class DuplicatesPipeline(object):
def __init__(self):
servers = SOLR_SERVERS
self.cnn = SolrConnection(servers)[SOLR_COLLECTION_DEFAULT]
self.cache_list = []
def process_item(self, item, spider):
if self.cnn.search(
{"q": "report_link:%s" % item["report_link"].encode("utf-8")}
).result.response.numFound != 0 or self.cache_duplicate(item["report_link"]):
raise DropItem("Duplicate item found: %s" % item["report_link"])
else:
if "report_revision_time_standard" in item:
delta = datetime.timedelta(hours=8)
dateTimezone = item["report_revision_time_standard"] - delta
item["report_revision_time_standard"] = dateTimezone.strftime("%Y-%m-%dT%H:%M:%S") + "Z"
return item
def cache_duplicate(self, report_link):
if report_link in self.cache_list:
return True
else:
if len(self.cache_list) > SOLR_CACHE_MAX_ELEMENTS_PER_SPIDER:
self.cache_list = []
else:
self.cache_list.append(report_link)
return False
class DuplicatesPipeline(object):
def __init__(self):
servers = SOLR_SERVERS
self.cnn = SolrConnection(servers)[SOLR_COLLECTION_DEFAULT]
self.cache_list = []
def process_item(self, item, spider):
if self.cnn.search({"q":'report_link:%s' % item['report_link'].encode('utf-8')}).result.response.numFound != 0 \
or self.cache_duplicate(item["report_link"]):
raise DropItem("Duplicate item found: %s" % item['report_link'])
else:
if "report_revision_time_standard" in item:
delta = datetime.timedelta(hours=8)
dateTimezone = item["report_revision_time_standard"] - delta
item["report_revision_time_standard"] = dateTimezone.strftime(
'%Y-%m-%dT%H:%M:%S') + "Z"
return item
def cache_duplicate(self, report_link):
if report_link in self.cache_list:
return True
else:
if len(self.cache_list) > SOLR_CACHE_MAX_ELEMENTS_PER_SPIDER:
self.cache_list = []
else:
self.cache_list.append(report_link)
return False
class SolrBackend(object):
def __init__(self, table, core="collection1"):
self.table = table
self.core = core
self.url = 'localhost:8983'
try:
self.interface = SolrConnection(self.url)[self.core]
except Exception as e:
logger.warning("Cannot connect to Solr: %s" % e)
raise RuntimeError("Cannot connect to Solr: %s" % e)
def get_ids(self, queryset):
return [r.id for r in queryset.select(self.table._id)]
def indexes(self, *fieldnames):
self.fieldnames = fieldnames
def after_insert(self, fields, id):
document = [{'id': id}]
for name in self.fieldnames:
if name in fields:
document[0][name] = str(fields[name])
self.interface.add(document)
self.interface.commit()
return True
def after_update(self, queryset, fields):
""" caveat, this should work but only if ALL indexed fields are updated at once """
ids = self.get_ids(queryset)
documents = []
for id in ids:
self.interface.delete({'q':'id:%i'%id})
document = {'id':id}
for name in self.fieldnames:
if name in fields:
document[name] = str(fields[name])
documents.append(document)
self.interface.add(documents)
self.interface.commit()
return True
def index_table(self, query, db):
for row in db(query).select():
self.interface.delete({'q':'id:%i'%row.id})
self.interface.commit()
documents = []
for row in db(query).select():
document = {'id':row.id}
for name in self.fieldnames:
document[name] = str(row[name])
documents.append(document)
self.interface.add(documents)
self.interface.commit()
return True
def update(self, query, fields, db, **core_fields):
rows = db(query).select(*fields)
documents = []
for row in rows:
document={}
for key in row.keys():
for core_field in core_fields:
if core_field in key:
document[key] = str(row[key])
if key == 'id':
self.interface.delete({'q':'id:%i'%row[key]})
documents.append(document)
self.interface.add(documents)
self.interface.commit()
return True
def before_delete(self, queryset):
self.ids = self.get_ids(queryset)
return False
def after_delete(self, queryset):
self.ids = self.get_ids(queryset)
for id in self.ids:
self.interface.delete({'q':'id:%i'%id})
self.interface.commit()
return True
def meta_search(self, limit, offset, mode, compact, sort, **fieldkeys):
query = ''
items = len(fieldkeys)
count = 0
# Convert to solrcloudpy search
for fieldkey in fieldkeys:
query += " %s:%s " % (fieldkey, fieldkeys[fieldkey])
count += 1
if items > 1 and count < items:
query += mode
se = SearchOptions()
se.commonparams.q(query).rows(limit).sort(sort).start(offset)
response = self.interface.search(se)
if compact:
return [r['id'] for r in response.result['response'].docs]
return response.result['response']
class SolrBackend(object):
def __init__(self, table, core="collection1"):
self.table = table
self.core = core
self.url = 'localhost:8983'
try:
self.interface = SolrConnection(self.url)[self.core]
except Exception as e:
logger.warning("Cannot connect to Solr: %s" % e)
raise RuntimeError("Cannot connect to Solr: %s" % e)
def get_ids(self, queryset):
return [r.id for r in queryset.select(self.table._id)]
def indexes(self, *fieldnames):
self.fieldnames = fieldnames
def after_insert(self, fields, id):
document = [{'id': id}]
for name in self.fieldnames:
if name in fields:
document[0][name] = unicode(fields[name])
self.interface.add(document)
self.interface.commit()
return True
def after_update(self, queryset, fields):
""" caveat, this should work but only if ALL indexed fields are updated at once """
ids = self.get_ids(queryset)
documents = []
for id in ids:
self.interface.delete(id)
document = {'id':id}
for name in self.fieldnames:
if name in fields:
document[name] = unicode(fields[name])
documents.append(document)
self.interface.add(documents)
self.interface.commit()
return True
def update(self, query, fields, db, **core_fields):
'''
Usage:
'''
rows = db(query).select(*fields)
documents = []
for row in rows:
document={}
for key in row.keys():
for core_field in core_fields:
if core_field in row[key]:
document[core_fields[core_field]] = unicode(row[key][core_field])
if core_field == 'id':
self.interface.delete(row[key][core_field])
documents.append(document)
self.interface.add(documents)
self.interface.commit()
return True
def before_delete(self, queryset):
self.ids = self.get_ids(queryset)
return False
def after_delete(self):
for id in self.ids:
self.interface.delete(id=id)
self.interface.commit()
return True
def meta_search(self, limit, offset, mode, compact, sort, **fieldkeys):
query = ''
items = len(fieldkeys)
count = 0
# Convert to solrcloudpy search
for fieldkey in fieldkeys:
query += " %s:%s " % (fieldkey, fieldkeys[fieldkey])
count += 1
if items > 1 and count < items:
query += mode
se = SearchOptions()
se.commonparams.q(query).rows(limit).sort(sort).start(offset)
print se
response = self.interface.search(se)
if compact:
return [r['id'] for r in response.result['response'].docs]
return response.result['response']
