def handleQuery(self, query, fieldName, formatStr):
out = ByteArrayOutputStream()
req = SearchRequest(query)
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", "1000")
self.indexer.search(req, out)
res = SolrResult(ByteArrayInputStream(out.toByteArray()))
hits = HashSet()
if (res.getNumFound() > 0):
results = res.getResults()
for searchRes in results:
searchResList = searchRes.getList(fieldName)
if (searchResList.isEmpty()==False):
for hit in searchResList:
if self.term is not None:
if hit.find(self.term) != -1:
hits.add(hit)
else:
hits.add(hit)
self.writer.print("[")
hitnum = 0
for hit in hits:
if (hitnum > 0):
self.writer.print(","+formatStr % {"hit":hit})
else:
self.writer.print(formatStr % {"hit":hit})
hitnum += 1
self.writer.print("]")
else:
self.writer.println("[\"\"]")
self.writer.close()
def export(self, exportType):
exportQuery = "%s:%s" % (self.facetField, self.facetFieldValue)
outputType = "text/%s; charset=UTF-8" % type
responseHeader = "attachment; filename=%s.%s" % (self.facetFieldValue, exportType)
try:
out = ByteArrayOutputStream()
recnumreq = SearchRequest(exportQuery)
recnumreq.setParam("fl","create_timestamp")
recnumreq.setParam("rows", "0")
self.indexer.search(recnumreq, out)
recnumres = SolrResult(ByteArrayInputStream(out.toByteArray()))
self.__rowsFoundSolr = "%s" % recnumres.getNumFound()
except:
self.errorMsg = "Export query failure. The issue has been logged (%s - %s)." % (sys.exc_info()[0], sys.exc_info()[1])
self.log.error("Export query threw an exception (package type was %s): %s - %s" % (self.facetFieldValue, sys.exc_info()[0], sys.exc_info()[1]))
return
out = ByteArrayOutputStream()
req = SearchRequest(exportQuery)
req.setParam("wt", exportType)
req.setParam("rows", self.__rowsFoundSolr)
self.indexer.search(req, out)
self.response.setHeader("Content-Disposition", responseHeader)
writer = self.response.getPrintWriter(outputType)
writer.println(out.toString("UTF-8"))
writer.close()
def handleWorkflowStep(self):
out = ByteArrayOutputStream()
req = SearchRequest("workflow_step_label:[* TO *]" )
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", "1000")
self.indexer.search(req, out)
res = SolrResult(ByteArrayInputStream(out.toByteArray()))
hits = HashSet()
if (res.getNumFound() > 0):
recordTypeResults = res.getResults()
for recordTypeResult in recordTypeResults:
recordTypeList = recordTypeResult.getList("workflow_step_label")
if (recordTypeList.isEmpty()==False):
for hit in recordTypeList:
hits.add(hit)
self.writer.println("[")
hitnum = 0
for hit in hits:
if (hitnum > 0):
self.writer.println(",{\"value\": \"%s\",\n\"label\": \"%s\"}" % (hit,hit))
else:
self.writer.println("{\"value\": \"%s\",\n\"label\": \"%s\"}" % (hit,hit))
hitnum += 1
self.writer.println("]")
else:
self.writer.println("[\"\"]")
self.writer.close()
def handleQuery(self, query, fieldName, formatStr):
out = ByteArrayOutputStream()
req = SearchRequest(query)
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", "1000")
self.indexer.search(req, out)
res = SolrResult(ByteArrayInputStream(out.toByteArray()))
hits = HashSet()
if (res.getNumFound() > 0):
results = res.getResults()
for searchRes in results:
searchResList = searchRes.getList(fieldName)
if (searchResList.isEmpty() == False):
for hit in searchResList:
if self.term is not None:
if hit.find(self.term) != -1:
hits.add(hit)
else:
hits.add(hit)
self.writer.print("[")
hitnum = 0
for hit in hits:
if (hitnum > 0):
self.writer.print("," + formatStr % {"hit": hit})
else:
self.writer.print(formatStr % {"hit": hit})
hitnum += 1
self.writer.print("]")
else:
self.writer.println("[\"\"]")
self.writer.close()
def handleGrantNumber(self):
out = ByteArrayOutputStream()
req = SearchRequest("grant_numbers:%s*" % self.term)
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", "1000")
self.indexer.search(req, out)
res = SolrResult(ByteArrayInputStream(out.toByteArray()))
hits = HashSet()
if (res.getNumFound() > 0):
creatorResults = res.getResults()
for creatorRes in creatorResults:
creatorList = creatorRes.getList("grant_numbers")
if (creatorList.isEmpty()==False):
for hit in creatorList:
hits.add(hit)
self.writer.print("[")
hitnum = 0
for hit in hits:
if (hitnum > 0):
self.writer.print(",\"%s\"" % hit)
else:
self.writer.print("\"%s\"" % hit)
hitnum += 1
self.writer.print("]")
else:
self.writer.println("[\"\"]")
self.writer.close()
def _searchSets(self, startPage=1):
req = SearchRequest(self.getQuery())
req.setParam("fq", 'item_type:"object"')
req.setParam("rows", str(self.getRecordsPerPage()))
req.setParam("start", str((startPage - 1) * self.getRecordsPerPage()))
req.addParam("fq", self.getFilterQuery())
req.setParam("fl", self.getReturnFields())
req.setParam("sort", "date_object_modified desc, f_dc_title asc")
if not self.isAdmin():
req.addParam("fq", self.getSecurityQuery())
out = ByteArrayOutputStream()
self.indexer.search(req, out)
result = SolrResult(ByteArrayInputStream(out.toByteArray()))
self._setPaging(result.getNumFound())
result.getJsonObject().put("lastPage", str(self.paging.getLastPage()))
result.getJsonObject().put("curPage", str(startPage))
return result
def _searchSets(self, startPage=1):
req = SearchRequest(self.getQuery())
req.setParam("fq", 'item_type:"object"')
req.setParam("rows", str(self.getRecordsPerPage()))
req.setParam("start", str((startPage - 1) * self.getRecordsPerPage()))
req.addParam("fq", self.getFilterQuery())
req.setParam("fl", self.getReturnFields())
req.setParam("sort", "date_object_modified desc, f_dc_title asc")
if not self.isAdmin():
req.addParam("fq", self.getSecurityQuery())
out = ByteArrayOutputStream()
self.indexer.search(req, out)
result = SolrResult(ByteArrayInputStream(out.toByteArray()))
self._setPaging(result.getNumFound())
result.getJsonObject().put("lastPage", str(self.paging.getLastPage()))
result.getJsonObject().put("curPage", str(startPage))
return result
class ReportsData:
def __init__(self):
pass
def __activate__(self, context):
#import pydevd;pydevd.settrace()
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.services = self.vc("Services")
self.log = context["log"]
self.__harvestList = 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 = self.services.getIndexer()
# Security prep work
isAdmin = self.vc("page").authentication.is_admin()
if not isAdmin:
print "ERROR: User is not an admin '"
return None
req = SearchRequest('eventType:harvestStart')
req.setParam("rows", "100")
out = ByteArrayOutputStream()
indexer.searchByIndex(req, out, "eventLog")
self.__harvestList = SolrResult(ByteArrayInputStream(out.toByteArray()))
def getHarvestlist(self):
return self.__harvestList.getResults()
def getItemCount(self):
return self.__harvestList.getNumFound()
