def setUp(self):
SeecrTestCase.setUp(self)
self.observer1 = CallTrace('observer1',
emptyGeneratorMethods=['message'],
returnValues={
'function': 41,
'gen': (i for i in [41]),
'noop': None
})
self.observer2 = object()
self.observer3 = CallTrace('observer3',
emptyGeneratorMethods=['message'],
returnValues={
'function': 42,
'gen': (i for i in [42]),
'noop': None
})
self.dna = be(
(Observable(), (FilterMessages(
disallowed=['message', 'function', 'gen', 'noop']),
(self.observer1, )),
(FilterMessages(allowed=['message', 'function', 'gen', 'noop']),
(self.observer2, )),
(FilterMessages(allowed=['message', 'function', 'gen', 'noop']),
(self.observer3, ))))
def uploadHelix(lucene, termNumerator, storageComponent, drilldownFields,
fieldRegistry):
indexHelix = (Fields2LuceneDoc('record', fieldRegistry=fieldRegistry),
(termNumerator, ), (lucene, ))
return \
(SruRecordUpdate(),
(TransactionScope('record'),
(Venturi(should=[{'partname': 'record', 'xpath': '.'}], namespaces={'doc': 'http://meresco.org/namespace/example'}),
(FilterMessages(allowed=['delete']),
(lucene,),
(storageComponent,)
),
(FilterMessages(allowed=['add']),
(Xml2Fields(),
(RenameField(lambda name: name.split('.', 1)[-1]),
(FilterField(lambda name: 'fieldHier' not in name),
indexHelix,
),
(FilterField(lambda name: name == 'intfield1'),
(RenameField(lambda name: SORTED_PREFIX + name),
indexHelix,
)
),
(FilterField(lambda name: name in ['field2', 'field3']),
(RenameField(lambda name: UNTOKENIZED_PREFIX + name),
indexHelix,
)
),
)
),
(FieldHier(),
indexHelix,
)
),
(XmlPrintLxml(fromKwarg='lxmlNode', toKwarg='data'),
(storageComponent,)
)
)
)
)
def testWriteLogMustNotFail(self):
logwriter = CallTrace('logwriter',
emptyGeneratorMethods=['someMessage'])
logwriter.exceptions['writeLog'] = ValueError
top = be((Observable(), (LogCollector('default'), (logwriter, ),
(FilterMessages(allowed=['someMessage']),
(LogKeyValue(dict(name='A')), )))))
with stderr_replaced() as err:
try:
consume(top.all.someMessage())
except ValueError:
self.fail("Should not raise an error; Only print it")
self.assertTrue('ValueError' in err.getvalue(), err.getvalue())
def testScope(self):
logwriter = CallTrace('logwriter',
emptyGeneratorMethods=['someMessage'])
top = be(
(Observable(), (LogCollector('default'), (logwriter, ),
(FilterMessages(allowed=['someMessage']), (
LogKeyValue(dict(name='A')),
(LogKeyValue(dict(name='B')), ),
(
LogCollectorScope('scope_one'),
(LogKeyValue(dict(name='C')), ),
),
(LogKeyValue(dict(name='D')), ),
(
LogCollectorScope('scope_two'),
(
LogKeyValue(dict(name='E')),
(
LogCollectorScope('scope_two_one'),
(LogKeyValue(dict(name='F')), ),
),
),
(LogKeyValue(dict(name='G')), ),
),
)))))
consume(top.all.someMessage())
self.assertEqual(['someMessage', 'writeLog'],
logwriter.calledMethodNames())
self.assertEqual(
{
'name': ['A', 'B', 'D'],
'scope_one': {
'name': ['C'],
},
'scope_two': {
'name': ['E', 'G'],
'scope_two_one': {
'name': ['F']
}
}
}, logwriter.calledMethods[-1].kwargs['collectedLog'])
def createDownloadHelix(reactor, periodicDownload, oaiDownload,
dbStorageComponent):
return \
(periodicDownload, # Scheduled connection to a remote (response / request)...
(XmlParseLxml(fromKwarg="data", toKwarg="lxmlNode", parseOptions=dict(huge_tree=True, remove_blank_text=True)), # Convert from plain text to lxml-object.
(oaiDownload, # Implementation/Protocol of a PeriodicDownload...
(UpdateAdapterFromOaiDownloadProcessor(), # Maakt van een SRU update/delete bericht (lxmlNode) een relevante message: 'delete' of 'add' message.
# (FilterMessages(['delete']), # Filtert delete messages
# # (LogComponent("Delete msg:"),),
# # Write a 'deleted' part to the storage, that holds the (Record)uploadId.
# # (WriteTombstone(),
# # (storageComponent,),
# # )
# ),
(FilterMessages(allowed=['add']),
# (LogComponent("AddToNBNRES"),),
(Resolver(ro=False, nsMap=NAMESPACEMAP),
(dbStorageComponent,),
),
# (XmlXPath(['//document:document/document:part[@name="normdoc"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
# (XmlParseLxml(fromKwarg='data', toKwarg='lxmlNode'),
# (LogComponent("NORMDOC"),), #TODO: get urn:nbn and location from document.
# # (RewritePartname(NL_DIDL_NORMALISED_PREFIX), # Hernoemt partname van 'record' naar "metadata".
# # (XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
# # (storageComponent,) # Schrijft oai:metadata (=origineel) naar storage.
# # )
# # )
# )
# )
)
)
)
)
)
def createDownloadHelix(reactor, periodicDownload, oaiDownload,
storageComponent, oaiJazz):
return \
(periodicDownload, # Scheduled connection to a remote (response / request)...
(XmlParseLxml(fromKwarg="data", toKwarg="lxmlNode", parseOptions=dict(huge_tree=True, remove_blank_text=True)), # Convert from plain text to lxml-object.
(oaiDownload, # Implementation/Protocol of a PeriodicDownload...
(UpdateAdapterFromOaiDownloadProcessor(), # Maakt van een SRU update/delete bericht (lxmlNode) een relevante message: 'delete' of 'add' message.
(FilterMessages(['delete']), # Filtert delete messages
# (LogComponent("Delete Update"),),
(storageComponent,), # Delete from storage
(oaiJazz,), # Delete from OAI-pmh repo
# Write a 'deleted' part to the storage, that holds the (Record)uploadId.
(WriteTombstone(),
(storageComponent,),
)
),
(FilterMessages(allowed=['add']),
# (LogComponent("ADD"),),
(XmlXPath(['//document:document/document:part[@name="normdoc"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
# (LogComponent("NORMDOC"),),
(XmlParseLxml(fromKwarg='data', toKwarg='lxmlNode'),
(RewritePartname(NL_DIDL_NORMALISED_PREFIX), # Hernoemt partname van 'record' naar "metadata".
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,) # Schrijft oai:metadata (=origineel) naar storage.
