class OaiData:
def __init__(self):
self.tokensDB = None
def __activate__(self, context):
if self.tokensDB is None:
self.tokensDB = TokensDatabase(context)
# Set up configuration
self.systemConfig = JsonSimpleConfig()
self.oaiConfig = None
self.getMetadataFormats()
self.velocityContext = context
self.services = context["Services"]
self.log = context["log"]
self.sessionState = context["sessionState"]
self.portalDir = context["portalDir"]
self.__result = None
# Check if the OAI request has an overriding portal ('set') to the URL
paramSet = self.vc("formData").get("set")
self.__portalName = context["page"].getPortal().getName()
illegalSet = False
if paramSet is not None:
portals = self.vc("page").getPortals().keySet()
if portals.contains(paramSet):
self.__portalName = paramSet
else:
illegalSet = True
self.__metadataPrefix = ""
self.__sessionExpiry = self.systemConfig.getInteger(None, ["portal", "oai-pmh", "sessionExpiry"])
# Check if there's a resumption token in the formData
self.__currentToken = None
resumptionToken = self.vc("formData").get("resumptionToken")
if resumptionToken is not None:
# This could still be be null
self.__currentToken = self.tokensDB.getToken(resumptionToken)
# Process/parse the request we've received for validity
self.vc("request").setAttribute("Content-Type", "text/xml")
self.__request = OaiPmhVerb(context, self.tokensDB, self.__currentToken)
if self.getError() is None and illegalSet:
self.__request.setError("badArgument", "Set '%s' is not valid!" % paramSet)
# If there are no errors... and the request requires some additional
# data (like a search result) do so now. Everything else can be
# handled in the templates.
if self.getError() is None and \
self.getVerb() in ["GetRecord", "ListIdentifiers", "ListRecords"]:
# Find the metadata prefix requested
self.__metadataPrefix = self.vc("formData").get("metadataPrefix")
if self.__metadataPrefix is None:
self.__metadataPrefix = self.__currentToken.getMetadataPrefix()
# Only list records if the metadata format is enabled in this view
if self.isInView(self.__metadataPrefix):
self.__search()
# Get from velocity context
def vc(self, index):
if self.velocityContext[index] is not None:
return self.velocityContext[index]
else:
self.log.error("ERROR: Requested context entry '" + index + "' doesn't exist")
return None
def isInView(self, format, view = None):
# Sanity check
if format is None or format == "":
return False
# Default to current poral
if view is None:
view = self.__portalName
# Make sure there is some config for this format
formatConfig = self.getMetadataFormats().get(format)
if formatConfig is None:
return False
# Is it visible everywhere?
allViews = formatConfig.getBoolean(False, ["enabledInAllViews"])
if allViews:
return True
# Check if it is visible in this view
else:
allowedViews = formatConfig.getStringList(["enabledViews"])
if view in allowedViews:
return True
# Rejection
return False
def getID(self, item):
identifier = item.getFirst("oai_identifier")
# Fallback to the default
if identifier is None or identifier == "":
return "oai:fascinator.usq.edu.au:" + item.getFirst("id")
# Use the indexed value
return identifier
def isDeleted(self, item):
return bool(item.getFirst("oai_deleted"))
def getSet(self, item):
set = item.getFirst("oai_set")
# Fallback to the portal name
if set is None or set == "":
return self.__portalName
# Use the required set
return set
def getVerb(self):
return self.getRequest().getVerb()
def getError(self):
return self.getRequest().getError()
def getResponseDate(self):
return time.strftime("%Y-%m-%dT%H:%M:%SZ")
def getRequest(self):
return self.__request
def getResult(self):
return self.__result
def getElement(self, elementName, values):
elementStr = ""
if values:
for value in values:
elementStr += "<%s>%s</%s>" % (elementName, value, elementName)
return elementStr
def __search(self):
self.__result = SolrResult(None)
portal = self.services.getPortalManager().get(self.__portalName)
recordsPerPage = portal.recordsPerPage
# Resolve our identifier
id = self.vc("formData").get("identifier")
query = "*:*"
if id is not None and id != "":
# A default TF2 OID
if id.startswith("oai:fascinator.usq.edu.au:"):
idString = id.replace("oai:fascinator.usq.edu.au:", "")
idString = self.__escapeQuery(idString)
query = "id:" + idString
# Or a custom OAI ID
else:
idString = self.__escapeQuery(id)
query = "oai_identifier:" + idString
req = SearchRequest(query)
req.setParam("facet", "true")
req.setParam("rows", str(recordsPerPage))
req.setParam("facet.field", portal.facetFieldList)
req.setParam("facet.limit", str(portal.facetCount))
req.setParam("sort", "f_dc_title asc")
portalQuery = portal.query
if portalQuery:
req.addParam("fq", portalQuery)
req.addParam("fq", "item_type:object")
# Date data... is supplied
fromDate = self.__request.getFromDate()
untilDate = self.__request.getUntilDate()
if fromDate is not None:
fromStr = fromDate.isoformat() + "Z"
self.log.debug("From Date: '{}'", fromStr)
if untilDate is not None:
untilStr = untilDate.isoformat() + "Z"
self.log.debug("Until Date: '{}'", untilStr)
queryStr = "last_modified:[%s TO %s]" % (fromStr, untilStr)
else:
queryStr = "last_modified:[%s TO *]" % (fromStr)
self.log.debug("Date query: '{}'", queryStr)
req.addParam("fq", queryStr)
else:
if untilDate is not None:
untilStr = untilDate.isoformat() + "Z"
self.log.debug("Until Date: '{}'", untilDate.isoformat())
queryStr = "last_modified:[* TO %s]" % (untilStr)
self.log.debug("Date query: '{}'", queryStr)
req.addParam("fq", queryStr)
# Check if there's resumption token exist in the formData
newToken = None
if self.__currentToken is not None:
start = int(self.__currentToken.getStart())
totalFound = int(self.__currentToken.getTotalFound())
nextTokenStart = start + recordsPerPage
if nextTokenStart < totalFound:
newToken = self.__currentToken
newToken.resetExpiry(self.__sessionExpiry)
newToken.setStart(nextTokenStart)
# or start a new resumption token
else:
start = 0
newToken = ResumptionToken(None, recordsPerPage, \
self.__metadataPrefix, self.__sessionExpiry)
req.setParam("start", str(start))
out = ByteArrayOutputStream()
self.services.indexer.search(req, out)
self.__result = SolrResult(ByteArrayInputStream(out.toByteArray()))
totalFound = self.__result.getNumFound()
if totalFound == 0:
newToken = None
# If an ID was requested, and not found, this is an error
if id is not None and id != "":
self.__request.setError("idDoesNotExist", "ID: '%s' not found" % id)
else:
self.__request.setError("noRecordsMatch", "No records match this request")
# We need to store this for NEW tokens
elif self.__currentToken is None:
# Assuming there are enough results to even keep the token
if newToken.getStart() < totalFound:
newToken.setTotalFound(totalFound)
else:
newToken = None
# Check if we need to remove the resumption token
else:
if self.__result.getResults().size() < recordsPerPage:
self.tokensDB.removeToken(self.__currentToken)
# Store/update the resumption token
if newToken is not None:
# Brand new token
if self.__currentToken is None:
self.tokensDB.storeToken(newToken)
# Or update an old token
else:
self.tokensDB.updateToken(newToken)
self.__currentToken = newToken
def getToken(self):
if self.isInView(self.__metadataPrefix):
return self.__currentToken
return None
def getMetadataFormats(self):
if self.oaiConfig is None:
self.oaiConfig = self.systemConfig.getJsonSimpleMap(["portal", "oai-pmh", "metadataFormats"])
return self.oaiConfig
def encodeXml(self, string):
return StringEscapeUtils.escapeXml(string);
def getPayload(self, oid, metadataFileName):