class ReportsData:
def __init__(self):
pass
def __activate__(self, context):
#import pydevd;pydevd.settrace()
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.services = self.vc("Services")
self.log = context["log"]
self.__harvestList = 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 = self.services.getIndexer()
# Security prep work
isAdmin = self.vc("page").authentication.is_admin()
if not isAdmin:
print "ERROR: User is not an admin '"
return None
req = SearchRequest('eventType:harvestStart')
req.setParam("rows", "100")
out = ByteArrayOutputStream()
indexer.searchByIndex(req, out, "eventLog")
self.__harvestList = SolrResult(ByteArrayInputStream(
out.toByteArray()))
def getHarvestlist(self):
return self.__harvestList.getResults()
def getItemCount(self):
return self.__harvestList.getNumFound()
class HomeData:
def __init__(self):
pass
def __activate__(self, context):
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.__latest = None
self.__steps = None
self.__alerts = None
self.__result = None
self.__stages = None
self.__search()
# Get from velocity context
def vc(self, index):
if self.velocityContext[index] is not None:
return self.velocityContext[index]
else:
self.velocityContext["log"].error(
"ERROR: Requested context entry '{}' doesn't exist", index)
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("*:*")
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")
req.setParam("facet", "true")
req.setParam("facet.field", "workflow_step")
if not isAdmin:
req.addParam("fq", security_query)
out = ByteArrayOutputStream()
indexer.search(req, out)
steps = SolrResult(ByteArrayInputStream(out.toByteArray()))
self.__steps = steps.getFacets().get("workflow_step")
wfConfig = JsonSimple(
FascinatorHome.getPathFile("harvest/workflows/dataset.json"))
jsonStageList = wfConfig.getJsonSimpleList(["stages"])
stages = []
for jsonStage in jsonStageList:
wfStage = WorkflowStage(jsonStage, self.__steps)
stages.append(wfStage)
self.__stages = stages
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", "25")
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.__result = SolrResult(ByteArrayInputStream(out.toByteArray()))
req.addParam("fq", "workflow_step:%s" % stages[0].getName())
out = ByteArrayOutputStream()
indexer.search(req, out)
self.__alerts = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest(
"last_modified:[NOW-1MONTH TO *] AND workflow_step:live")
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()))
self.vc("sessionState").set("fq", 'item_type:"object"')
def getLatest(self):
return self.__latest.getResults()
def getAlerts(self):
return self.__alerts.getResults()
def getItemCount(self):
return self.__result.getNumFound()
def getStages(self):
return self.__stages
class SearchData:
def __activate__(self, context):
self.services = context["Services"]
self.page = context["page"]
self.formData = context["formData"]
self.portalId = context["portalId"]
self.sessionState = context["sessionState"]
self.request = context["request"]
self.pageName = context["pageName"]
self.log = context["log"]
self.__portal = context["page"].getPortal()
self.__useSessionNavigation = self.__portal.getBoolean(
True, ["portal", "use-session-navigation"])
self.__result = None
if self.__useSessionNavigation:
self.__pageNum = self.sessionState.get("pageNum", 1)
else:
self.__pageNum = 1
self.__selected = ArrayList()
self.__fqParts = []
self.__searchField = self.formData.get("searchField", "full_text")
self.__sortField = self.formData.get("sort-field")
self.__sortOrder = self.formData.get("sort-order")
if not (self.__sortField or self.__sortOrder):
# use form data not specified, check session
sortField = self.__portal.sortFieldDefault or "score"
sortOrder = self.__portal.sortFieldDefaultOrder or "desc"
#print "f:%s,o:%s" % (sortField, sortOrder)
self.__sortField = self.sessionState.get("sortField", sortField)
self.__sortOrder = self.sessionState.get("sortOrder", sortOrder)
self.sessionState.set("sortField", self.__sortField)
self.sessionState.set("sortOrder", self.__sortOrder)
self.__sortBy = "%s %s" % (self.__sortField, self.__sortOrder)
# reset the query and facet selections when changing views
lastPortalId = self.sessionState.get("lastPortalId")
if lastPortalId != self.portalId:
self.log.info(
"Portal changed. Resetting saved search values in session.")
self.sessionState.remove("fq")
self.sessionState.remove("pageNum")
self.sessionState.remove("sortField")
self.sessionState.remove("sortOrder")
self.__pageNum = 1
self.sessionState.set("lastPortalId", self.portalId)
self.__search()
def usingSessionNavigation(self):
return self.__restful
def getPortalName(self):
return self.__portal.getDescription()
def getSearchField(self):
return self.__searchField
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.getFirst("annotatesUri")
ids.add(annotatesUri)
self.log.debug("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)
## uncomment to ensure guest users not logged in cannot see alerts in browse page
# self.filterOutWorkflowStepForUnAuth(req, "inbox")
req.setParam("start", str((self.__pageNum - 1) * recordsPerPage))
self.log.debug(" * search.py: %s, page: %s" %
(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,
self.__result.getNumFound(),
self.__portal.recordsPerPage)
def filterOutWorkflowStepForUnAuth(self, req, workflow_step):
if self.page.authentication.get_username(
) == 'guest' and not self.page.authentication.is_logged_in():
step_filter = [workflow_step]
no_alerts_for_non_auth = ' - workflow_step:("' + ''.join(
step_filter) + '")'
req.addParam("fq", no_alerts_for_non_auth)
def __escapeQuery(self, q):
temp = ""
chars = "+-&|!(){}[]^\"~*?:\\"
for c in q:
if c in chars:
temp += "\%s" % c
else:
temp += c
return temp
# eq = q
# # escape all solr/lucene special chars
# # from http://lucene.apache.org/java/2_4_0/queryparsersyntax.html#Escaping%20Special%20Characters
# for c in "+-&|!(){}[]^\"~*?:\\":
# eq = eq.replace(c, "\\%s" % c)
# ## Escape UTF8
# try:
# return URLEncoder.encode(eq, "UTF-8")
# except UnsupportedEncodingException, e:
# print "Error during UTF8 escape! ", repr(eq)
# return eq
def getQueryTime(self):
return int(self.__result.getQueryTime()) / 1000.0
def getPaging(self):
return self.__paging
def getResult(self):
return self.__result
def getFacetField(self, key):
return self.__portal.facetFields.get(key)
def getFacetName(self, key):
return self.__portal.facetFields.get(key).getString(None, ["label"])
def getFacetCounts(self, key):
if self.__useSessionNavigation:
facetData = self.__result.getFacets()
if facetData is None:
return LinkedHashMap()
if not facetData.containsKey(key):
return LinkedHashMap()
return facetData.get(key).values()
else:
return LinkedHashMap()
