def _getDescriptiveMetadata(self, lxmlNode):
## This always normalizes to rdf namespace, without warning/message
descriptiveMetadataItem = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@rdf:resource="info:eu-repo/semantics/descriptiveMetadata"]',
namespaces=self._nsMap)
if len(descriptiveMetadataItem
) == 0: #Fallback to @resource (no rdf nmsp), if available...
descriptiveMetadataItem = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@resource="info:eu-repo/semantics/descriptiveMetadata"]',
namespaces=self._nsMap)
if len(descriptiveMetadataItem) > 0:
self.do.logMsg(self._identifier, LOGGER3, prefix=STR_DIDL)
if len(descriptiveMetadataItem
) == 0: #Fallback to dip namespace, if available...
descriptiveMetadataItem = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Item[didl:Descriptor/didl:Statement/dip:ObjectType/text()="info:eu-repo/semantics/descriptiveMetadata"]',
namespaces=self._nsMap)
if len(descriptiveMetadataItem) > 0:
self.do.logMsg(self._identifier, LOGGER4, prefix=STR_DIDL)
if len(descriptiveMetadataItem) > 0:
#look for first DMI containing MODS:
dmi_mods = None
dmItem = None
for dmi in descriptiveMetadataItem:
node = dmi.xpath('self::didl:Item//mods:mods',
namespaces=self._nsMap)
if len(node) > 0: #Found MODS:
dmi_mods = node[0]
dmItem = dmi
break
else:
raise ValidateException(
formatExceptionLine(EXCEPTION6, prefix=STR_DIDL))
item_template = """<didl:Item>
<didl:Descriptor>
<didl:Statement mimeType="application/xml">
<rdf:type rdf:resource="info:eu-repo/semantics/descriptiveMetadata"/>
</didl:Statement>
</didl:Descriptor>
%s%s<didl:Component>
<didl:Resource mimeType="application/xml">
%s
</didl:Resource>
</didl:Component>
</didl:Item>""" % (
self._getIdentifierDescriptor(dmItem),
self._getDateModifiedDescriptor(dmItem), tostring(dmi_mods))
else:
raise ValidateException(
formatExceptionLine(EXCEPTION7, prefix=STR_DIDL))
return item_template
def _normalizeRecord(self, lxmlNode):
# MODS normalisation in 4 steps:
# 1. Get Mods from the lxmlNode.
# 2. Normalize it
# 3. Put it back in place.
# 4. return the lxmlNode containing the normalized MODS.
#1: Get Mods from the lxmlNode:
lxmlMODS = lxmlNode.xpath('(//mods:mods)[1]', namespaces=self._nsMap)
## Our normalisation functions to call:
modsFunctions = [self._convertFullMods2GHMods]
if len(lxmlMODS) > 0:
#2: Normalize it
str_norm_mods = ''
for function in modsFunctions:
str_norm_mods += function(lxmlMODS[0])
#3: Put it back in DIDL/place:
lxmlMODS[0].getparent().replace(lxmlMODS[0],
etree.fromstring(str_norm_mods))
else: #This should never happen @runtime: record should have been validated up front...