)
)
)
),
(XmlXPath(['//document:document/document:part[@name="record"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
(XmlParseLxml(fromKwarg='data', toKwarg='lxmlNode'),
# TODO: Check indien conversies misgaan, dat ook de meta en header part niet naar storage gaan: geen enkel part als het even kan...
# Schrijf 'header' partname naar storage:
(XmlXPath(['/oai:record/oai:header'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP),
(RewritePartname("header"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,) # Schrijft OAI-header naar storage.
)
)
),
# Schrijf 'metadata' partname naar storage:
# Op gharvester21 gaat dit niet goed: Daar is het root element <metadata> in het 'metadata' part, in plaats van <DIDL>.
# Liever hier een child::node(), echter gaat deze syntax mis i.c.m. XmlXPath component??
(XmlXPath(['/oai:record/oai:metadata/didl:DIDL'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP),
# (LogComponent("METADATA_PART"),),
(RewritePartname("metadata"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,) # Schrijft metadata naar storage.
)
)
)
)
),
(XmlXPath(['//document:document/document:part[@name="record"]/text()'], fromKwarg='lxmlNode',
toKwarg='data', namespaces=NAMESPACEMAP),
(XmlParseLxml(fromKwarg='data', toKwarg='lxmlNode'),
(NlDidlCombined(nsMap=NAMESPACEMAP, fromKwarg='lxmlNode'),
# Create combined format from stored metadataPart and normalized part.
(XmlPrintLxml(fromKwarg='lxmlNode', toKwarg='data'), # Convert it to plaintext
(RewritePartname(NL_DIDL_COMBINED_PREFIX), # Rename combined partName
(storageComponent,) # Write combined partName to storage
)
)
)
)
),
(XmlXPath(['//document:document/document:part[@name="meta"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
(RewritePartname("meta"),
(storageComponent,) # Schrijft harvester 'meta' data naar storage.
)
),
(OaiAddRecord(metadataPrefixes=[('metadata', 'http://standards.iso.org/ittf/PubliclyAvailableStandards/MPEG-21_schema_files/did/didmodel.xsd', 'urn:mpeg:mpeg21:2002:02-DIDL-NS'),
(NL_DIDL_NORMALISED_PREFIX, '', NAMESPACEMAP.gmhnorm),
(NL_DIDL_COMBINED_PREFIX, '', NAMESPACEMAP.gmhcombined)]), #[(partname, schema, namespace)]
# (LogComponent("OaiAddRecord:"),),
(storageComponent,),
(oaiJazz,) # Assert partNames header and meta are available from storage!
),
(ResurrectTombstone(),
(storageComponent,),
),
),
# (FilterMessages(allowed=['add']),
# # (LogComponent("UnDelete"),),
# (ResurrectTombstone(),
# (storageComponent,),
# )
# )
)
)
)
)
def main(reactor, port, statePath, **ignored):
oaiSuspendRegister = SuspendRegister()
oaiJazz = be((OaiJazz(join(statePath, 'oai')), (oaiSuspendRegister, )))
# WST:
# strategie = HashDistributeStrategy() # filename (=partname) is also hashed: difficult to read by human eye...
strategie = Md5HashDistributeStrategy()
storeComponent = StorageComponent(
join(statePath, 'store'),
strategy=strategie,
partsRemovedOnDelete=[NORMALISED_DOC_NAME])
return \
(Observable(),
# (scheduledCommitPeriodicCall,),
# (DebugPrompt(reactor=reactor, port=port+1, globals=locals()),),
(ObservableHttpServer(reactor=reactor, port=port),
(BasicHttpHandler(),
(IpFilter(allowedIps=['127.0.0.1']),
(PathFilter('/oaix', excluding=['/oaix/info']),
(OaiPmh(repositoryName='Gateway',
adminEmail='*****@*****.**',
supportXWait=True,
batchSize=2000 # Override default batch size of 200.
),
(oaiJazz,),
(oaiSuspendRegister,),
(StorageAdapter(),
(storeComponent,),
),
)
),
(PathFilter('/oaix/info'),
(OaiInfo(reactor=reactor, oaiPath='/oai'),
(oaiJazz,),
)
),
),
(PathFilter('/update'),
(SruRecordUpdate(sendRecordData=False, logErrors=True,),
(FilterMessages(allowed=['delete']),
(storeComponent,),
(oaiJazz,),
),
(FilterMessages(allowed=['add']),
# Does not work? See comments in component...
# (AddMetadataFormat(fromKwarg="lxmlNode", name='md_format'),
# (LogComponent("AddMetadataFormat"),),
# ),
(XmlXPath(['srw:recordData/*'], fromKwarg='lxmlNode'), # Stuurt IEDERE matching node in een nieuw bericht door.
# (LogComponent("TO LONG CONVERTER:"),),
(AddMetadataNamespace(dateformat="%Y-%m-%dT%H:%M:%SZ", fromKwarg='lxmlNode'), # Adds metadataNamespace to meta part in the message.
(NormaliseOaiRecord(fromKwarg='lxmlNode'), # Normalises record to: long & original parts. Raises ValidationException if no 'known' metadataformat
(XmlPrintLxml(fromKwarg='lxmlNode', toKwarg='data', pretty_print=False),
(RewritePartname(NORMALISED_DOC_NAME), # Rename converted part.
(storeComponent,), # Store converted/renamed part.
)
)
),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[NORMALISED_DOC_NAME]),
(oaiJazz,),
)
)
)
)
)
)
)
)
)
def main(reactor,
port,
statePath,
lucenePort,
gatewayPort,
quickCommit=False,
**ignored):
######## START Lucene Integration ###############################################################
defaultLuceneSettings = LuceneSettings(
commitTimeout=30,
readonly=True,
)
http11Request = be((
HttpRequest1_1(),
(SocketPool(reactor=reactor,
unusedTimeout=5,
limits=dict(totalSize=100, destinationSize=10)), ),
))
luceneIndex = luceneAndReaderConfig(
defaultLuceneSettings.clone(readonly=True), http11Request, lucenePort)
luceneRoHelix = be(
(AdapterToLuceneQuery(defaultCore=DEFAULT_CORE,
coreConverters={
DEFAULT_CORE:
QueryExpressionToLuceneQueryDict(
UNQUALIFIED_TERM_FIELDS,
luceneSettings=luceneIndex.settings),
}), (
MultiLucene(host='localhost',
port=lucenePort,
defaultCore=DEFAULT_CORE),
(luceneIndex, ),
(http11Request, ),
)))
######## END Lucene Integration ###############################################################
fieldnameRewrites = {
# UNTOKENIZED_PREFIX+'genre': UNTOKENIZED_PREFIX+'dc:genre',
}
def fieldnameRewrite(name):
return fieldnameRewrites.get(name, name)
def drilldownFieldnamesTranslate(fieldname):
untokenizedName = untokenizedFieldname(fieldname)
if untokenizedName in untokenizedFieldnames:
fieldname = untokenizedName
return fieldnameRewrite(fieldname)
convertToComposedQuery = ConvertToComposedQuery(
resultsFrom=DEFAULT_CORE,
matches=[],
drilldownFieldnamesTranslate=drilldownFieldnamesTranslate)
strategie = Md5HashDistributeStrategy()
storage = StorageComponent(join(statePath, 'store'),
strategy=strategie,
partsRemovedOnDelete=[
HEADER_PARTNAME, META_PARTNAME,
METADATA_PARTNAME, OAI_DC_PARTNAME,
LONG_PARTNAME, SHORT_PARTNAME,
OPENAIRE_PARTNAME
])
oaiJazz = OaiJazz(join(statePath, 'oai'))
oaiJazz.updateMetadataFormat(
OAI_DC_PARTNAME, "http://www.openarchives.org/OAI/2.0/oai_dc.xsd",
"http://purl.org/dc/elements/1.1/")
oai_oa_cerifJazz = OaiJazz(join(statePath, 'oai_cerif'))
oai_oa_cerifJazz.updateMetadataFormat(
OPENAIRE_PARTNAME,
"https://www.openaire.eu/schema/cris/current/openaire-cerif-profile.xsd",
"https://www.openaire.eu/cerif-profile/1.1/")
