def setUp(self):
coreAConverter = QueryExpressionToLuceneQueryDict(
[('fieldA', 1.0)], luceneSettings=LuceneSettings())
coreBConverter = QueryExpressionToLuceneQueryDict(
[('fieldB', 1.0)], luceneSettings=LuceneSettings())
self.converter = AdapterToLuceneQuery(defaultCore='A',
coreConverters=dict(
A=coreAConverter,
B=coreBConverter))
self.observer = CallTrace('Query responder',
methods={'executeQuery': executeQueryMock})
self.dna = be((Observable(), (
self.converter,
(self.observer, ),
)))
def testComposedQuery(self):
self.response = JsonDict({
"total": 887,
"queryTime": 6,
"hits": [{"id": "record:1", "score": 0.1234}]
}).dumps()
cq = ComposedQuery('coreA')
q = QueryExpressionToLuceneQueryDict([], LuceneSettings()).convert(cqlToExpression("field=value"))
cq.setCoreQuery("coreA", q)
consume(self._multiLucene.executeComposedQuery(cq))
self.assertEqual(1, len(self.post))
self.assertEqual("/query/", self.post[0]['path'])
self.assertEqual({
"_sortKeys": [],
"resultsFrom": "coreA",
"_matches": {},
"_facets": {},
"_otherCoreFacetFilters": {},
"_rankQueries": {},
"_drilldownQueries": {},
"_unites": [],
"_queries": {"coreA": {"term": {"field": "field", "value": "value"}, "type": "TermQuery"}},
"cores": ["coreA"],
"_filterQueries": {}
}, loads(self.post[0]['data']))
def testAddTypeAndMissingValueToSortField(self):
self.response = JsonDict({
"total": 887,
"queryTime": 6,
"hits": [{"id": "record:1", "score": 0.1234}]
}).dumps()
cq = ComposedQuery('coreA')
q = QueryExpressionToLuceneQueryDict([], LuceneSettings()).convert(cqlToExpression("field=value"))
cq.setCoreQuery('coreB', q)
cq.sortKeys = [dict(sortBy='sortField', core='coreA', sortDescending=True)]
cq.addMatch(dict(core='coreA', uniqueKey='A'), dict(core='coreB', key='B'))
consume(self._multiLucene.executeComposedQuery(cq))
self.assertEqual({
"_sortKeys": [{'core': 'coreA', 'sortBy': 'sortField', 'sortDescending': True, 'type': 'String', 'missingValue': 'STRING_FIRST'}],
"resultsFrom": "coreA",
'_matches': {'coreA->coreB': [{'core': 'coreA', 'uniqueKey': 'A'}, {'core': 'coreB', 'key': 'B'}]},
"_facets": {},
"_otherCoreFacetFilters": {},
"_rankQueries": {},
"_drilldownQueries": {},
"_unites": [],
'_queries': {'coreB': {'term': {'field': 'field', 'value': 'value'}, 'type': 'TermQuery'}},
"cores": ["coreB", "coreA"],
"_filterQueries": {}
}, loads(self.post[0]['data']))
class LuceneQueryComposer(object):
def __init__(self, **kwargs):
self._queryExpressionToLuceneQuery = QueryExpressionToLuceneQueryDict(
**kwargs)
def compose(self, ast):
return self._queryExpressionToLuceneQuery.convert(cqlToExpression(ast))
def _prepareConverter(self):
unqualifiedFields = getattr(self, 'unqualifiedFields',
[("unqualified", 1.0)])
return QueryExpressionToLuceneQueryDict(
unqualifiedTermFields=unqualifiedFields,
luceneSettings=self._prepareLuceneSettings(),
ignoreStemmingForWords=getattr(self, '_ignoredStemmingForWords',
None))
def convert(self, expression=None, cql=None):
if expression is None:
expression = cqlToExpression(parseCql(cql))
unqualifiedFields = getattr(self, 'unqualifiedFields', [("unqualified", 1.0)])
settings = LuceneSettings()
if hasattr(self, '_analyzer'):
settings.analyzer = self._analyzer
if hasattr(self, 'fieldRegistry'):
settings.fieldRegistry = self.fieldRegistry
else:
settings.fieldRegistry = FieldRegistry()
settings.fieldRegistry.register("intField", fieldDefinition=INTFIELD)
settings.fieldRegistry.register("longField", fieldDefinition=LONGFIELD)
converter = QueryExpressionToLuceneQueryDict(
unqualifiedTermFields=unqualifiedFields,
luceneSettings=settings,
ignoreStemmingForWords=getattr(self, '_ignoredStemmingForWords', None)
)
return converter.convert(expression)
def testExecuteQuery(self):
self.response = JsonDict({
"total": 887,
"queryTime": 6,
"times": {"searchTime": 3},
"hits": [{
"id": "record:1", "score": 0.1234,
"duplicateCount": {"__key__": 2},
"duplicates": {"__grouping_key__": [{"id": 'record:1'}, {"id": 'record:2'}]}
}],
"drilldownData": [