# TODO : What were these doing? Hiding file path facets unless some facets are selected?
#if name.find("/") == -1 or self.hasSelectedFacets():
# values.put(name, count)
def getFacetDisplay(self):
return self.__portal.facetDisplay
def hasSelectedFacets(self):
return (self.__selected is not None and len(self.__selected) > 1) and \
not (self.__portal.query in self.__selected and len(self.__selected) == 2)
def getSelectedFacets(self):
return self.__selected
def isPortalQueryFacet(self, fq):
return fq == self.__portal.query
def isSelected(self, fq):
return fq in self.__selected
def getSelectedFacetIds(self):
## Returns a native Python array
## Python has indicator of u' for unicode string which causes JS complain.
## Because MD5 hash is made from letters we can encode them in ascii
return [
DigestUtils.md5Hex(fq).encode("ascii", "ignore")
for fq in self.__selected
]
def getFileName(self, path):
return os.path.splitext(os.path.basename(path))[0]
def getFacetQuery(self, name, value):
return '%s:"%s"' % (name, value)
# Packaging support
def getActiveManifestTitle(self):
return self.__getActiveManifest().getTitle()
def getActiveManifestId(self):
return self.sessionState.get("package/active/id")
def getSelectedItemsCount(self):
return self.__getActiveManifest().size()
def isSelectedForPackage(self, oid):
result = self.__getActiveManifest().getNode("node-%s" % oid)
return (result is not None)
def getManifestItemTitle(self, oid, defaultValue):
result = self.__getActiveManifest().getNode("node-%s" % oid)
if result is None:
return defaultValue
return result.getTitle()
def __getActiveManifest(self):
activeManifest = self.sessionState.get("package/active")
if not activeManifest:
activeManifest = Manifest(None)
activeManifest.setTitle("New package")
activeManifest.setViewId(self.__portal.getName())
self.sessionState.set("package/active", activeManifest)
return activeManifest
def isSelectableForPackage(self, oid):
return oid != self.getActiveManifestId()
def getSortFields(self):
return self.__portal.sortFields
def getSortField(self):
return self.__sortField
def getSortOrder(self):
return self.__sortOrder
# RESTful style URL support methods
def getPageQuery(self, page):
prefix = ""
if self.__fqParts:
prefix = "/" + "/".join(self.__fqParts)
suffix = ""
if page > 1:
suffix = "/page/%s" % page
return prefix + suffix
def getFacetQueryUri(self, name, value):
return "%s/%s" % (name, value)
def getFacetValue(self, facetValue):
return facetValue.split("/")[-1]
def getLimitQueryWith(self, fq):
limits = ArrayList(self.__fqParts)
limits.add("category/" + fq)
return "/".join(limits)
def getFacetIndent(self, facetValue):
return len(facetValue.split("/"))
def getLimitQueryWithout(self, fq):
limits = ArrayList(self.__fqParts)
limits.remove("category/" + fq)
if limits.isEmpty():
return ""
return "/".join(limits)
def __parseUri(self, uri):
page = 1
fq = []
fqParts = []
if uri != "":
parts = uri.split("/")
partType = None
facetKey = None
facetValues = None
for part in parts:
if partType == "page":
facetKey = None
page = int(part)
elif partType == "category":
partType = "category-value"
facetValues = None
facetKey = part
elif partType == "category-value":
if facetValues is None:
facetValues = []
if part in ["page", "category"]:
partType = part
facetQuery = '%s:"%s"' % (facetKey,
"/".join(facetValues))
fq.append(facetQuery)
fqParts.append("category/%s/%s" %
(facetKey, "/".join(facetValues)))
facetKey = None
facetValues = None
else:
facetValues.append(URLDecoder.decode(part))
else:
partType = part
if partType == "category-value":
facetQuery = '%s:"%s"' % (facetKey, "/".join(facetValues))
fq.append(facetQuery)
fqParts.append("category/%s/%s" %
(facetKey, "/".join(facetValues)))
return page, fq, fqParts
class HomeData:
def __init__(self):
pass
def __activate__(self, context):
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.services = self.vc("Services")
self.__latest = None
self.__mine = None
self.__workflows = None
self.__result = None
self.__steps = None
self.__selfservicesStages = None
self.__search()
# Get from velocity context Mint version
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 = self.services.getIndexer()
portalQuery = self.services.getPortalManager().get(self.vc("portalId")).getQuery()
portalSearchQuery = self.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("\"", "'"))
# Load in the services UI workflow
selfSubmitWfConfig = JsonSimple(FascinatorHome.getPathFile("harvest/workflows/servicesUI.json"))
selfSubmitJsonStageList = selfSubmitWfConfig.getJsonSimpleList(["stages"])
servicesStages = []
for jsonStage in selfSubmitJsonStageList:
wfStage = WorkflowStage(jsonStage, self.__steps)
servicesStages.append(wfStage)
self.__selfservicesStages = servicesStages
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 getServicesStages(self):
return self.__servicesStages
class HomeData:
def __init__(self):
pass
def __activate__(self, context):
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.__latest = None
self.__steps = None
self.__alerts = None
self.__result = None
self.__stages = None
self.__embargoes = None
self.__search()
# Get from velocity context
def vc(self, index):
if self.velocityContext[index] is not None:
return self.velocityContext[index]
else:
self.velocityContext["log"].error("ERROR: Requested context entry '{}' doesn't exist", index)
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("*:*")
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")
req.setParam("facet", "true")
req.setParam("facet.field", "workflow_step")
if not isAdmin:
req.addParam("fq", security_query)
out = ByteArrayOutputStream()
indexer.search(req, out)
steps = SolrResult(ByteArrayInputStream(out.toByteArray()))
self.__steps = steps.getFacets().get("workflow_step")
wfConfig = JsonSimple(FascinatorHome.getPathFile("harvest/workflows/dataset.json"))