# All of the following OAI-PMH sets shall be recognized by the CRIS, even if not all of them are populated.
oai_oa_cerifJazz.updateSet("openaire_cris_projects",
"OpenAIRE_CRIS_projects")
oai_oa_cerifJazz.updateSet("openaire_cris_orgunits",
"OpenAIRE_CRIS_orgunits")
oai_oa_cerifJazz.updateSet("openaire_cris_persons",
"OpenAIRE_CRIS_persons")
oai_oa_cerifJazz.updateSet("openaire_cris_patents",
"OpenAIRE_CRIS_patents")
oai_oa_cerifJazz.updateSet("openaire_cris_products",
"OpenAIRE_CRIS_products")
oai_oa_cerifJazz.updateSet("openaire_cris_publications",
"OpenAIRE_CRIS_publications")
oai_oa_cerifJazz.updateSet("openaire_cris_funding",
"OpenAIRE_CRIS_funding")
oai_oa_cerifJazz.updateSet("openaire_cris_events", "OpenAIRE_CRIS_events")
oai_oa_cerifJazz.updateSet("openaire_cris_equipments",
"OpenAIRE_CRIS_equipments")
cqlClauseConverters = [
RenameFieldForExact(
untokenizedFields=untokenizedFieldnames,
untokenizedPrefix=UNTOKENIZED_PREFIX,
).filterAndModifier(),
SearchTermFilterAndModifier(
shouldModifyFieldValue=lambda *args: True,
fieldnameModifier=fieldnameRewrite).filterAndModifier(),
]
periodicGateWayDownload = PeriodicDownload(
reactor,
host='localhost',
port=gatewayPort,
schedule=Schedule(
period=1 if quickCommit else 10
), # WST: Interval in seconds before sending a new request to the GATEWAY in case of an error while processing batch records.(default=1). IntegrationTests need 1 second! Otherwise tests will fail!
name='api',
autoStart=True)
oaiDownload = OaiDownloadProcessor(path='/oaix',
metadataPrefix=NORMALISED_DOC_NAME,
workingDirectory=join(
statePath, 'harvesterstate',
'gateway'),
userAgentAddition='ApiServer',
xWait=True,
name='api',
autoCommit=False)
executeQueryHelix = \
(FilterMessages(allowed=['executeQuery']),
(CqlMultiSearchClauseConversion(cqlClauseConverters, fromKwarg='query'),
(DrilldownQueries(),
(convertToComposedQuery,
(luceneRoHelix,),
)
)
)
)
return \
(Observable(),
createDownloadHelix(reactor, periodicGateWayDownload, oaiDownload, storage, oaiJazz, oai_oa_cerifJazz),
(ObservableHttpServer(reactor, port, compressResponse=True),
(BasicHttpHandler(),
(PathFilter(["/oai"]),
(OaiPmh(repositoryName="NARCIS OAI-pmh", adminEmail="*****@*****.**", externalUrl="http://oai.narcis.nl"),
(oaiJazz,),
(StorageAdapter(),
(storage,)
),
(OaiBranding(
url="http://www.narcis.nl/images/logos/logo-knaw-house.gif",
link="http://oai.narcis.nl",
title="Narcis - The gateway to scholarly information in The Netherlands"),
),
(OaiProvenance(
nsMap=NAMESPACEMAP,
baseURL=('meta', '//meta:repository/meta:baseurl/text()'),
harvestDate=('meta', '//meta:record/meta:harvestdate/text()'),
metadataNamespace=('meta', '//meta:record/meta:metadataNamespace/text()'),
identifier=('header','//oai:identifier/text()'),
datestamp=('header', '//oai:datestamp/text()')
),
(storage,)
)
)
),
(PathFilter(["/cerif"]),
(OaiPmhDans(repositoryName="OpenAIRE CERIF", adminEmail="*****@*****.**", repositoryIdentifier="services.nod.dans.knaw.nl", externalUrl="http://services.nod.dans.knaw.nl"), #TODO: pathFilter should resemble proxy path
(oai_oa_cerifJazz,),
(StorageAdapter(),
(storage,)
),
(OaiOpenAIREDescription(
serviceid='organisation:ORG1242054',
acronym='services.nod.dans.knaw.nl',
name='NARCIS',
description='Compliant with the OpenAIRE Guidelines for CRIS Managers v.1.1.',
website='https://www.narcis.nl',
baseurl='http://services.nod.dans.knaw.nl/oa-cerif',
subjectheading='',
orgunitid='organisation:ORG1242054',
owneracronym='DANS'),
),
# (OaiBranding(
# url="http://www.narcis.nl/images/logos/logo-knaw-house.gif",
# link="http://oai.narcis.nl",
# title="Narcis - The gateway to scholarly information in The Netherlands"),
# ),
(OaiProvenance(
nsMap=NAMESPACEMAP,
baseURL=('meta', '//meta:repository/meta:baseurl/text()'),
harvestDate=('meta', '//meta:record/meta:harvestdate/text()'),
metadataNamespace=('meta', '//meta:record/meta:metadataNamespace/text()'),
identifier=('header','//oai:identifier/text()'),
datestamp=('header', '//oai:datestamp/text()')
),
(storage,)
)
)
),
(PathFilter(['/sru']),
(SruParser(
host='sru.narcis.nl',
port=80,
defaultRecordSchema='knaw_short',
defaultRecordPacking='xml'),
(SruLimitStartRecord(limitBeyond=4000),
(SruHandler(
includeQueryTimes=False,
extraXParameters=[],
enableCollectLog=False),
(SruTermDrilldown(),),
executeQueryHelix,
(StorageAdapter(),
(storage,)
)
)
)
)
),
(PathFilter('/rss'),
(Rss( supportedLanguages = ['nl','en'], # defaults to first, if requested language is not available or supplied.
