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 __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()
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()