raise ValidateException(
formatExceptionLine(EXCEPTION1, prefix=STR_MODS))
#4: Return the lxmlNode containing the normalized MODS:
#print(etree.tostring(lxmlNode, pretty_print=True))
return lxmlNode
def _tlOrigininfo(self, childNode):
hasDateIssued = False
## Select all children from originInfo having 'encoding' attribute:
children = childNode.xpath(
"self::mods:originInfo/child::*[@encoding='w3cdtf' or @encoding='iso8601']",
namespaces=self._nsMap)
if len(children) > 0:
for child in children:
if self._validateISO8601(child.text):
child.text = self._granulateDate(child.text)
child.set('encoding', 'w3cdtf')
if child.tag == ('{%s}dateIssued') % self._nsMap['mods']:
hasDateIssued = True
else:
child.getparent().remove(child)
if not hasDateIssued:
raise ValidateException(
formatExceptionLine(EXCEPTION7, prefix=STR_MODS))
for child in childNode.xpath("self::mods:originInfo/mods:publisher",
namespaces=self._nsMap):
if not child.text:
child.getparent().remove(child)
return childNode if len(childNode) > 0 else None
def _isValidTitleInfoTag(self, lxmlNode):
for title in lxmlNode.iterfind(('{%s}title') % self._nsMap['mods']):
if not title.text:
raise ValidateException(
formatExceptionLine(EXCEPTION3, prefix=STR_MODS))
for subtitle in lxmlNode.iterfind(
('{%s}subTitle') % self._nsMap['mods']):
if not subtitle.text:
subtitle.getparent().remove(subtitle)
return True
def _normalizeTitleinfo(self, modsNode):
## Select all titleInfo's
hasTitleInfo = False
for child in modsNode.iterfind(
('{%s}titleInfo') % self._nsMap['mods']):
hasTitleInfo = True
if not self._isValidTitleInfoTag(child):
modsNode.remove(child)
if not hasTitleInfo:
raise ValidateException(
formatExceptionLine(EXCEPTION2, prefix=STR_MODS))
def _getHumanStartPage(self, lxmlNode):
didl_hsp_item = lxmlNode.xpath(
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@rdf:resource="info:eu-repo/semantics/humanStartPage"]',
namespaces=self._nsMap)
if len(didl_hsp_item) == 0:
didl_hsp_item = lxmlNode.xpath(
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@resource="info:eu-repo/semantics/humanStartPage"]',
namespaces=self._nsMap)
if len(didl_hsp_item) > 0:
self.do.logMsg(self._identifier, LOGGER9, prefix=STR_DIDL)
if len(didl_hsp_item) == 0:
didl_hsp_item = lxmlNode.xpath(
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/dip:ObjectType/text()="info:eu-repo/semantics/humanStartPage"]',
namespaces=self._nsMap)
if len(didl_hsp_item) > 0:
self.do.logMsg(self._identifier, LOGGER10, prefix=STR_DIDL)
if len(didl_hsp_item) == 0:
self.do.logMsg(self._identifier, LOGGER11, prefix=STR_DIDL)
return ""
uriref = didl_hsp_item[0].xpath(
'self::didl:Item/didl:Component/didl:Resource/@ref',
namespaces=self._nsMap)
mimetype = didl_hsp_item[0].xpath(
'self::didl:Item/didl:Component/didl:Resource/@mimeType',
namespaces=self._nsMap)
if len(mimetype) == 0:
self.do.logMsg(self._identifier, LOGGER13, prefix=STR_DIDL)
if len(mimetype) > 0 and not comm.isMimeType(mimetype[0]):
self.do.logMsg(self._identifier,
LOGGER12 + mimetype[0],
prefix=STR_DIDL)
if len(uriref) == 0 or not comm.isURL(uriref[0]):
raise ValidateException(
formatExceptionLine(EXCEPTION11, prefix=STR_DIDL))
return """<didl:Item>
<didl:Descriptor>
<didl:Statement mimeType="application/xml">
<rdf:type rdf:resource="info:eu-repo/semantics/humanStartPage"/>
</didl:Statement>
</didl:Descriptor>
<didl:Component>
<didl:Resource ref="%s" mimeType="%s"/>
</didl:Component>
</didl:Item>""" % (escapeXml(comm.urlQuote(
uriref[0].strip())), escapeXml(mimetype[0]))
def _validateNames(self, modsNode):
for name in modsNode.iterfind(('{%s}name') % self._nsMap['mods']):
for roleterm in name.iterfind(
('.//{%s}roleTerm') % self._nsMap['mods']):
if roleterm.text: roleterm.text = roleterm.text.strip()
role = name.xpath(
"self::mods:name/mods:role/mods:roleTerm[@type='code' and @authority='marcrelator']/text()",