title = {'nl':'NARCIS', 'en':'NARCIS'},
description = {'nl':'NARCIS: De toegang tot de Nederlandse wetenschapsinformatie', 'en':'NARCIS: The gateway to Dutch scientific information'},
link = {'nl':'http://www.narcis.nl/?Language=nl', 'en':'http://www.narcis.nl/?Language=en'},
maximumRecords = 20),
executeQueryHelix,
(RssItem(
nsMap=NAMESPACEMAP,
title = ('knaw_short', {'nl':'//short:metadata/short:titleInfo[not (@xml:lang)]/short:title/text()', 'en':'//short:metadata/short:titleInfo[@xml:lang="en"]/short:title/text()'}),
description = ('knaw_short', {'nl':'//short:abstract[not (@xml:lang)]/text()', 'en':'//short:abstract[@xml:lang="en"]/text()'}),
pubdate = ('knaw_short', '//short:dateIssued/short:parsed/text()'),
linkTemplate = 'http://www.narcis.nl/%(wcpcollection)s/RecordID/%(oai_identifier)s/Language/%(language)s',
wcpcollection = ('meta', '//*[local-name() = "collection"]/text()'),
oai_identifier = ('meta', '//meta:record/meta:id/text()'),
language = ('Dummy: Language is auto provided by the calling RSS component, but needs to be present to serve the linkTemplate.')
),
(StorageAdapter(),
(storage,)
)
)
)
)
)
)
)
def createDownloadHelix(reactor, periodicDownload, oaiDownload,
storageComponent, oaiJazz, oai_oa_cerifJazz):
return \
(periodicDownload, # Scheduled connection to a remote (response / request)...
(XmlParseLxml(fromKwarg="data", toKwarg="lxmlNode", parseOptions=dict(huge_tree=True, remove_blank_text=True)), # Convert from plain text to lxml-object.
(oaiDownload, # Implementation/Protocol of a PeriodicDownload...
(UpdateAdapterFromOaiDownloadProcessor(), # Maakt van een SRU update/delete bericht (lxmlNode) een relevante message: 'delete' of 'add' message.
(FilterMessages(['delete']), # Filtert delete messages
# (LogComponent("Delete Update"),),
(storageComponent,), # Delete from storage
(oaiJazz,), # Delete from OAI-pmh repo
(oai_oa_cerifJazz,),
# Write a 'deleted' part to the storage, that holds the (Record)uploadId.
(WriteTombstone(),
(storageComponent,),
)
),
(FilterMessages(allowed=['add']),
(XmlXPath(['/oai:record/oai:metadata/document:document/document:part[@name="record"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
(XmlParseLxml(fromKwarg='data', toKwarg='lxmlNode'),
(FilterWcpCollection(allowed=['research']),
(XmlXPath(['/oai:record/oai:metadata/norm:md_original/child::*'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Origineel 'metadata' formaat
(XsltCrosswalk([join(dirname(abspath(__file__)), '..', '..', 'xslt', 'cerif-project.xsl')], fromKwarg="lxmlNode"),
(RewritePartname(OPENAIRE_PARTNAME),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,)
)
)
)
)
),
(FilterWcpCollection(allowed=['person']),
(XmlXPath(['/oai:record/oai:metadata/norm:md_original/child::*'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Origineel 'metadata' formaat
(XsltCrosswalk([join(dirname(abspath(__file__)), '..', '..', 'xslt', 'cerif-person.xsl')], fromKwarg="lxmlNode"),
(RewritePartname(OPENAIRE_PARTNAME),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,)
)
)
)
)
),
(FilterWcpCollection(allowed=['organisation']),
(XmlXPath(['/oai:record/oai:metadata/norm:md_original/child::*'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Origineel 'metadata' formaat
(XsltCrosswalk([join(dirname(abspath(__file__)), '..', '..', 'xslt', 'cerif-orgunit.xsl')], fromKwarg="lxmlNode"),
(RewritePartname(OPENAIRE_PARTNAME),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,)
)
)
)
)
),
(FilterWcpCollection(allowed=['dataset']), # START CERIF CONVERSION FOR DATASET COLLECTION: cerif-dataset / cerif-software.
(XmlXPath(['/oai:record/oai:metadata/norm:normalized/long:knaw_long'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Genormaliseerd 'long' formaat.
(XsltCrosswalk([join(dirname(abspath(__file__)), '..', '..', 'xslt', 'cerif-product.xsl')], fromKwarg="lxmlNode"),
(RewritePartname(OPENAIRE_PARTNAME),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,)
)
)
)
)
),
(FilterWcpCollection(allowed=['publication']), # START CERIF CONVERSION FOR PUBLICATIONS COLLECTION
(XmlXPath(['/oai:record/oai:metadata/norm:normalized/long:knaw_long'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Genormaliseerd 'long' formaat.
(FilterKnawLongGenre(allowed=['patent']), # START PATENTS CONVERSION
(XsltCrosswalk([join(dirname(abspath(__file__)), '..', '..', 'xslt', 'cerif-patent.xsl')], fromKwarg="lxmlNode"),
(RewritePartname(OPENAIRE_PARTNAME),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,)
)
)
)
),
(FilterKnawLongGenre(disallowed=['patent']), # START Publication CONVERSION
(XsltCrosswalk([join(dirname(abspath(__file__)), '..', '..', 'xslt', 'cerif-publication.xsl')], fromKwarg="lxmlNode"),
(RewritePartname(OPENAIRE_PARTNAME),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,)
)
)
)
)
)
),
(XmlXPath(['/oai:record/oai:metadata/norm:md_original/child::*'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Origineel 'metadata' formaat
(RewritePartname("metadata"), # Hernoemt partname van 'record' naar "metadata".
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,) # Schrijft oai:metadata (=origineel) naar storage.
)
)
),
(XmlXPath(['/oai:record/oai:metadata/norm:normalized/long:knaw_long'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Genormaliseerd 'long' formaat.
(RewritePartname("knaw_long"), # Hernoemt partname van 'record' naar "knaw_long".
(FilterWcpCollection(disallowed=['person', 'research', 'organisation']),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,), # Schrijft 'long' (=norm:normdoc) naar storage.
)
),
(ShortConverter(fromKwarg='lxmlNode'), # creeer 'knaw_short' subset formaat.
(RewritePartname("knaw_short"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,) # Schrijft 'short' naar storage.
)
)
),
(FilterWcpCollection(disallowed=['person', 'research', 'organisation']),
(DcConverter(fromKwarg='lxmlNode'), # Hernoem partname van 'record' naar "oai_dc".
(RewritePartname("oai_dc"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,) # Schrijft 'oai_dc' naar storage.
)
)
)
)
)
),
# TODO: Check indien conversies misgaan, dat ook de meta en header part niet naar storage gaan: geen 1 part als het even kan...
# Schrijf 'header' partname naar storage:
(XmlXPath(['/oai:record/oai:header'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP),
(RewritePartname("header"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,) # Schrijft OAI-header naar storage.