jsonStageList = wfConfig.getJsonSimpleList(["stages"])
stages = []
for jsonStage in jsonStageList:
wfStage = WorkflowStage(jsonStage, self.__steps)
stages.append(wfStage)
self.__stages = stages
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", "25")
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.__result = SolrResult(ByteArrayInputStream(out.toByteArray()))
req.addParam("fq", "workflow_step:%s" % stages[0].getName())
out = ByteArrayOutputStream()
indexer.search(req, out)
self.__alerts = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest("last_modified:[NOW-1MONTH TO *] AND workflow_step:live")
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()))
self._searchEmbargoes()
self.vc("sessionState").set("fq", 'item_type:"object"')
def getLatest(self):
return self.__latest.getResults()
def getAlerts(self):
return self.__alerts.getResults()
def getItemCount(self):
return self.__result.getNumFound()
def getStages(self):
return self.__stages
def getEmbargoes(self):
return self.__embargoes.getResults()
def _searchEmbargoes(self):
req = SearchRequest("item_type:object")
req.setParam("fq", 'redbox\:embargo.redbox\:isEmbargoed:on')
req.addParam("fq", 'workflow_step:final-review')
req.addParam("fq", "")
req.setParam("fl","id,date_embargoed,dc_title")
req.setParam("rows", "25")
req.setParam("sort", "date_embargoed asc, dc_title asc");
out = ByteArrayOutputStream()
indexer = Services.getIndexer()
indexer.search(req, out)
self.__embargoes = SolrResult(ByteArrayInputStream(out.toByteArray()))
self.velocityContext["log"].info("searchEmbargoes call ended" + str(self.__embargoes))
def formatDate(self, date):
dfSource = SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss")
dfTarget = SimpleDateFormat("dd/MM/yyyy")
return dfTarget.format(dfSource.parse(date))
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
self.lastPage = False
# 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:
# Split out the start component from the actual resumption token
(resumptionTokenPart, start) = resumptionToken.strip().split(":")
# This could still be be null
self.__currentToken = self.tokensDB.getToken(resumptionTokenPart)
# Code to handle null token is handled later on
if self.__currentToken is not None:
self.__currentToken.setStart(start)
# 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 (start + recordsPerPage) >= totalFound:
self.tokensDB.removeToken(self.__currentToken)
self.lastPage = True
# 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) and not self.lastPage:
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 SearchData:
def __activate__(self, context):
self.services = context["Services"]
self.page = context["page"]
self.formData = context["formData"]
self.portalId = context["portalId"]
self.sessionState = context["sessionState"]
self.request = context["request"]
self.pageName = context["pageName"]
self.log = context["log"]
self.__portal = context["page"].getPortal()
self.__useSessionNavigation = self.__portal.getBoolean(True, ["portal", "use-session-navigation"])
self.__result = None
if self.__useSessionNavigation:
self.__pageNum = self.sessionState.get("pageNum", 1)
else:
self.__pageNum = 1
self.__selected = ArrayList()
self.__fqParts = []
self.__searchField = self.formData.get("searchField", "full_text")
self.__sortField = self.formData.get("sort-field")
self.__sortOrder = self.formData.get("sort-order")
if not (self.__sortField or self.__sortOrder):
# use form data not specified, check session
sortField = self.__portal.sortFieldDefault or "score"
sortOrder = self.__portal.sortFieldDefaultOrder or "desc"
#print "f:%s,o:%s" % (sortField, sortOrder)
self.__sortField = self.sessionState.get("sortField", sortField)
self.__sortOrder = self.sessionState.get("sortOrder", sortOrder)
self.sessionState.set("sortField", self.__sortField)
self.sessionState.set("sortOrder", self.__sortOrder)
self.__sortBy = "%s %s" % (self.__sortField, self.__sortOrder)
# reset the query and facet selections when changing views
lastPortalId = self.sessionState.get("lastPortalId")
if lastPortalId != self.portalId:
self.sessionState.remove("fq")
self.sessionState.remove("pageNum")
self.sessionState.remove("sortField")
self.sessionState.remove("sortOrder")
self.__pageNum = 1
self.sessionState.set("lastPortalId", self.portalId)
self.__search()
def usingSessionNavigation(self):
return self.__restful
def getPortalName(self):
return self.__portal.getDescription()
def getSearchField(self):
return self.__searchField
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.getFirst("annotatesUri")
ids.add(annotatesUri)
self.log.debug("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,
self.__result.getNumFound(),
self.__portal.recordsPerPage)
def __escapeQuery(self, q):
temp = ""
chars = "+-&|!(){}[]^\"~*?:\\"
for c in q:
if c in chars:
temp += "\%s" % c
else:
temp += c
return temp
# eq = q
# # escape all solr/lucene special chars
# # from http://lucene.apache.org/java/2_4_0/queryparsersyntax.html#Escaping%20Special%20Characters
# for c in "+-&|!(){}[]^\"~*?:\\":
# eq = eq.replace(c, "\\%s" % c)
# ## Escape UTF8
# try:
# return URLEncoder.encode(eq, "UTF-8")
# except UnsupportedEncodingException, e:
# print "Error during UTF8 escape! ", repr(eq)
# return eq
def getQueryTime(self):
return int(self.__result.getQueryTime()) / 1000.0;
def getPaging(self):
return self.__paging
def getResult(self):
return self.__result
def getFacetField(self, key):
return self.__portal.facetFields.get(key)
def getFacetName(self, key):
return self.__portal.facetFields.get(key).getString(None, ["label"])
def getFacetCounts(self, key):
if self.__useSessionNavigation:
facetData = self.__result.getFacets()
if facetData is None:
return LinkedHashMap()
if not facetData.containsKey(key):
return LinkedHashMap()
return facetData.get(key).values()
else:
return LinkedHashMap()