namespaces=self._nsMap)
for namepart in name.iterfind(
('{%s}namePart') % self._nsMap['mods']):
if not namepart.text: # Remove empty nameparts
name.remove(namepart)
if not role or len(role) < 1 or name.find(
('{%s}namePart') % self._nsMap['mods']
) is None: ## Geen roleterm gevonden, of lege string voor type code en authority marcrelator, of geen nameParts: Verwijder dit name element:
modsNode.remove(name)
elif len(role) > 0 and not self.__isValidRoleTerm(role[0]):
raise ValidateException(
formatExceptionLine(EXCEPTION4 + role[0], prefix=STR_MODS))
if len(modsNode.xpath("//mods:mods/mods:name",
namespaces=self._nsMap)) <= 0:
raise ValidateException(
formatExceptionLine(EXCEPTION5, prefix=STR_MODS))
def _validateGenre(self, modsNode):
fqGenre = None
bln_hasValid = False
## Loop all 'genre' elements as separate nodes:
for genre in modsNode.iterfind('{' + self._nsMap.get('mods') +
'}genre'):
for key, value in GENRES_SEMANTIEK.iteritems():
if genre.text.strip().lower().find(
key) >= 0: #found a (lowercased) genre
fqGenre = value
break
if fqGenre is not None and not bln_hasValid:
bln_hasValid = True
genre.text = fqGenre
else:
modsNode.remove(genre)
if not bln_hasValid:
raise ValidateException(
formatExceptionLine(EXCEPTION6, prefix=STR_MODS))
def _getObjectfiles(self, lxmlNode):
of_container = ''
objectfiles = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@rdf:resource="info:eu-repo/semantics/objectFile"]',
namespaces=self._nsMap)
if len(objectfiles) == 0:
objectfiles = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@resource="info:eu-repo/semantics/objectFile"]',
namespaces=self._nsMap)
if len(objectfiles) > 0:
self.do.logMsg(self._identifier, LOGGER6, prefix=STR_DIDL)
if len(objectfiles) == 0:
objectfiles = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Item[didl:Descriptor/didl:Statement/dip:ObjectType/text()="info:eu-repo/semantics/objectFile"]',
namespaces=self._nsMap)
if len(objectfiles) > 0:
self.do.logMsg(self._identifier, LOGGER7, prefix=STR_DIDL)
for objectfile in objectfiles:
#1:Define correct ObjectFile descriptor:
of_container += '<didl:Item><didl:Descriptor><didl:Statement mimeType="application/xml"><rdf:type rdf:resource="info:eu-repo/semantics/objectFile"/></didl:Statement></didl:Descriptor>'
#2: Check geldige Identifier (feitelijk verplicht, hoewel vaak niet geimplemeteerd...)
pi = objectfile.xpath(
'self::didl:Item/didl:Descriptor/didl:Statement/dii:Identifier/text()',
namespaces=self._nsMap)
if len(pi) > 0:
of_container += descr_templ % ('<dii:Identifier>' + escapeXml(
pi[0].strip()) + '</dii:Identifier>')
#3: Check op geldige AccessRights:
arights = objectfile.xpath(
'self::didl:Item/didl:Descriptor/didl:Statement/dcterms:accessRights/text()',
namespaces=self._nsMap)
if len(arights) > 0:
for key, value in accessRights.iteritems():
if arights[0].strip().lower().find(key) >= 0:
of_container += descr_templ % (
'<dcterms:accessRights>' + value +
'</dcterms:accessRights>')
break
else:
raise ValidateException(
formatExceptionLine(arights[0] + EXCEPTION12,
prefix=STR_DIDL))
else:
raise ValidateException(
formatExceptionLine(EXCEPTION8, prefix=STR_DIDL))
#4: Check geldige datemodified (feitelijk verplicht, hoewel vaak niet geimplemeteerd...)
modified = objectfile.xpath(
'self::didl:Item/didl:Descriptor/didl:Statement/dcterms:modified/text()',
namespaces=self._nsMap)
if len(modified) > 0 and comm.isISO8601(modified[0]):
of_container += descr_templ % ('<dcterms:modified>' +
modified[0].strip() +
'</dcterms:modified>')
#5: Check for 'file' description:
descr = objectfile.xpath(
'self::didl:Item/didl:Descriptor/didl:Statement/dc:description/text()',
namespaces=self._nsMap)
if len(descr) > 0:
of_container += descr_templ % ('<dc:description>' + escapeXml(
descr[0].strip()) + '</dc:description>')