)
)
),
(FilterWcpCollection(allowed=['publication']),
# (LogComponent("PUBLICATION"),),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['publication']),
(oaiJazz,),
),
(XmlXPath(["//long:knaw_long[long:accessRights ='openAccess']"], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['oa_publication', 'openaire']),
(oaiJazz,),
)
),
(XmlXPath(["//long:knaw_long/long:metadata[long:genre ='doctoralthesis']"], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['thesis']),
(oaiJazz,),
)
),
(XmlXPath(['//long:knaw_long/long:metadata/long:grantAgreements/long:grantAgreement[long:code[contains(.,"greement/EC/") or contains(.,"greement/ec/")]][1]'], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['ec_fundedresources', 'openaire']),
(oaiJazz,),
)
),
(XmlXPath(["//long:knaw_long/long:metadata[long:genre ='patent']"], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[OPENAIRE_PARTNAME], setSpecs=["openaire_cris_patents"]),
(oai_oa_cerifJazz,),
)
),
(XmlXPath(["//long:knaw_long/long:metadata[long:genre !='patent']"], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[OPENAIRE_PARTNAME], setSpecs=["openaire_cris_publications"]),
(oai_oa_cerifJazz,),
)
),
),
(FilterWcpCollection(allowed=['dataset']),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['dataset']),
(oaiJazz,),
),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[OPENAIRE_PARTNAME], setSpecs=["openaire_cris_products"]),
(oai_oa_cerifJazz,),
)
),
# Add NOD OpenAIRE Cerif to OpenAIRE-PMH repo.
(FilterWcpCollection(allowed=['research']),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[OPENAIRE_PARTNAME], setSpecs=["openaire_cris_projects"]),
(oai_oa_cerifJazz,),
)
),
(FilterWcpCollection(allowed=['person']),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[OPENAIRE_PARTNAME], setSpecs=['openaire_cris_persons']),
(oai_oa_cerifJazz,),
)
),
(FilterWcpCollection(allowed=['organisation']),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[OPENAIRE_PARTNAME], setSpecs=['openaire_cris_orgunits']),
(oai_oa_cerifJazz,),
)
)
)
), # Schrijf 'meta' partname naar storage:
(XmlXPath(['/oai:record/oai:metadata/document:document/document:part[@name="meta"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
(RewritePartname("meta"),
(storageComponent,) # Schrijft harvester 'meta' data naar storage.
)
)
),
(FilterMessages(allowed=['add']),
# (LogComponent("UnDelete"),),
(ResurrectTombstone(),
(storageComponent,),
)
)
)
)
)
)
def createUploadHelix(storageComponent, oaiJazz, loggerComponent):
return \
(TransactionScope('batch'),
(TransactionScope('record'),
(Venturi(
should=[ # Order DOES matter: First part goes first!
{'partname':'header', 'xpath':'/document:document/document:part[@name="header"]/text()', 'asString':False},
{'partname':'meta', 'xpath':'/document:document/document:part[@name="meta"]/text()', 'asString':False},
{'partname':'metadata', 'xpath':'/document:document/document:part[@name="metadata"]/text()', 'asString':False}
],
namespaceMap=namespacesMap),
# Remove all delete msgs from storage and OAI:
(FilterMessages(allowed=['delete']),
#(DNADebug(enabled=False, prefix='DELETE'),
(storageComponent,),
(oaiJazz,)
#)
),
(FilterMessages(allowed=['add']),
## Write harvestdate (=now()) to meta part (OAI provenance)
(FilterPartByName(included=['meta']),
(AddHarvestDateToMetaPart(verbose=False),)
),
# Store ALL (original)parts retrieved by Venturi (required ('should') and optional ('could') parts).
# Write all uploadParts to storage (header, meta & metadata)
(XmlPrintLxml(fromKwarg='lxmlNode', toKwarg='data'),
(storageComponent,)
),
(FilterPartByName(included=['metadata']), # Normalize 'metadata' part:
#(DNADebug(enabled=False, prefix='add metadata'),
# Validate DIDL and MODS part against their xsd-schema:
(Validate([('DIDL container','//didl:DIDL', 'didl.xsd'), ('MODS metadata', '//mods:mods', 'mods-3-6.xsd')], nsMap=namespacesMap),
(Normalize_nl_DIDL(nsMap=namespacesMap), # Normalize DIDL in metadataPart
(loggerComponent,),
(Normalize_nl_MODS(nsMap=namespacesMap), # Normalize MODS in metadataPart.
(loggerComponent,),
(XmlPrintLxml(fromKwarg='lxmlNode', toKwarg='data'), # Convert it from etree.ElementTree to plaintext
(RewritePartname(NL_DIDL_NORMALISED_PREFIX), # Rename normalized partName from 'metadata' to 'nl_didl_norm'
#(DNADebug(enabled=False, prefix='to storage'),
(storageComponent,) # Write normalized partName to storage
#)
)
),
# Create and store Combined format:
(NL_DIDL_combined(nsMap=namespacesMap), # Create combined format from stored metadataPart and normalized part.
(XmlPrintLxml(fromKwarg='lxmlNode', toKwarg='data'), # Convert it to plaintext
(RewritePartname(NL_DIDL_COMBINED_PREFIX), # Rename combined partName
(storageComponent,) # Write combined partName to storage
)
)
),
# Add parts to OAI repository/index
#(DNADebug(enabled=False, prefix='ADD2OAI'),
(OaiAddRecordWithDefaults(metadataFormats=[('metadata', 'http://standards.iso.org/ittf/PubliclyAvailableStandards/MPEG-21_schema_files/did/didmodel.xsd', 'urn:mpeg:mpeg21:2002:02-DIDL-NS'),
(NL_DIDL_NORMALISED_PREFIX, '', 'http://gh.kb-dans.nl/normalised/v0.9/'),
(NL_DIDL_COMBINED_PREFIX, '', 'http://gh.kb-dans.nl/combined/v0.9/')]),
(storageComponent,),
(oaiJazz,) # Assert partNames header and meta are available from storage!
) #! OaiAddRecord
#) #!Debug
)
)
)
#) #Debug
) #!FilterPartNames(allowed=['metadata']
) # !FilterMessages(allowed=['add']
) # !venturi
) # !record
) # !batch
def createDownloadHelix(reactor, periodicDownload, oaiDownload,
storageComponent, oaiJazz):
return \
(periodicDownload, # Scheduled connection to a remote (response / request)...
(XmlParseLxml(fromKwarg="data", toKwarg="lxmlNode", parseOptions=dict(huge_tree=True, remove_blank_text=True)), # Convert from plain text to lxml-object.
(oaiDownload, # Implementation/Protocol of a PeriodicDownload...
(UpdateAdapterFromOaiDownloadProcessor(), # Maakt van een SRU update/delete bericht (lxmlNode) een relevante message: 'delete' of 'add' message.