# TODO : What were these doing? Hiding file path facets unless some facets are selected?
#if name.find("/") == -1 or self.hasSelectedFacets():
# values.put(name, count)
def getFacetDisplay(self):
return self.__portal.facetDisplay
def hasSelectedFacets(self):
return (self.__selected is not None and len(self.__selected) > 1) and \
not (self.__portal.query in self.__selected and len(self.__selected) == 2)
def getSelectedFacets(self):
return self.__selected
def isPortalQueryFacet(self, fq):
return fq == self.__portal.query
def isSelected(self, fq):
return fq in self.__selected
def getSelectedFacetIds(self):
return [md5.new(fq).hexdigest() for fq in self.__selected]
def getFileName(self, path):
return os.path.splitext(os.path.basename(path))[0]
def getFacetQuery(self, name, value):
return '%s:"%s"' % (name, value)
# Packaging support
def getActiveManifestTitle(self):
return self.__getActiveManifest().getTitle()
def getActiveManifestId(self):
return self.sessionState.get("package/active/id")
def getSelectedItemsCount(self):
return self.__getActiveManifest().size()
def isSelectedForPackage(self, oid):
result = self.__getActiveManifest().getNode("node-%s" % oid)
return (result is not None)
def getManifestItemTitle(self, oid, defaultValue):
result = self.__getActiveManifest().getNode("node-%s" % oid)
if result is None:
return defaultValue
return result.getTitle()
def __getActiveManifest(self):
activeManifest = self.sessionState.get("package/active")
if not activeManifest:
activeManifest = Manifest(None)
activeManifest.setTitle("New package")
activeManifest.setViewId(self.__portal.getName())
self.sessionState.set("package/active", activeManifest)
return activeManifest
def isSelectableForPackage(self, oid):
return oid != self.getActiveManifestId()
def getSortFields(self):
return self.__portal.sortFields
def getSortField(self):
return self.__sortField
def getSortOrder(self):
return self.__sortOrder
# RESTful style URL support methods
def getPageQuery(self, page):
prefix = ""
if self.__fqParts:
prefix = "/" + "/".join(self.__fqParts)
suffix = ""
if page > 1:
suffix = "/page/%s" % page
return prefix + suffix
def getFacetQueryUri(self, name, value):
return "%s/%s" % (name, value)
def getFacetValue(self, facetValue):
return facetValue.split("/")[-1]
def getLimitQueryWith(self, fq):
limits = ArrayList(self.__fqParts)
limits.add("category/" + fq)
return "/".join(limits)
def getFacetIndent(self, facetValue):
return len(facetValue.split("/"))
def getLimitQueryWithout(self, fq):
limits = ArrayList(self.__fqParts)
limits.remove("category/" + fq)
if limits.isEmpty():
return ""
return "/".join(limits)
def __parseUri(self, uri):
page = 1
fq = []
fqParts = []
if uri != "":
parts = uri.split("/")
partType = None
facetKey = None
facetValues = None
for part in parts:
if partType == "page":
facetKey = None
page = int(part)
elif partType == "category":
partType = "category-value"
facetValues = None
facetKey = part
elif partType == "category-value":
if facetValues is None:
facetValues = []
if part in ["page", "category"]:
partType = part
facetQuery = '%s:"%s"' % (facetKey, "/".join(facetValues))
fq.append(facetQuery)
fqParts.append("category/%s/%s" % (facetKey, "/".join(facetValues)))
facetKey = None
facetValues = None
else:
facetValues.append(URLDecoder.decode(part))
else:
partType = part
if partType == "category-value":
facetQuery = '%s:"%s"' % (facetKey, "/".join(facetValues))
fq.append(facetQuery)
fqParts.append("category/%s/%s" % (facetKey, "/".join(facetValues)))
return page, fq, fqParts
def __reportSearch(self):
self.reportId = self.request.getParameter("id")
self.format = self.request.getParameter("format")
self.report = self.reportManager.getReports().get(self.reportId)
self.reportQuery = self.report.getQueryAsString()
self.log.debug("Report query: " + self.reportQuery)
#Get a total number of records
try:
out = ByteArrayOutputStream()
recnumreq = SearchRequest(self.reportQuery)
recnumreq.setParam("rows", "0")
self.indexer.search(recnumreq, out)
recnumres = SolrResult(ByteArrayInputStream(out.toByteArray()))
self.__rowsFoundSolr = "%s" % recnumres.getNumFound()
except:
self.errorMsg = "Query failure. The issue has been logged (%s - %s)." % (sys.exc_info()[0], sys.exc_info()[1])
self.log.error("Reporting threw an exception (report was %s): %s - %s" % (self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1]))
return
#Setup the main query
req = SearchRequest(self.reportQuery)
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", self.__rowsFoundSolr)
try:
#Now do the master search
out = ByteArrayOutputStream()
self.indexer.search(req, out)
self.__reportResult = SolrResult(ByteArrayInputStream(out.toByteArray()))
self.__checkResults()
except:
self.errorMsg = "Query failure. The issue has been logged (%s - %s)." % (sys.exc_info()[0], sys.exc_info()[1])
self.log.error("Reporting threw an exception (report was %s): %s - %s" % (self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1]))
return
#At this point the display template has enough to go with.