# First get the Object from storage
object = None
try:
object = self.services.getStorage().getObject(oid)
except StorageException, e:
return None
# Check whether the payload exists
try:
return object.getPayload(metadataFileName)
except StorageException, e:
return None
class HomeData:
def __init__(self):
pass
def __activate__(self, context):
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.__latest = None
self.__mine = None
self.__workflows = None
self.__result = None
self.__search()
# Get from velocity context
def vc(self, index):
if self.velocityContext[index] is not None:
return self.velocityContext[index]
else:
print "ERROR: Requested context entry '" + index + "' doesn't exist"
return None
def __search(self):
indexer = Services.getIndexer()
portalQuery = Services.getPortalManager().get(self.vc("portalId")).getQuery()
portalSearchQuery = Services.getPortalManager().get(self.vc("portalId")).getSearchQuery()
# Security prep work
current_user = self.vc("page").authentication.get_username()
security_roles = self.vc("page").authentication.get_roles_list()
security_filter = 'security_filter:("' + '" OR "'.join(security_roles) + '")'
security_exceptions = 'security_exception:"' + current_user + '"'
owner_query = 'owner:"' + current_user + '"'
security_query = "(" + security_filter + ") OR (" + security_exceptions + ") OR (" + owner_query + ")"
isAdmin = self.vc("page").authentication.is_admin()
req = SearchRequest("last_modified:[NOW-1MONTH TO *]")
req.setParam("fq", 'item_type:"object"')
if portalQuery:
req.addParam("fq", portalQuery)
if portalSearchQuery:
req.addParam("fq", portalSearchQuery)
req.setParam("rows", "10")
req.setParam("sort", "last_modified desc, f_dc_title asc");
if not isAdmin:
req.addParam("fq", security_query)
out = ByteArrayOutputStream()
indexer.search(req, out)
self.__latest = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest(owner_query)
req.setParam("fq", 'item_type:"object"')
if portalQuery:
req.addParam("fq", portalQuery)
if portalSearchQuery:
req.addParam("fq", portalSearchQuery)
req.setParam("rows", "10")
req.setParam("sort", "last_modified desc, f_dc_title asc");
if not isAdmin:
req.addParam("fq", security_query)
out = ByteArrayOutputStream()
indexer.search(req, out)
self.__mine = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest('workflow_security:"' + current_user + '"')
req.setParam("fq", 'item_type:"object"')
if portalQuery:
req.addParam("fq", portalQuery)
if portalSearchQuery:
req.addParam("fq", portalSearchQuery)
req.setParam("rows", "10")
req.setParam("sort", "last_modified desc, f_dc_title asc");
if not isAdmin:
req.addParam("fq", security_query)
out = ByteArrayOutputStream()
indexer.search(req, out)
self.__workflows = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest("*:*")
req.setParam("fq", 'item_type:"object"')
if portalQuery:
req.addParam("fq", portalQuery)
if portalSearchQuery:
req.addParam("fq", portalSearchQuery)
req.addParam("fq", "")
req.setParam("rows", "0")
if not isAdmin:
req.addParam("fq", security_query)
out = ByteArrayOutputStream()
indexer.search(req, out)
self.vc("sessionState").set("fq", 'item_type:"object"')
#sessionState.set("query", portalQuery.replace("\"", "'"))
self.__result = SolrResult(ByteArrayInputStream(out.toByteArray()))
def getLatest(self):
return self.__latest.getResults()
def getMine(self):
return self.__mine.getResults()
def getWorkflows(self):
return self.__workflows.getResults()
def getItemCount(self):
return self.__result.getNumFound()
def __search(self):
requireEscape = False
recordsPerPage = self.__portal.recordsPerPage
uri = URLDecoder.decode(self.request.getAttribute("RequestURI"))
query = None
pagePath = self.__portal.getName() + "/" + self.pageName
if query is None or query == "":
query = self.formData.get("query")
