def update_dublincore(job, mets, sip_uuid):
"""
Add new dmdSec for updated Dublin Core info relating to entire SIP.
Case: No DC in DB, no DC in METS: Do nothing
Case: No DC in DB, DC in METS: Mark as deleted.
Case: DC in DB is untouched (METADATA_STATUS_REINGEST): Do nothing
Case: New DC in DB with METADATA_STATUS_ORIGINAL: Add new DC
Case: DC in DB with METADATA_STATUS_UPDATED: mark old, create updated
"""
# Check for DC in DB with METADATA_STATUS_UPDATED or METADATA_STATUS_ORIGINAL
untouched = models.DublinCore.objects.filter(
metadataappliestoidentifier=sip_uuid,
metadataappliestotype_id=createmets2.SIPMetadataAppliesToType,
status=models.METADATA_STATUS_REINGEST,
).exists()
if untouched:
# No new or updated DC found - return early
job.pyprint("No updated or new DC metadata found")
return mets
# Get structMap element related to SIP DC info
objects_div = mets.get_file(label="objects", type="Directory")
job.pyprint("Existing dmdIds for DC metadata:", objects_div.dmdids)
# Create element
dc_elem = createmets2.getDublinCore(createmets2.SIPMetadataAppliesToType, sip_uuid)
if dc_elem is None:
if objects_div.dmdsecs:
job.pyprint("DC metadata was deleted")
# Create 'deleted' DC element
dc_elem = etree.Element(
ns.dctermsBNS + "dublincore",
nsmap={"dcterms": ns.dctermsNS, "dc": ns.dcNS},
)
dc_elem.set(
ns.xsiBNS + "schemaLocation",
ns.dctermsNS
+ " http://dublincore.org/schemas/xmls/qdc/2008/02/11/dcterms.xsd",
)
else:
# No new or updated DC found - return early
job.pyprint("No updated or new DC metadata found")
return mets
dmdsec = objects_div.add_dublin_core(dc_elem)
job.pyprint("Adding new DC in dmdSec with ID", dmdsec.id_string)
if len(objects_div.dmdsecs) > 1:
objects_div.dmdsecs[-2].replace_with(dmdsec)
return mets
def test_get_dublincore(self):
"""It should create a Dublin Core element from the database info."""
# Generate DC element from DB
dc_elem = create_mets_v2.getDublinCore(self.siptypeuuid, self.sipuuid)
# Verify created correctly
assert dc_elem is not None
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 == 'Yuki'
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_get_dublincore(self):
"""It should create a Dublin Core element from the database info."""
# Generate DC element from DB
dc_elem = create_mets_v2.getDublinCore(self.siptypeuuid, self.sipuuid)
# Verify created correctly
assert dc_elem is not None
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 == "Yuki"
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_get_dublincore_none_found(self):
"""It should not create a Dublin Core element if no info found."""
sipuuid = 'dnednedn-5bd2-4249-84a1-2f00f725b981'
dc_elem = create_mets_v2.getDublinCore(self.siptypeuuid, sipuuid)
assert dc_elem is None
def create_mets_file(aic, aips, job):
""" Create AIC METS file with AIP information. """
# Prepare constants
nsmap = {"mets": ns.metsNS, "xlink": ns.xlinkNS, "xsi": ns.xsiNS}
now = timezone.now().strftime("%Y-%m-%dT%H:%M:%S")
# Set up structure
E = ElementMaker(namespace=ns.metsNS, nsmap=nsmap)
mets = E.mets(
E.metsHdr(CREATEDATE=now),
E.dmdSec(E.mdWrap(E.xmlData(), MDTYPE="DC"),
ID="dmdSec_1"), # mdWrap # dmdSec
E.fileSec(E.fileGrp()),
E.structMap(
E.div(TYPE="Archival Information Collection", DMDID="dmdSec_1"),
TYPE="logical", # structMap
),
)
mets.attrib["{{{ns}}}schemaLocation".format(
ns=nsmap["xsi"]
)] = "http://www.loc.gov/METS/ http://www.loc.gov/standards/mets/version1121/mets.xsd"
# Add Dublin Core info
xml_data = mets.find("mets:dmdSec/mets:mdWrap/mets:xmlData",
namespaces=ns.NSMAP)
dublincore = create_mets_v2.getDublinCore(
create_mets_v2.SIPMetadataAppliesToType, aic["uuid"])
# Add <extent> with number of AIPs
extent = etree.SubElement(dublincore, ns.dctermsBNS + "extent")
extent.text = "{} AIPs".format(len(aips))
xml_data.append(dublincore)
# Add elements for each AIP
file_grp = mets.find("mets:fileSec/mets:fileGrp", namespaces=ns.NSMAP)
struct_div = mets.find("mets:structMap/mets:div", namespaces=ns.NSMAP)
for aip in aips:
file_id = "{name}-{uuid}".format(name=aip["name"], uuid=aip["uuid"])
etree.SubElement(file_grp, ns.metsBNS + "file", ID=file_id)
label = aip["label"] or aip["name"]
div = etree.SubElement(struct_div, ns.metsBNS + "div", LABEL=label)
etree.SubElement(div, ns.metsBNS + "fptr", FILEID=file_id)
job.pyprint(etree.tostring(mets, pretty_print=True))
# Write out the file
file_uuid = str(uuid.uuid4())
basename = os.path.join("metadata", "METS.{}.xml".format(file_uuid))
filename = os.path.join(aic["dir"], basename)
with open(filename, "w") as f:
f.write(
etree.tostring(mets,
pretty_print=True,
xml_declaration=True,
encoding="utf-8"))
fileOperations.addFileToSIP(
filePathRelativeToSIP="%SIPDirectory%" + basename,
fileUUID=file_uuid,
sipUUID=aic["uuid"],
taskUUID=str(uuid.uuid4()), # Unsure what should go here
date=now,
sourceType="aip creation",
use="metadata",
)
# To make this work with the createMETS2 (for SIPs)
databaseFunctions.insertIntoDerivations(file_uuid, file_uuid)
# Insert the count of AIPs in the AIC into UnitVariables, so it can be
# indexed later
UnitVariable.objects.create(
unittype="SIP",
unituuid=aic["uuid"],
variable="AIPsinAIC",
variablevalue=str(len(aips)),
)