#We just need to handle the CSV now
if (self.format == "csv"):
#Setup the main query - we need to requery to make sure we return
#only the required fields. We'll use the specific IDs that met the
#__checkResults check
req = SearchRequest(self.reportQuery)
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", self.__rowsFoundSolr)
req.setParam("csv.mv.separator",";")
#we need to get a list of the matching IDs from Solr
#this doesn't work for long queries so it's abandoned
#but left here commented to make sure we don't try it again
#idQry = ""
#for item in self.getProcessedResultsList():
# idQry += item.get("id") + " OR "
#req.setParam("fq", 'id:(%s)' % idQry[:len(idQry)-4])
#Create a list of IDs for reference when preparing the CSV
idQryList = []
for item in self.getProcessedResultsList():
idQryList.append(item.get("id"))
#Setup SOLR query with the required fields
self.fields = self.systemConfig.getArray("redbox-reports","csv-output-fields")
#We must have an ID field and it must be the first field
fieldString = "id,"
if self.fields is not None:
for field in self.fields:
fieldString = fieldString+ field.get("field-name")+","
fieldString = fieldString[:-1]
req.setParam("fl",fieldString)
out = ByteArrayOutputStream()
try:
self.indexer.search(req, out, self.format)
except:
#We can't get the result back from SOLR so fail back to the template display
self.errorMsg = "Query failure. Failed to load the data - this issue has been logged (%s - %s)." % (sys.exc_info()[0], sys.exc_info()[1])
self.log.error("Reporting threw an exception (report was %s); Error: %s - %s" % (self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1]))
return
try:
csvResponseString = String(out.toByteArray(),"utf-8")
csvResponseLines = csvResponseString.split("\n")
except:
#We can't get the result back from SOLR so fail back to the template display
self.errorMsg = "Query failure. Failed to prepare the CSV - this issue has been logged (%s - %s)." % (sys.exc_info()[0], sys.exc_info()[1])
self.log.error("Reporting threw an exception (report was %s); Error: %s - %s" % (self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1]))
return
fileName = self.urlEncode(self.report.getLabel())
self.log.debug("Generating CSV report with file name: " + fileName)
self.response.setHeader("Content-Disposition", "attachment; filename=%s.csv" % fileName)
sw = StringWriter()
parser = CSVParser()
writer = CSVWriter(sw)
count = 0
prevLine = ""
badRowFlag = False
for line in csvResponseLines:
if badRowFlag:
#In this section of code we'll handle errors by either trying to fix the problem
#or by adding an error line in the CSV. We'll then move to the next row and keep going
try:
self.log.debug("Reporting - trying to append the previous line with the previous faulty one. Line appears as: %s" % prevLine + line)
csvLine = parser.parseLine(prevLine + line)
badRowFlag = False
prevLine = ""
self.log.debug("Reporting - remedy appears to have worked. Line appears as: %s" % prevLine + line)
except:
#We tried to rescue the file but failed on the second run so give up
writer.writeNext(["Failed to transfer record to CSV - check logs"])
self.log.error("Reporting threw an exception (report was %s); Error: %s - %s; Result line: %s" % (self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1], prevLine + line))
else:
try:
csvLine = parser.parseLine(line)
badRowFlag = False
prevLine = ""
except:
#This can happen if there's a newline in the index data
#so we raise the badRowFlag and see if we can join this
#row to the next one to fix it
self.log.debug("Reporting threw an exception but I'll see if it's just a formatting issue (report was %s); Error: %s - %s; Result line: %s" % (self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1], line))
badRowFlag = True
prevLine = line
continue
if count == 0 :
#Header row
count += 1
for idx, csvValue in enumerate(csvLine):
csvLine[idx] = self.findDisplayLabel(csvValue)
elif csvLine[0] not in idQryList:
#ignore
continue
writer.writeNext(csvLine)
#Now send off the CSV
self.out = self.response.getOutputStream("text/csv")
self.out.print(sw.toString())
self.out.close()
def __activate__(self, context):
self.log = context["log"]
response = context["response"]
writer = response.getPrintWriter("text/plain; charset=UTF-8")
auth = context["page"].authentication
sessionState = context["sessionState"]
result = JsonObject()
result.put("status", "error")
result.put("message", "An unknown error has occurred")
if auth.is_admin():
services = context["Services"]
formData = context["formData"]
func = formData.get("func")
oid = formData.get("oid")
portalId = formData.get("portalId")
portalManager = services.portalManager
if func == "reharvest":
if oid:
self.log.debug("Reharvesting object '{}'", oid)
self.sendMessage(oid)
result.put("status", "ok")
result.put("message", "Object '%s' queued for reharvest")
elif portalId:
self.log.debug("Reharvesting view '{}'", portalId)
sessionState.set("reharvest/running/" + portalId, "true")
# TODO security filter - not necessary because this requires admin anyway?
portal = portalManager.get(portalId)
query = "*:*"
if portal.query != "":
query = portal.query
if portal.searchQuery != "":
if query == "*:*":
query = portal.searchQuery
else:
query = query + " AND " + portal.searchQuery
# query solr to get the objects to reharvest
rows = 25
req = SearchRequest(query)
req.setParam("fq", 'item_type:"object"')
req.setParam("rows", str(rows))
req.setParam("fl", "id")
done = False
count = 0
while not done:
req.setParam("start", str(count))
out = ByteArrayOutputStream()
services.indexer.search(req, out)
json = SolrResult(
ByteArrayInputStream(out.toByteArray()))
objectIds = HashSet(json.getFieldList("id"))
if not objectIds.isEmpty():
for oid in objectIds:
self.sendMessage(oid)
count = count + rows
total = json.getNumFound()
self.log.debug("Queued {} of {}...", min(count, total),
total)
done = (count >= total)
sessionState.remove("reharvest/running/" + portalId)
result.put("status", "ok")
result.put(
"message",
"Objects in '%s' queued for reharvest" % portalId)
else:
response.setStatus(500)
result.put("message",
"No object or view specified for reharvest")
elif func == "reindex":
if oid:
self.log.debug("Reindexing object '{}'", oid)
self.sendMessage(oid)
result.put("status", "ok")
result.put("message",
"Object '%s' queued for reindex" % oid)
else:
response.setStatus(500)
result.put("message", "No object specified to reindex")
else:
response.setStatus(500)
result.put("message", "Unknown action '%s'" % func)
else:
response.setStatus(500)
result.put("message",
"Only administrative users can access this API")
writer.println(result.toString())
writer.close()
def __reportSearch(self):
self.reportId = self.request.getParameter("id")
self.format = self.request.getParameter("format")
self.report = self.reportManager.getReports().get(self.reportId)
self.reportQuery = self.report.getQueryAsString()
self.log.debug("Report query: " + self.reportQuery)
#Get a total number of records
try:
out = ByteArrayOutputStream()
recnumreq = SearchRequest(self.reportQuery)
recnumreq.setParam("rows", "0")
self.indexer.search(recnumreq, out)
recnumres = SolrResult(ByteArrayInputStream(out.toByteArray()))
self.__rowsFoundSolr = "%s" % recnumres.getNumFound()
except:
self.errorMsg = "Query failure. The issue has been logged (%s - %s)." % (
sys.exc_info()[0], sys.exc_info()[1])
self.log.error(
"Reporting threw an exception (report was %s): %s - %s" %
(self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1]))
return
#Setup the main query
req = SearchRequest(self.reportQuery)
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", self.__rowsFoundSolr)
try:
#Now do the master search
out = ByteArrayOutputStream()
self.indexer.search(req, out)
self.__reportResult = SolrResult(
ByteArrayInputStream(out.toByteArray()))
self.__checkResults()
except:
self.errorMsg = "Query failure. The issue has been logged (%s - %s)." % (
sys.exc_info()[0], sys.exc_info()[1])
self.log.error(
"Reporting threw an exception (report was %s): %s - %s" %
(self.report.getLabel(), sys.exc_info()[0], sys.exc_info()[1]))
return
#At this point the display template has enough to go with.