requireEscape = True
if query is None or query == "":
query = "*:*"
if query == "*:*":
self.__query = ""
else:
self.__query = query
if requireEscape:
query = self.__escapeQuery(query)
query = "%s:%s" % (self.__searchField, query)
self.sessionState.set("query", self.__query)
# find objects with annotations matching the query
if query != "*:*":
anotarQuery = self.__query
if requireEscape:
anotarQuery = self.__escapeQuery(anotarQuery)
annoReq = SearchRequest(anotarQuery)
annoReq.setParam("facet", "false")
annoReq.setParam("rows", str(99999))
annoReq.setParam("sort", "dateCreated asc")
annoReq.setParam("start", str(0))
anotarOut = ByteArrayOutputStream()
self.services.indexer.annotateSearch(annoReq, anotarOut)
resultForAnotar = SolrResult(ByteArrayInputStream(anotarOut.toByteArray()))
resultForAnotar = resultForAnotar.getResults()
ids = HashSet()
for annoDoc in resultForAnotar:
annotatesUri = annoDoc.get("annotatesUri")
ids.add(annotatesUri)
print "Found annotation for %s" % annotatesUri
# add annotation ids to query
query += ' OR id:("' + '" OR "'.join(ids) + '")'
portalSearchQuery = self.__portal.searchQuery
if portalSearchQuery == "":
portalSearchQuery = query
else:
if query != "*:*":
query += " AND " + portalSearchQuery
else:
query = portalSearchQuery
req = SearchRequest(query)
req.setParam("facet", "true")
req.setParam("rows", str(recordsPerPage))
req.setParam("facet.field", self.__portal.facetFieldList)
req.setParam("facet.sort", Boolean.toString(self.__portal.getFacetSort()))
req.setParam("facet.limit", str(self.__portal.facetCount))
req.setParam("sort", self.__sortBy)
# setup facets
if self.__useSessionNavigation:
action = self.formData.get("verb")
value = self.formData.get("value")
fq = self.sessionState.get("fq")
if fq is not None:
self.__pageNum = 1
req.setParam("fq", fq)
if action == "add_fq":
self.__pageNum = 1
req.addParam("fq", URLDecoder.decode(value, "UTF-8"))
elif action == "remove_fq":
self.__pageNum = 1
req.removeParam("fq", URLDecoder.decode(value, "UTF-8"))
elif action == "clear_fq":
self.__pageNum = 1
req.removeParam("fq")
elif action == "select-page":
self.__pageNum = int(value)
else:
navUri = uri[len(pagePath):]
self.__pageNum, fq, self.__fqParts = self.__parseUri(navUri)
savedfq = self.sessionState.get("savedfq")
limits = []
if savedfq:
limits.extend(savedfq)
if fq:
limits.extend(fq)
self.sessionState.set("savedfq", limits)
for q in fq:
req.addParam("fq", URLDecoder.decode(q, "UTF-8"))
portalQuery = self.__portal.query
if portalQuery:
req.addParam("fq", portalQuery)
req.addParam("fq", 'item_type:"object"')
if req.getParams("fq"):
self.__selected = ArrayList(req.getParams("fq"))
if self.__useSessionNavigation:
self.sessionState.set("fq", self.__selected)
self.sessionState.set("searchQuery", portalSearchQuery)
self.sessionState.set("pageNum", self.__pageNum)
# Make sure 'fq' has already been set in the session
if not self.page.authentication.is_admin():
current_user = self.page.authentication.get_username()
security_roles = self.page.authentication.get_roles_list()
security_filter = 'security_filter:("' + '" OR "'.join(security_roles) + '")'
security_exceptions = 'security_exception:"' + current_user + '"'
owner_query = 'owner:"' + current_user + '"'
security_query = "(" + security_filter + ") OR (" + security_exceptions + ") OR (" + owner_query + ")"
req.addParam("fq", security_query)
req.setParam("start", str((self.__pageNum - 1) * recordsPerPage))
print " * search.py:", req.toString(), self.__pageNum
out = ByteArrayOutputStream()
self.services.indexer.search(req, out)
self.__result = SolrResult(ByteArrayInputStream(out.toByteArray()))
if self.__result is not None:
self.__paging = Pagination(self.__pageNum,
int(self.__result.getNumFound()),
self.__portal.recordsPerPage)
