def run_morphline(self,
request,
collection_name,
morphline,
input_path,
query=None):
workspace_path = self._upload_workspace(morphline)
notebook = Notebook(name='Indexer job for %s' % collection_name,
isManaged=True)
if query:
q = Notebook(document=Document2.objects.get_by_uuid(user=self.user,
uuid=query))
notebook_data = q.get_data()
snippet = notebook_data['snippets'][0]
api = get_api(request, snippet)
destination = '__hue_%s' % notebook_data['uuid'][:4]
location = '/user/%s/__hue-%s' % (request.user,
notebook_data['uuid'][:4])
sql, success_url = api.export_data_as_table(notebook_data,
snippet,
destination,
is_temporary=True,
location=location)
input_path = '${nameNode}%s' % location
notebook.add_hive_snippet(snippet['database'], sql)
notebook.add_java_snippet(
clazz='org.apache.solr.hadoop.MapReduceIndexerTool',
app_jar=CONFIG_INDEXER_LIBS_PATH.get(),
arguments=[
u'--morphline-file',
u'morphline.conf',
u'--output-dir',
u'${nameNode}/user/%s/indexer' % self.username,
u'--log4j',
u'log4j.properties',
u'--go-live',
u'--zk-host',
zkensemble(),
u'--collection',
collection_name,
input_path,
],
files=[{
u'path': u'%s/log4j.properties' % workspace_path,
u'type': u'file'
}, {
u'path': u'%s/morphline.conf' % workspace_path,
u'type': u'file'
}])
return notebook.execute(request, batch=True)
def run_morphline(self, request, collection_name, morphline, input_path, query=None, start_time=None, lib_path=None):
workspace_path = self._upload_workspace(morphline)
task = make_notebook(
name=_('Indexing into %s') % collection_name,
editor_type='notebook',
on_success_url=reverse('search:browse', kwargs={'name': collection_name}),
pub_sub_url='assist.collections.refresh',
is_task=True,
is_notebook=True,
last_executed=start_time
)
if query:
q = Notebook(document=Document2.objects.get_by_uuid(user=self.user, uuid=query))
notebook_data = q.get_data()
snippet = notebook_data['snippets'][0]
api = get_api(request, snippet)
destination = '__hue_%s' % notebook_data['uuid'][:4]
location = '/user/%s/__hue-%s' % (request.user, notebook_data['uuid'][:4])
sql, _success_url = api.export_data_as_table(notebook_data, snippet, destination, is_temporary=True, location=location)
input_path = '${nameNode}%s' % location
task.add_hive_snippet(snippet['database'], sql)
client = SolrClient(self.user)
extra_args = ['-Dmapreduce.job.user.classpath.first=true'] if client.is_solr_six_or_more() else []
task.add_java_snippet(
clazz='org.apache.solr.hadoop.MapReduceIndexerTool',
app_jar=lib_path if lib_path is not None else CONFIG_INDEXER_LIBS_PATH.get(),
arguments=extra_args + [
u'--morphline-file',
u'morphline.conf',
u'--output-dir',
u'${nameNode}/user/%s/indexer' % self.username,
u'--log4j',
u'log4j.properties',
u'--go-live',
u'--zk-host',
client.get_zookeeper_host(),
u'--collection',
collection_name,
input_path,
],
files=[
{u'path': u'%s/log4j.properties' % workspace_path, u'type': u'file'},
{u'path': u'%s/morphline.conf' % workspace_path, u'type': u'file'}
]
)
return task.execute(request, batch=True)
def run_morphline(self, request, collection_name, morphline, input_path, query=None):
workspace_path = self._upload_workspace(morphline)
notebook = Notebook(
name='Indexer job for %s' % collection_name,
isManaged=True
)
if query:
q = Notebook(document=Document2.objects.get_by_uuid(user=self.user, uuid=query))
notebook_data = q.get_data()
snippet = notebook_data['snippets'][0]
api = get_api(request, snippet)
destination = '__hue_%s' % notebook_data['uuid'][:4]
location = '/user/%s/__hue-%s' % (request.user, notebook_data['uuid'][:4])
sql, success_url = api.export_data_as_table(notebook_data, snippet, destination, is_temporary=True, location=location)
input_path = '${nameNode}%s' % location
notebook.add_hive_snippet(snippet['database'], sql)
notebook.add_java_snippet(
clazz='org.apache.solr.hadoop.MapReduceIndexerTool',
app_jar=CONFIG_INDEXER_LIBS_PATH.get(),
arguments=[
u'--morphline-file',
u'morphline.conf',
u'--output-dir',
u'${nameNode}/user/%s/indexer' % self.username,
u'--log4j',
u'log4j.properties',
u'--go-live',
u'--zk-host',
zkensemble(),
u'--collection',
collection_name,
input_path,
],
files=[
{u'path': u'%s/log4j.properties' % workspace_path, u'type': u'file'},
{u'path': u'%s/morphline.conf' % workspace_path, u'type': u'file'}
]
)
return notebook.execute(request, batch=True)