#We just need to handle the CSV now
if (self.format == "csv"):
#Setup the main query - we need to requery to make sure we return
#only the required fields. We'll use the specific IDs that met the
#__checkResults check
req = SearchRequest(self.reportQuery)
req.setParam("fq", 'item_type:"object"')
req.setParam("fq", 'workflow_id:"dataset"')
req.setParam("rows", self.__rowsFoundSolr)
req.setParam("csv.mv.separator", ";")
#we need to get a list of the matching IDs from Solr
#this doesn't work for long queries so it's abandoned
#but left here commented to make sure we don't try it again
#idQry = ""
#for item in self.getProcessedResultsList():
# idQry += item.get("id") + " OR "
#req.setParam("fq", 'id:(%s)' % idQry[:len(idQry)-4])
#Create a list of IDs for reference when preparing the CSV
idQryList = []
for item in self.getProcessedResultsList():
idQryList.append(item.get("id"))
#Setup SOLR query with the required fields
self.fields = self.systemConfig.getArray("redbox-reports",
"csv-output-fields")
#We must have an ID field and it must be the first field
fieldString = "id,"
if self.fields is not None:
for field in self.fields:
fieldString = fieldString + field.get("field-name") + ","
fieldString = fieldString[:-1]
req.setParam("fl", fieldString)
out = ByteArrayOutputStream()
try:
self.indexer.search(req, out, self.format)
except:
#We can't get the result back from SOLR so fail back to the template display
self.errorMsg = "Query failure. Failed to load the data - this issue has been logged (%s - %s)." % (
sys.exc_info()[0], sys.exc_info()[1])
self.log.error(
"Reporting threw an exception (report was %s); Error: %s - %s"
% (self.report.getLabel(), sys.exc_info()[0],
sys.exc_info()[1]))
return
try:
csvResponseString = String(out.toByteArray(), "utf-8")
csvResponseLines = csvResponseString.split("\n")
except:
#We can't get the result back from SOLR so fail back to the template display
self.errorMsg = "Query failure. Failed to prepare the CSV - this issue has been logged (%s - %s)." % (
sys.exc_info()[0], sys.exc_info()[1])
self.log.error(
"Reporting threw an exception (report was %s); Error: %s - %s"
% (self.report.getLabel(), sys.exc_info()[0],
sys.exc_info()[1]))
return
fileName = self.urlEncode(self.report.getLabel())
self.log.debug("Generating CSV report with file name: " + fileName)
self.response.setHeader("Content-Disposition",
"attachment; filename=%s.csv" % fileName)
sw = StringWriter()
parser = CSVParser()
writer = CSVWriter(sw)
count = 0
prevLine = ""
badRowFlag = False
for line in csvResponseLines:
if badRowFlag:
#In this section of code we'll handle errors by either trying to fix the problem
#or by adding an error line in the CSV. We'll then move to the next row and keep going
try:
self.log.debug(
"Reporting - trying to append the previous line with the previous faulty one. Line appears as: %s"
% prevLine + line)
csvLine = parser.parseLine(prevLine + line)
badRowFlag = False
prevLine = ""
self.log.debug(
"Reporting - remedy appears to have worked. Line appears as: %s"
% prevLine + line)
except:
#We tried to rescue the file but failed on the second run so give up
writer.writeNext(
["Failed to transfer record to CSV - check logs"])
self.log.error(
"Reporting threw an exception (report was %s); Error: %s - %s; Result line: %s"
% (self.report.getLabel(), sys.exc_info()[0],
sys.exc_info()[1], prevLine + line))
else:
try:
csvLine = parser.parseLine(line)
badRowFlag = False
prevLine = ""
except:
#This can happen if there's a newline in the index data
#so we raise the badRowFlag and see if we can join this
#row to the next one to fix it
self.log.debug(
"Reporting threw an exception but I'll see if it's just a formatting issue (report was %s); Error: %s - %s; Result line: %s"
% (self.report.getLabel(), sys.exc_info()[0],
sys.exc_info()[1], line))
badRowFlag = True
prevLine = line
continue
if count == 0:
#Header row
count += 1
for idx, csvValue in enumerate(csvLine):
csvLine[idx] = self.findDisplayLabel(csvValue)
elif csvLine[0] not in idQryList:
#ignore
continue
writer.writeNext(csvLine)
#Now send off the CSV
self.out = self.response.getOutputStream("text/csv")
self.out.print(sw.toString())
self.out.close()
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
self.lastPage = False
# 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:
token = self.tokensDB.getToken(resumptionToken)
self.__currentToken = token
# 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()
if resumptionToken is None:
self.__buildResumptionTokenSets()
else:
self.__result = SolrResult(self.__currentToken.getResultJson())
# Only list records if the metadata format is enabled in this view
if self.isInView(self.__metadataPrefix) == False:
self.__result = None
# 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 __buildResumptionTokenSets(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
start = 0
req.setParam("start", str(start))
out = ByteArrayOutputStream()
self.services.indexer.search(req, out)
self.__result = SolrResult(ByteArrayInputStream(out.toByteArray()))
totalFound = self.__result.getNumFound()
self.log.debug("Total found:" + str(totalFound))
if totalFound > recordsPerPage:
startRow = 0
random.seed()
resumptionToken = "%016x" % random.getrandbits(128)
nextResumptionToken = resumptionToken
firstLoop = True
while True:
self.log.debug("Current Resumption Token: " + resumptionToken)
req.setParam("start", str(startRow))
out = ByteArrayOutputStream()
self.services.indexer.search(req, out)
result = SolrResult(ByteArrayInputStream(out.toByteArray()))
tokenObject = ResumptionToken(resumptionToken,
self.__metadataPrefix,
nextResumptionToken,
result.toString())
if firstLoop:
self.__currentToken = ResumptionToken(
None, self.__metadataPrefix, resumptionToken, None)
tokenObject = None
firstLoop = False
startRow = startRow + recordsPerPage
if startRow > totalFound:
tokenObject = ResumptionToken(resumptionToken,
self.__metadataPrefix, "",
result.toString())
self.tokensDB.storeToken(tokenObject)
break
if tokenObject is not None:
self.tokensDB.storeToken(tokenObject)
resumptionToken = nextResumptionToken
nextResumptionToken = "%016x" % random.getrandbits(128)
self.log.debug("Resumption Token: " + resumptionToken +
" next resumption token:" + nextResumptionToken)
def getToken(self):
if self.isInView(self.__metadataPrefix) and not self.lastPage:
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
def __activate__(self, context):
response = context["response"]
log = context["log"]
writer = response.getPrintWriter("text/plain; charset=UTF-8")
auth = context["page"].authentication
sessionState = context["sessionState"]
result = JsonObject()
result.put("status", "error")
result.put("message", "An unknown error has occurred")
if auth.is_admin():
services = context["Services"]
formData = context["formData"]
func = formData.get("func")
oid = formData.get("oid")
portalId = formData.get("portalId")
portalManager = services.portalManager
if func == "reharvest":
# One object
if oid:
log.info(" * Reharvesting object '{}'", oid)
portalManager.reharvest(oid)
result.put("status", "ok")
result.put("message", "Object '%s' queued for reharvest")
# The whole portal
elif portalId:
log.info(" * Reharvesting view '{}'", portalId)
sessionState.set("reharvest/running/" + portalId, "true")