(FilterMessages(['delete']), # Filtert delete messages
# (LogComponent("Delete Update"),),
(storageComponent,), # Delete from storage
(oaiJazz,), # Delete from OAI-pmh repo
# Write a 'deleted' part to the storage, that holds the (Record)uploadId.
(WriteTombstone(),
(storageComponent,),
)
),
(FilterMessages(allowed=['add']),
# TODO: onderstaande toKwarg='data' kan eruit. Dan de volgende regel ook:-)
(XmlXPath(['/oai:record/oai:metadata/document:document/document:part[@name="record"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
(XmlParseLxml(fromKwarg='data', toKwarg='lxmlNode'),
(XmlXPath(['/oai:record/oai:metadata/norm:md_original/child::*'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Origineel 'metadata' formaat
(RewritePartname("metadata"), # Hernoemt partname van 'record' naar "metadata".
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,) # Schrijft oai:metadata (=origineel) naar storage.
)
)
),
(XmlXPath(['/oai:record/oai:metadata/norm:normalized/long:knaw_long'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP), # Genormaliseerd 'long' formaat.
(RewritePartname("knaw_long"), # Hernoemt partname van 'record' naar "knaw_long".
(FilterWcpCollection(disallowed=['person', 'research', "organisation"]),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,), # Schrijft 'long' (=norm:normdoc) naar storage.
)
),
(ShortConverter(fromKwarg='lxmlNode'), # creeer 'knaw_short' subset formaat.
(RewritePartname("knaw_short"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,) # Schrijft 'short' naar storage.
)
)
),
(FilterWcpCollection(disallowed=['person', 'research', "organisation"]),
(DcConverter(fromKwarg='lxmlNode'), # Hernoem partname van 'record' naar "oai_dc".
(RewritePartname("oai_dc"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=True),
(storageComponent,) # Schrijft 'oai_dc' naar storage.
)
)
)
)
)
),
# TODO: Check indien conversies misgaan, dat ook de meta en header part niet naar storage gaan: geen 1 part als het even kan...
# Schrijf 'header' partname naar storage:
(XmlXPath(['/oai:record/oai:header'], fromKwarg='lxmlNode', namespaces=NAMESPACEMAP),
(RewritePartname("header"),
(XmlPrintLxml(fromKwarg="lxmlNode", toKwarg="data", pretty_print=False),
(storageComponent,) # Schrijft OAI-header naar storage.
)
)
),
(FilterWcpCollection(allowed=['publication']),
# (LogComponent("PUBLICATION"),),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['publication'], name='NARCISPORTAL'), #TODO: Skip name='NARCISPORTAL'
(oaiJazz,),
),
(XmlXPath(["//long:knaw_long[long:accessRights ='openAccess']"], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
# (LogComponent("OPENACCESS"),),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['oa_publication', 'openaire'], name='NARCISPORTAL'),
(oaiJazz,),
)
),
(XmlXPath(["//long:knaw_long/long:metadata[long:genre ='doctoralthesis']"], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['thesis'], name='NARCISPORTAL'),
(oaiJazz,),
)
),
(XmlXPath(['//long:knaw_long/long:metadata/long:grantAgreements/long:grantAgreement[long:code[contains(.,"greement/EC/") or contains(.,"greement/ec/")]][1]'], fromKwarg='lxmlNode', namespaceMap=NAMESPACEMAP),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['ec_fundedresources', 'openaire'], name='NARCISPORTAL'),
(oaiJazz,),
)
)
),
(FilterWcpCollection(allowed=['dataset']),
# (LogComponent("DATASET"),),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=["oai_dc"], setSpecs=['dataset'], name='NARCISPORTAL'),
(oaiJazz,),
)
)
)
), # Schrijf 'meta' partname naar storage:
(XmlXPath(['/oai:record/oai:metadata/document:document/document:part[@name="meta"]/text()'], fromKwarg='lxmlNode', toKwarg='data', namespaces=NAMESPACEMAP),
(RewritePartname("meta"),
(storageComponent,) # Schrijft harvester 'meta' data naar storage.
)
)
),
(FilterMessages(allowed=['add']), # TODO: Remove this line.
# (LogComponent("UnDelete"),),
(ResurrectTombstone(),
(storageComponent,),
)
)
)
)
)
)
def main(reactor, port, statePath, **ignored):
oaiSuspendRegister = SuspendRegister()
oaiJazz = be((OaiJazz(join(statePath, 'oai'),
alwaysDeleteInPrefixes=[NORMALISED_DOC_NAME]),
(oaiSuspendRegister, )))
normLogger = Logger(join(statePath, 'normlogger'))
# strategie = HashDistributeStrategy() # filename (=partname) is also hashed: difficult to read by human eye...
strategie = Md5HashDistributeStrategy()
storeComponent = StorageComponent(
join(statePath, 'store'),
strategy=strategie,
partsRemovedOnDelete=[NORMALISED_DOC_NAME])
return \
(Observable(),
# (scheduledCommitPeriodicCall,),
# (DebugPrompt(reactor=reactor, port=port+1, globals=locals()),),
(ObservableHttpServer(reactor=reactor, port=port),
(BasicHttpHandler(),
(IpFilter(allowedIps=['127.0.0.1']),
(PathFilter('/oaix', excluding=['/oaix/info']),
(OaiPmh(repositoryName='Gateway',
adminEmail='*****@*****.**',
supportXWait=True,
batchSize=2000 # Override default batch size of 200.
),
(oaiJazz,),
(oaiSuspendRegister,),
(StorageAdapter(),
(storeComponent,),
),
)
),
(PathFilter('/oaix/info'),
(OaiInfo(reactor=reactor, oaiPath='/oai'),
(oaiJazz,),
)
),
),
(PathFilter('/update'),
(SruRecordUpdate(sendRecordData=False, logErrors=True,),
(FilterMessages(allowed=['delete']),
(storeComponent,),
(oaiJazz,),
),
(FilterMessages(allowed=['add']),
# (LogComponent("LXML:"),),
(Validate([('DIDL container','//didl:DIDL', 'didl.xsd'), ('MODS metadata', '//mods:mods', 'mods-3-6.xsd')]),
# (LogComponent("VALIDATED:"),),
(AddMetadataDocumentPart(partName='normdoc', fromKwarg='lxmlNode'),
(NormaliseDIDL(nsMap=namespacesMap, fromKwarg='lxmlNode'), # Normalise DIDL in partname=normdoc metadata
(normLogger,),
(NormaliseMODS(nsMap=namespacesMap, fromKwarg='lxmlNode'), # Normalise MODS in partname=normdoc metadata
(normLogger,),
(XmlPrintLxml(fromKwarg='lxmlNode', toKwarg='data'),
(RewritePartname(NORMALISED_DOC_NAME), # Rename converted part.
(storeComponent,), # Store converted/renamed part.