## SKIPPING: Not in EduStandaard.
#6.0: Check for embargo:
# embargo = objectfile.xpath('self::didl:Item/didl:Descriptor/didl:Statement/dcterms:available/text()', namespaces=self._nsMap)
# if len(embargo) > 0 and comm.isISO8601(embargo[0]):
# of_container += descr_templ % ('<dcterms:available>'+embargo[0].strip()+'</dcterms:available>')
## SKIPPING: Not in EduStandaard.
#6.1: Check for dateSubmitted:
# dembargo = objectfile.xpath('self::didl:Item/didl:Descriptor/didl:Statement/dcterms:dateSubmitted/text()', namespaces=self._nsMap)
# if len(dembargo) > 0 and comm.isISO8601(dembargo[0]):
# of_container += descr_templ % ('<dcterms:dateSubmitted>'+dembargo[0].strip()+'</dcterms:dateSubmitted>')
# else:
# #6.2: Check for issued (depricated, normalize to dateSubmitted):
# issued = objectfile.xpath('self::didl:Item/didl:Descriptor/didl:Statement/dcterms:issued/text()', namespaces=self._nsMap)
# if len(issued) > 0 and comm.isISO8601(issued[0]):
# of_container += descr_templ % ('<dcterms:dateSubmitted>'+issued[0].strip()+'</dcterms:dateSubmitted>')
#7: Check for published version(author/publisher):
pubVersion = objectfile.xpath(
'self::didl:Item/didl:Descriptor/didl:Statement/rdf:type/@rdf:resource',
namespaces=self._nsMap)
if len(
pubVersion
) > 0: ## Both (author/publisher) may be available: we'll take the first one...
for key, value in pubVersions.iteritems():
if pubVersion[0].strip().lower().find(key) >= 0:
of_container += descr_templ % (
'<rdf:type rdf:resource="' + value + '"/>')
break
#8:Check for MANDATORY resources and mimetypes:
didl_resources = objectfile.xpath(
'self::didl:Item/didl:Component/didl:Resource[@mimeType and @ref]',
namespaces=self._nsMap)
resources = ''
_url_list = []
for resource in didl_resources:
mimeType = resource.xpath('self::didl:Resource/@mimeType',
namespaces=self._nsMap)
uri = resource.xpath('self::didl:Resource/@ref',
namespaces=self._nsMap)
## We need both mimeType and URI: (MIMETYPE is required by DIDL schema, @ref not).
if len(mimeType) > 0 and len(uri) > 0:
if not comm.isMimeType(mimeType[0]):
self.do.logMsg(self._identifier,
LOGGER8 + mimeType[0],
prefix=STR_DIDL)
if comm.isURL(uri[0].strip()):
resources += """<didl:Resource mimeType="%s" ref="%s"/>""" % (
escapeXml(mimeType[0].strip()),
escapeXml(comm.urlQuote(uri[0].strip())))
_url_list.append(
"""<didl:Resource mimeType="%s" ref="%s"/>""" %
(escapeXml(mimeType[0].strip()),
escapeXml(comm.urlQuote(uri[0].strip()))))
else:
raise ValidateException(
formatExceptionLine(EXCEPTION9 + uri[0],
prefix=STR_DIDL))
if resources != '':
of_container += """<didl:Component>
%s
</didl:Component>""" % (resources)
else:
raise ValidateException(
formatExceptionLine(EXCEPTION10, prefix=STR_DIDL))
of_container += '</didl:Item>'
return of_container
def _getTopItem(self, lxmlNode):
## Wrappers:
pid, modified, mimetype, pidlocation = '', '', "application/xml", ''
#1: Get persistentIdentifier:
pidlist = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Descriptor/didl:Statement/dii:Identifier/text()',
namespaces=self._nsMap)
if len(pidlist) > 0:
pid = pidlist[0].strip()
if not comm.isURNNBN(pid):
raise ValidateException(
formatExceptionLine(EXCEPTION0 + pid, prefix=STR_DIDL))
else:
raise ValidateException(
formatExceptionLine(EXCEPTION1, prefix=STR_DIDL))
#2: Get toplevel modificationDate: comm.isISO8601()
tl_modified = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Descriptor/didl:Statement/dcterms:modified/text()',