# TODO security filter - not necessary because this requires admin anyway?
portal = portalManager.get(portalId)
query = "*:*"
if portal.query != "":
query = portal.query
if portal.searchQuery != "":
if query == "*:*":
query = portal.searchQuery
else:
query = query + " AND " + portal.searchQuery
# query solr to get the objects to reharvest
rows = 25
req = SearchRequest(query)
req.setParam("fq", 'item_type:"object"')
req.setParam("rows", str(rows))
req.setParam("fl", "id")
done = False
count = 0
while not done:
req.setParam("start", str(count))
out = ByteArrayOutputStream()
services.indexer.search(req, out)
json = SolrResult(ByteArrayInputStream(out.toByteArray()))
objectIds = HashSet(json.getFieldList("id"))
if not objectIds.isEmpty():
portalManager.reharvest(objectIds)
count = count + rows
total = json.getNumFound()
log.info(" * Queued {} of {}...", (min(count, total), total))
done = (count >= total)
sessionState.remove("reharvest/running/" + portalId)
result.put("status", "ok")
result.put("message", "Objects in '%s' queued for reharvest" % portalId)
else:
response.setStatus(500)
result.put("message", "No object or view specified for reharvest")
elif func == "reindex":
if oid:
log.info(" * Reindexing object '{}'", oid)
services.indexer.index(oid)
services.indexer.commit()
result.put("status", "ok")
result.put("message", "Object '%s' queued for reindex" % portalId)
else:
response.setStatus(500)
result.put("message", "No object specified to reindex")
else:
response.setStatus(500)
result.put("message", "Unknown action '%s'" % func)
else:
response.setStatus(500)
result.put("message", "Only administrative users can access this API")
writer.println(result.toString())
writer.close()
class HomeData:
def __init__(self):
pass
def __activate__(self, context):
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.services = self.vc("Services")
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 = self.services.getIndexer()
portalQuery = self.services.getPortalManager().get(self.vc("portalId")).getQuery()
portalSearchQuery = self.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()
class HarvestreportData:
def __init__(self):
pass
def __activate__(self, context):
self.velocityContext = context
self.vc("sessionState").remove("fq")
self.services = self.vc("Services")
self.request = self.vc("request")
self.log = context["log"]
self.__harvestedRecords = None
uri = URLDecoder.decode(self.request.getAttribute("RequestURI"))
self.__harvestId = os.path.basename(uri)
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 = self.services.getIndexer()
# Security prep work
isAdmin = self.vc("page").authentication.is_admin()
if not isAdmin:
print "ERROR: User is not an admin '"
return None
req = SearchRequest('harvestId:"' + self.__harvestId + '"')
req.setParam("fq", 'eventType:modify')
out = ByteArrayOutputStream()
try:
indexer.searchByIndex(req, out, "eventLog")
except:
print traceback.format_exc();
print repr(traceback.print_exc())
traceback.print_stack(file=sys.stdout)
self.__harvestedRecords = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest('harvestId:"' + self.__harvestId + '"')
req.setParam("fq", 'eventType:modify')
req.setParam("fq", 'isNew:true')
out = ByteArrayOutputStream()
indexer.searchByIndex(req, out, "eventLog")
self.__latest = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest('harvestId:"' + self.__harvestId + '"')
req.setParam("fq", 'eventType:modify')
req.setParam("fq", 'isModified:true')
out = ByteArrayOutputStream()
indexer.searchByIndex(req, out, "eventLog")
self.__modified = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest('harvestId:"' + self.__harvestId + '" AND eventType:"modify" AND isModified:false')
req.setParam("fq", 'isNew:false')
out = ByteArrayOutputStream()
indexer.searchByIndex(req, out, "eventLog")
self.__unmodified = SolrResult(ByteArrayInputStream(out.toByteArray()))
req = SearchRequest('harvestId:"' + self.__harvestId + '"')
req.setParam("fq", 'eventType:harvestEnd')
out = ByteArrayOutputStream()
indexer.searchByIndex(req, out, "eventLog")
endItem = SolrResult(ByteArrayInputStream(out.toByteArray()))
endTimeList = endItem.getFieldList('eventTime')
date = None;
repoType = None;
repoName = None;
if(endTimeList.size() > 0):
date = endTimeList.get(0)
repoTypeList = endItem.getFieldList('repository_type')
if(repoTypeList.size() > 0):
repoType = repoTypeList.get(0)
repoNameList = endItem.getFieldList('repository_name')
if(repoTypeList.size() > 0):
repoName = repoNameList.get(0)
req = SearchRequest('repository_type:"' + repoType + '" AND repository_name:"' + repoName + '" AND eventType:"modify" AND isNew:true')
req.setParam("fq", "eventTime:[* TO " + date + "]")
out = ByteArrayOutputStream()
indexer.searchByIndex(req, out, "eventLog")
self.__allRecords = SolrResult(ByteArrayInputStream(out.toByteArray()))
def getItemcount(self):
return self.__harvestedRecords.getNumFound()
def getNewcount(self):
return self.__latest.getNumFound()
def getModifiedcount(self):
return self.__modified.getNumFound()
def getUnmodifiedcount(self):
return self.__unmodified.getNumFound()
def getTotalcount(self):
return self.__allRecords.getNumFound()