)
),
(OaiAddDeleteRecordWithPrefixesAndSetSpecs(metadataPrefixes=[NORMALISED_DOC_NAME]),
(oaiJazz,),
)
)
)
)
)
)
)
)
)
)
)
def testEitherAllowedOrDisallowed(self):
self.assertRaises(
ValueError,
lambda: FilterMessages(allowed=['either'], disallowed=['or']))
def writerMain(writerReactor, statePath, luceneserverPort, gatewayPort, quickCommit=False):
http11Request = be(
(HttpRequest1_1(),
(SocketPool(reactor=writerReactor, unusedTimeout=5, limits=dict(totalSize=100, destinationSize=10)),),
)
)
indexCommitTimeout = 30
defaultLuceneSettings = LuceneSettings(
commitTimeout=indexCommitTimeout,
readonly=False,
)
luceneWriter = luceneAndReaderConfig(defaultLuceneSettings, http11Request, luceneserverPort)
periodicDownload = PeriodicDownload(
writerReactor,
host='localhost',
port=gatewayPort,
schedule=Schedule(period=1 if quickCommit else 10), # WST: Interval in seconds before sending a new request to the GATEWAY in case of an error while processing batch records.(default=1). IntegrationTests need 1 second! Otherwise tests will fail!
name='index',
autoStart=True)
oaiDownload = OaiDownloadProcessor(
path='/oaix',
metadataPrefix=NORMALISED_DOC_NAME,
workingDirectory=join(statePath, 'harvesterstate', 'gateway'),
userAgentAddition='idx-server',
xWait=True,
name='index',
autoCommit=False)
# Post commit naar Lucene(server):
scheduledCommitPeriodicCall = be(
(PeriodicCall(writerReactor, message='commit', name='Scheduled commit', schedule=Schedule(period=1 if quickCommit else 300), initialSchedule=Schedule(period=1)), # WST: Flushes data from memory to disk. IntegrationTests need 1 second! Otherwise tests will fail! (API).
(AllToDo(), # broadcast message to all components, despite of what kind of message...
# (periodicDownload,), # WST: periodicDownload does not do anything with a 'commit' message? So why send it to it???
(LuceneCommit(host='localhost', port=luceneserverPort,), # 'commit' message results in http post to /commit/ to Lucene server:
# (LogComponent("PERIODIC"),#), # [PERIODIC] httprequest1_1(*(), **{'body': None, 'host': 'localhost', 'request': '/commit/', 'port': 52501, 'method': 'POST'})
(http11Request,),
# ),
)
)
)
)
writerServer = \
(Observable(),
(scheduledCommitPeriodicCall,), # Stuur periodiek een 'Commit' naar de LuceneServer...
# (DebugPrompt(reactor=writerReactor, port=readerPort-1, globals=locals()),),
(periodicDownload, # Ga/connect (periodiek) naar de Gateway-server...
(XmlParseLxml(fromKwarg="data", toKwarg="lxmlNode", parseOptions=dict(huge_tree=True, remove_blank_text=True)),
(oaiDownload, # Haal OAI spulletjes van de Gateway...
(UpdateAdapterFromOaiDownloadProcessor(), # Maakt van een SRU update/delete bericht (lxmlNode) een relevante message: 'delete' of 'add' message.
# (LogComponent("SRU harvest van GATEWAY"),), #[SRU harvest van GATEWAY] add(*(), **{'partname': 'record', 'identifier': 'meresco:record:1', 'lxmlNode': '_ElementTree(<record xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"><header><identifier>meresco:record:1</identifier><datestamp>2016-07-13T15:31:10Z</datestamp></header><metadata><document xmlns="http://meres
(FilterMessages(allowed=['add']),
(XmlXPath(['/oai:record/oai:metadata/document:document'], fromKwarg='lxmlNode'),
# (LogComponent("NormdocToFieldsList"),),
(NormdocToFieldsList(), # Platte lijst met veldnamen en waardes...
(RecordPidToAuthNid(),),
# (LogComponent("NormdocToFieldsList"),), # [DcToFieldsList] add(*(), **{'fieldslist': [('dc:identifier', 'http://meresco.com?record=1'), ('dc:description', 'This is an example program about Search with Meresco'), ('dc:title', 'Example Program 1'), ('dc:creator', 'Seecr'), ('dc:publisher', 'Seecr'), ('dc:date', '2016'), ('dc:type', 'Example'), ('dc:subject', 'Search'), ('dc:language', 'en'), ('dc:rights', 'Open Source')], 'partname': 'record', 'identifier': 'meresco:record:1'})
# [NormdocToFieldsList] lookupNameIds(*(set(['doi:10.1002/lno.10611', 'wos:000423029300003']),), **{})
(FieldsListToLuceneDocument( # Maakt addDocument messege + creeert de facet/drilldown velden waarvan de value's tot max. 256 chars getruncated worden.
fieldRegistry=luceneWriter.settings.fieldRegistry, # o.a. drilldownfields definitie
untokenizedFieldnames=untokenizedFieldnames, # untokenized fields
indexFieldFactory=DcFields, # Creeert een "__all__", veldnaam en optioneel "untokenized.veldnaam"...