{"fieldname": "facet", "path": [], "terms": [{"term": "term", "count": 1}]}
],
"suggestions": {
"valeu": ["value"]
}
}).dumps()
query = QueryExpressionToLuceneQueryDict([], LuceneSettings()).convert(cqlToExpression("field=value"))
response = retval(self._lucene.executeQuery(
luceneQuery=query, start=1, stop=5,
facets=[dict(maxTerms=10, fieldname='facet')],
sortKeys=[dict(sortBy='field', sortDescending=False)],
suggestionRequest=dict(suggests=['valeu'], count=2, field='field1'),
dedupField="__key__",
clustering=True,
storedFields=["field"]
))
self.assertEqual(1, len(self.post))
self.assertEqual('/lucene/query/', self.post[0]['path'])
self.assertEqual({
"start": 1, "stop": 5,
"storedFields": ["field"],
"query": {"term": {"field": "field", "value": "value"}, "type": "TermQuery"},
"facets": [{"fieldname": "facet", "maxTerms": 10}],
"sortKeys": [{"sortBy": "field", "sortDescending": False, "type": "String", 'missingValue': 'STRING_LAST'}],
"suggestionRequest": dict(suggests=['valeu'], count=2, field='field1'),
"dedupField": "__key__",
"dedupSortField": None,
"clustering": True,
}, loads(self.post[0]['data']))
self.assertEqual(887, response.total)
self.assertEqual(6, response.queryTime)
self.assertEqual({'searchTime': 3}, response.times)
self.assertEqual(1, len(response.hits))
self.assertEqual("record:1", response.hits[0].id)
self.assertEqual(0.1234, response.hits[0].score)
self.assertEqual(dict(__key__=2), response.hits[0].duplicateCount)
self.assertEqual([
{"fieldname": "facet", "path": [], "terms": [{"term": "term", "count": 1}]}
], response.drilldownData)
self.assertEqual({'valeu': ['value']}, response.suggestions)
def testLuceneServerHostPortDynamic(self):
multiLucene = MultiLucene(defaultCore='core1')
def httprequest1_1Mock(**kwargs):
raise StopIteration(parseResponse(HTTP_RESPONSE))
yield
observer = CallTrace(
'observer',
returnValues=dict(luceneServer=('example.org', 1234)),
methods=dict(httprequest1_1=httprequest1_1Mock))
multiLucene.addObserver(observer)
query = QueryExpressionToLuceneQueryDict([], LuceneSettings()).convert(cqlToExpression("field=value"))
response = retval(multiLucene.executeComposedQuery(ComposedQuery('core1', query)))
self.assertEquals(887, response.total)
self.assertEquals(['luceneServer', 'httprequest1_1'], observer.calledMethodNames())
def testLuceneServerHostPortDynamic(self):
lucene = Lucene(name='lucene', settings=LuceneSettings(), readonly=True)
def httprequest1_1Mock(**kwargs):
raise StopIteration(parseResponse(HTTP_RESPONSE))
yield
observer = CallTrace(
'observer',
returnValues=dict(luceneServer=('example.org', 1234)),
methods=dict(httprequest1_1=httprequest1_1Mock))
lucene.addObserver(observer)
query = QueryExpressionToLuceneQueryDict([], LuceneSettings()).convert(cqlToExpression("field=value"))
response = retval(lucene.executeQuery(
luceneQuery=query, start=1, stop=5,
))
self.assertEquals(887, response.total)
self.assertEquals(['luceneServer', 'httprequest1_1'], observer.calledMethodNames())
def testDeleteByQuery(self):
query = QueryExpressionToLuceneQueryDict([], LuceneSettings()).convert(cqlToExpression("field=value"))
consume(self._lucene.delete(luceneQuery=query))
self.assertEqual(1, len(self.post))
self.assertEqual('/lucene/delete/', self.post[0]['path'])
self.assertEqual('{"query": {"term": {"field": "field", "value": "value"}, "type": "TermQuery"}}', self.post[0]['data'])
def __init__(self, **kwargs):
self._queryExpressionToLuceneQuery = QueryExpressionToLuceneQueryDict(
**kwargs)
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 __init__(self, **kwargs):
self._queryExpressionToLuceneQuery = QueryExpressionToLuceneQueryDict(**kwargs)
class LuceneQueryComposer(object):
def __init__(self, **kwargs):
self._queryExpressionToLuceneQuery = QueryExpressionToLuceneQueryDict(**kwargs)
def compose(self, ast):
return self._queryExpressionToLuceneQuery.convert(cqlToExpression(ast))
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,)
)
)
)
)
)
)
)
