == "{http://purl.org/dc/elements/1.1/}language"
assert dc_elem[12].text == "en"
assert dc_elem[13].tag == "{http://purl.org/dc/elements/1.1/}rights"
assert dc_elem[13].text == "Public Domain"
assert dc_elem[14].tag == "{http://purl.org/dc/terms/}isPartOf"
assert dc_elem[14].text == "AIC#42"
def update_metadata_csv(job, mets, metadata_csv, sip_uuid, sip_dir, state):
job.pyprint("Parse new metadata.csv")
full_path = metadata_csv.currentlocation.replace("%SIPDirectory%", sip_dir, 1)
csvmetadata = createmetscsv.parseMetadataCSV(job, full_path)
# FIXME This doesn't support having both DC and non-DC metadata in dmdSecs
# If createDmdSecsFromCSVParsedMetadata returns more than 1 dmdSec, behaviour is undefined
for f, md in csvmetadata.items():
# Verify file is in AIP
job.pyprint("Looking for", f, "from metadata.csv in SIP")
# Find File with original or current locationg matching metadata.csv
# Prepend % to match the end of %SIPDirectory% or %transferDirectory%
file_obj = None
try:
file_obj = models.File.objects.get(
sip_id=sip_uuid, originallocation__endswith="%" + f
)
except models.File.DoesNotExist:
try:
file_obj = models.File.objects.get(
sip_id=sip_uuid, currentlocation__endswith="%" + f
)
except models.File.DoesNotExist:
pass
if file_obj is not None:
fsentry = mets.get_file(file_uuid=file_obj.uuid)
else:
fsentry = _get_directory_fsentry(mets, f)
if fsentry is None:
job.pyprint(f, "not found in database or METS file")
continue
job.pyprint(f, "found in database or METS file")
job.pyprint(f, "was associated with", fsentry.dmdids)
# Save existing dmdSecs
dc_dmdsecs = []
non_dc_dmdsecs = []
for dmdsec in fsentry.dmdsecs:
mdwrap = dmdsec.contents
if mdwrap.mdtype == "DC":
dc_dmdsecs.append(dmdsec)
elif (
mdwrap.mdtype == "OTHER"
and getattr(mdwrap, "othermdtype", None) == "CUSTOM"
):
non_dc_dmdsecs.append(dmdsec)
# Create dmdSec
new_dmdsecs = createmets2.createDmdSecsFromCSVParsedMetadata(job, md, state)
# Add both
for new_dmdsec in new_dmdsecs:
# need to strip new_d to just the DC part
new_dc = new_dmdsec.find(".//dcterms:dublincore", namespaces=ns.NSMAP)
if new_dc is not None:
new_metsrw_dmdsec = fsentry.add_dublin_core(new_dc)
_replace_original_dmdsec(dc_dmdsecs, new_metsrw_dmdsec)
else:
new_non_dc = new_dmdsec.find(
'.//mets:mdWrap[@MDTYPE="OTHER"][@OTHERMDTYPE="CUSTOM"]/mets:xmlData',
namespaces=ns.NSMAP,
)
if new_non_dc is not None:
new_metsrw_dmdsec = fsentry.add_dmdsec(
new_non_dc, "OTHER", othermdtype="CUSTOM"
)
_replace_original_dmdsec(non_dc_dmdsecs, new_metsrw_dmdsec)
job.pyprint(f, "now associated with", fsentry.dmdids)
return mets