namespaces=self._nsMap)
## Check op geldig/aanwezigheid tlModified, anders exception:
if len(tl_modified) > 0 and not comm.isISO8601(tl_modified[0]):
raise ValidateException(
formatExceptionLine(EXCEPTION2 + tl_modified[0],
prefix=STR_DIDL))
elif len(tl_modified) == 0:
raise ValidateException(
formatExceptionLine(EXCEPTION3, prefix=STR_DIDL))
## Get all modified dates:
all_modified = lxmlNode.xpath(
'//didl:Item/didl:Descriptor/didl:Statement/dcterms:modified/text()',
namespaces=self._nsMap)
## Get most recent date from all items, to add to toplevelItem:
if len(all_modified) > 0:
datedict = {}
for date in all_modified:
if comm.isISO8601(date.strip()):
#datedict[parseDate(date.strip())] = date.strip()
pd = parseDate(date.strip())
datedict["%s %s" %
(str(pd.date()), str(pd.time()))] = date.strip()
## Get first sorted key:
for key in reversed(sorted(datedict.iterkeys())):
modified = datedict[key]
break
if not tl_modified[0].strip() == modified:
self.do.logMsg(self._identifier, LOGGER1, prefix=STR_DIDL)
#3: Get PidResourceMimetype
mimetypelist = lxmlNode.xpath(
'//didl:DIDL/didl:Item/didl:Component/didl:Resource/@mimeType',
namespaces=self._nsMap)
if len(mimetypelist) > 0:
mimetype = mimetypelist[0].strip()
if not comm.isMimeType(mimetype):
self.do.logMsg(self._identifier,
LOGGER2 + mimetype,
prefix=STR_DIDL)
#4: Get PidResourceLocation:
pidlocation = self._findAndBindFirst(
lxmlNode,
'%s',
'//didl:DIDL/didl:Item/didl:Component/didl:Resource/@ref',
'//didl:DIDL/didl:Item/didl:Component/didl:Resource/text()'
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@rdf:resource="info:eu-repo/semantics/humanStartPage"]/didl:Component/didl:Resource/@ref', #DIDL 3.0
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@resource="info:eu-repo/semantics/humanStartPage"]/didl:Component/didl:Resource/@ref', #DIDL 3.0, without @rdf:resource
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/dip:ObjectType/text()="info:eu-repo/semantics/humanStartPage"]/didl:Component/didl:Resource/@ref', #fallback DIDL 2.3.1
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@rdf:resource="info:eu-repo/semantics/objectFile"]/didl:Component/didl:Resource/@ref', #fallback DIDL 3.0
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/rdf:type/@resource="info:eu-repo/semantics/objectFile"]/didl:Component/didl:Resource/@ref', #fallback DIDL 3.0, without @rdf:resource
'//didl:Item/didl:Item[didl:Descriptor/didl:Statement/dip:ObjectType/text()="info:eu-repo/semantics/objectFile"]/didl:Component/didl:Resource/@ref' #fallback DIDL 2.3.1
).strip()
if pidlocation == '':
raise ValidateException(
formatExceptionLine(EXCEPTION4, prefix=STR_DIDL))
if not comm.isURL(pidlocation):
raise ValidateException(
formatExceptionLine(EXCEPTION5 + pidlocation, prefix=STR_DIDL))
return """<didl:Item>
<didl:Descriptor><didl:Statement mimeType="application/xml"><dii:Identifier>%s</dii:Identifier></didl:Statement></didl:Descriptor>
<didl:Descriptor><didl:Statement mimeType="application/xml"><dcterms:modified>%s</dcterms:modified></didl:Statement></didl:Descriptor>
<didl:Component><didl:Resource mimeType="%s" ref="%s"/></didl:Component>""" % (
escapeXml(pid), modified, escapeXml(mimetype),
comm.urlQuote(pidlocation))
def _checkOriginInfoDateIssued(self, modsNode):
if len(
modsNode.xpath("//mods:mods/mods:originInfo/mods:dateIssued",
namespaces=self._nsMap)) <= 0:
raise ValidateException(
formatExceptionLine(EXCEPTION7, prefix=STR_MODS))