#rewriteIdentifier=(lambda idee: idee.split(':', 1)[-1]) # meresco:record:1' => 'record:1'
),
# (LogComponent("FieldsListToLuceneDocument"),), # [LUCENE_WRITER] addDocument(*(), **{'fields': [{'type': 'TextField', 'name': '__all__', 'value': 'http://meresco.com?record=1'}, {'type': 'TextField', 'name': 'dc:identifier', 'value': 'http://meresco.com?record=1'}, {'type': 'StringField', 'name': 'untokenized.dc:identifier', 'value': 'http://meresco.com?record=1'}, {'type': 'TextField', 'name': '__all__', 'value': 'This is an example program about Search with Meresco'}, {'type': 'TextField', 'name': 'dc:description', 'value': 'This is an example program about Search with Meresco'}, {'type': 'TextField', 'name': '__all__', 'value': 'Example Program 1'}, {'type': 'TextField', 'name': 'dc:title', 'value': 'Example Program 1'}, {'type': 'TextField', 'name': '__all__', 'value': 'Seecr'}, {'type': 'TextField', 'name': 'dc:creator', 'value': 'Seecr'}, {'type': 'TextField', 'name': '__all__', 'value': 'Seecr'}, {'type': 'TextField', 'name': 'dc:publisher', 'value': 'Seecr'}, {'type': 'TextField', 'name': '__all__', 'value': '2016'}, {'type': 'TextField', 'name': 'dc:date', 'value': '2016'}, {'path': ['2016'], 'type': 'FacetField', 'name': 'untokenized.dc:date'}, {'type': 'TextField', 'name': '__all__', 'value': 'Example'}, {'type': 'TextField', 'name': 'dc:type', 'value': 'Example'}, {'type': 'TextField', 'name': '__all__', 'value': 'Search'}, {'type': 'TextField', 'name': 'dc:subject', 'value': 'Search'}, {'path': ['Search'], 'type': 'FacetField', 'name': 'untokenized.dc:subject'}, {'type': 'TextField', 'name': '__all__', 'value': 'en'}, {'type': 'TextField', 'name': 'dc:language', 'value': 'en'}, {'type': 'TextField', 'name': '__all__', 'value': 'Open Source'}, {'type': 'TextField', 'name': 'dc:rights', 'value': 'Open Source'}], 'identifier': 'meresco:record:1'})
# [####LUCENE_WRITER] addDocument(*(), **{'fields': [{'type': 'TextField', 'name': '__all__', 'value': 'knaw'}, {'type': 'TextField', 'name': 'meta:id', 'value': 'knaw'}, {'type': 'TextField', 'name': '__all__', 'value': 'olddata'}, {'type': 'TextField', 'name': 'meta:set', 'value': 'olddata'}, {'type': 'TextField', 'name': '__all__', 'value': 'http://oai.knaw.nl/oai'}, {'type': 'TextField', 'name': 'meta:baseurl', 'value': 'http://oai.knaw.nl/oai'}, {'type': 'TextField', 'name': '__all__', 'value': 'knaw'}, {'type': 'TextField', 'name': 'meta:repositoryGroupId', 'value': 'knaw'}, {'type': 'TextField', 'name': '__all__', 'value': 'nl_didl'}, {'type': 'TextField', 'name': 'meta:metadataPrefix', 'value': 'nl_didl'}, {'type': 'TextField', 'name': '__all__', 'value': 'publication'}, {'type': 'TextField', 'name': 'meta_collection', 'value': 'publication'}, {'path': ['publication'], 'type': 'FacetField', 'name': 'untokenized.meta_collection'}], 'identifier': 'knaw:record:3'})
(luceneWriter,),
# ),
)
)
# )
# )
)
),
(FilterMessages(allowed=['delete']),
(luceneWriter,),
)
)
)
)
)
)
return writerServer
def main(reactor, port, statePath, lucenePort, **ignored):
######## START Lucene Integration ###############################################################
defaultLuceneSettings = LuceneSettings(
commitTimeout=30,
readonly=True,)
http11Request = be(
(HttpRequest1_1(),
(SocketPool(reactor=reactor, unusedTimeout=5, limits=dict(totalSize=100, destinationSize=10)),),
)
)
luceneIndex = luceneAndReaderConfig(defaultLuceneSettings.clone(readonly=True), http11Request, lucenePort)
luceneRoHelix = be(
(AdapterToLuceneQuery(
defaultCore=DEFAULT_CORE,
coreConverters={
DEFAULT_CORE: QueryExpressionToLuceneQueryDict(UNQUALIFIED_TERM_FIELDS, luceneSettings=luceneIndex.settings),
}
),
(MultiLucene(host='127.0.0.1', port=lucenePort, defaultCore=DEFAULT_CORE),
(luceneIndex,),
(http11Request,),
)
)
)
######## END Lucene Integration ###############################################################
fieldnameRewrites = {}
def fieldnameRewrite(name):
return fieldnameRewrites.get(name, name)
def drilldownFieldnamesTranslate(fieldname):
untokenizedName = untokenizedFieldname(fieldname)
if untokenizedName in untokenizedFieldnames:
fieldname = untokenizedName
return fieldnameRewrite(fieldname)
convertToComposedQuery = ConvertToComposedQuery(
resultsFrom=DEFAULT_CORE,
matches=[],
drilldownFieldnamesTranslate=drilldownFieldnamesTranslate
)
strategie = Md5HashDistributeStrategy()
storage = StorageComponent(join(statePath, 'store'), strategy=strategie, partsRemovedOnDelete=[HEADER_PARTNAME, META_PARTNAME, METADATA_PARTNAME, OAI_DC_PARTNAME, LONG_PARTNAME, SHORT_PARTNAME])
# Wat doet dit?
cqlClauseConverters = [
RenameFieldForExact(
untokenizedFields=untokenizedFieldnames,
untokenizedPrefix=UNTOKENIZED_PREFIX,
).filterAndModifier(),
SearchTermFilterAndModifier(
shouldModifyFieldValue=lambda *args: True,
fieldnameModifier=fieldnameRewrite
).filterAndModifier(),
]
executeQueryHelix = \
(FilterMessages(allowed=['executeQuery']),
(CqlMultiSearchClauseConversion(cqlClauseConverters, fromKwarg='query'),
(DrilldownQueries(),
(convertToComposedQuery,
(luceneRoHelix,),
)
)
)
)
return \
(Observable(),
(ObservableHttpServer(reactor, port, compressResponse=True),
(BasicHttpHandler(),
(PathFilter(['/sru']),
(SruParser(
host='sru.narcis.nl',
port=80,
defaultRecordSchema='knaw_short',
defaultRecordPacking='xml'),
(SruLimitStartRecord(limitBeyond=4000),
(SruHandler(
includeQueryTimes=False,
extraXParameters=[],
enableCollectLog=False), #2017-03-24T12:00:33Z 127.0.0.1 3.5K 0.019s - /sru OF (TRUE): 2017-03-24T11:58:53Z 127.0.0.1 2.3K 0.004s 1hits /sru maximumRecords=10&operation=searchRetrieve&query=untokenized.dd_year+exact+%221993%22&recordPacking=xml&recordSchema=knaw_short&startRecord=1&version=1.2
(SruTermDrilldown(),),
executeQueryHelix,
(StorageAdapter(),
(storage,)
)
)
)
)
),
(PathFilter('/rss'),
(Rss( supportedLanguages = ['nl','en'], # defaults to first, if requested language is not available or supplied.
title = {'nl':'NARCIS', 'en':'NARCIS'},
description = {'nl':'NARCIS: De toegang tot de Nederlandse wetenschapsinformatie', 'en':'NARCIS: The gateway to Dutch scientific information'},
link = {'nl':'http://www.narcis.nl/?Language=nl', 'en':'http://www.narcis.nl/?Language=en'},
maximumRecords = 20),
executeQueryHelix,
(RssItem(
nsMap=NAMESPACEMAP,
title = ('knaw_short', {'nl':'//short:metadata/short:titleInfo[not (@xml:lang)]/short:title/text()', 'en':'//short:metadata/short:titleInfo[@xml:lang="en"]/short:title/text()'}),
description = ('knaw_short', {'nl':'//short:abstract[not (@xml:lang)]/text()', 'en':'//short:abstract[@xml:lang="en"]/text()'}),
pubdate = ('knaw_short', '//short:dateIssued/short:parsed/text()'),
linkTemplate = 'http://www.narcis.nl/%(wcpcollection)s/RecordID/%(oai_identifier)s/Language/%(language)s',
wcpcollection = ('meta', '//*[local-name() = "collection"]/text()'),
oai_identifier = ('meta', '//meta:record/meta:id/text()'),
language = ('Dummy: Language is auto provided by the calling RSS component, but needs to be present to serve the linkTemplate.')
),
(StorageAdapter(),
(storage,)
)
)
)
)
)
)
)