def db_prepare(self):
# @TODO: race condition - use transactions (pool.runInteraction())
query = (select([Resources]).where(Resources.c.name == self.data['name']))
resource = yield tx_pool.runQuery(query)
if not resource:
query = (ResourceMeta.insert().values(
recursive_sizes=self.data['options']['recursive_foldersizes'],
web_user_agent=self.data['options']['user-agent'],
auth_user=self.data['options']['auth_user'],
auth_pass=self.data['options']['auth_pass']
).returning(ResourceMeta.c.id))
meta_id = yield tx_pool.runQuery(query)
meta_id = meta_id[0].id
query = (Resources.insert().values(
name=self.data['name'],
address=self.data['address'],
port=self.data['options']['port'],
protocol=FileProtocols().id_by_name(self.data['method']),
display_url=self.data['options']['display_url'],
date_crawl_start=datetime.now(),
basepath=self.data['basepath'],
meta_id=meta_id
).returning(Resources.c.id))
resource_id = yield tx_pool.runQuery(query)
self.data['resource_id'] = resource_id[0].id
query = (select([Resources]).where(Resources.c.id == self.data['resource_id']))
resource = yield tx_pool.runQuery(query)
resource = resource[0]
resource_meta = yield tx_pool.runQuery(select([ResourceMeta]).where(ResourceMeta.c.id == resource.meta_id))
resource_meta = resource_meta[0]
if resource_meta.busy:
raise Exception('This resource is already being crawled')
self.data['resource_id'] = resource.id
self.data['resource_meta_id'] = resource.meta_id
self.data['resource'] = resource
self.data['resource_meta'] = resource_meta
self.db_busy_crawling_toggle()
query = (Resources.update().where(Resources.c.id == resource.id).values(date_crawl_start=datetime.now()))
yield tx_pool.runOperation(query)
query = (Resources.update().where(Resources.c.id == resource.id).values(basepath=self.data['basepath']))
yield tx_pool.runOperation(query)
query = (Resources.update().where(Resources.c.id == resource.id).values(address=self.data['address']))
yield tx_pool.runOperation(query)
def db_calculate_filedistribution(resource_id):
log.msg("[%s] Calculating file distributions" % resource_id)
file_distribution = {}
query = (select([func.count()]).select_from(Files).where(Files.c.resource_id == resource_id))
total_file_count = yield tx_pool.runQuery(query)
total_file_count = int(total_file_count[0].count_1)
for k, v in FileCategories().data.iteritems():
query = (select([func.count()]).select_from(Files).where(Files.c.file_format == v).where(Files.c.resource_id == resource_id))
count = yield tx_pool.runQuery(query)
if count:
count = int(count[0].count_1)
pct = 100 * float(count)/float(total_file_count)
file_distribution[k] = "%.1f" % pct
else:
file_distribution[k] = 0
query = (ResourceMeta.update().where(ResourceMeta.c.id == resource_id).values(file_distribution=json.dumps(file_distribution)))
yield tx_pool.runOperation(query)
log.msg("[%s] Calculating file distributions DONE" % resource_id)
def db_finalize(self):
query = (Resources.update().where(Resources.c.id == self.data['resource_id']).values(date_crawl_end=datetime.now()))
yield tx_pool.runOperation(query)
query = (Files.delete().where(Files.c.resource_id == self.data['resource_id']))
yield tx_pool.runOperation(query)
query = (Files.update().where(Files.c.resource_id == '-%s' % str(self.data['resource_id'])).values(resource_id=self.data['resource_id']))
yield tx_pool.runOperation(query)
query = (ResourceMeta.update().where(ResourceMeta.c.id == self.data['resource_meta_id']).values(file_count=self.resource.db_files_inserts))
yield tx_pool.runOperation(query)
if 'recursive_foldersizes' in self.data['options']:
yield self.db_calculate_foldersizes()
self.db_calculate_filedistribution(resource_id=self.data['resource_id'])
self.db_busy_crawling_toggle()
def db_calculate_foldersizes(self):
log.msg("[%s] Recursively calculating folder sizes" % self.data['resource_id'])
dirs = ['/']
while dirs:
file_path = dirs[0]
query = (
select([Files.c.file_name, Files.c.file_path, Files.c.file_size, Files.c.file_isdir])
.where(Files.c.resource_id == self.data['resource_id'])
.where(Files.c.file_path == file_path))
all = yield tx_pool.runQuery(query)
for file_name in [z.file_name for z in all if z.file_isdir]:
query = (
select([func.sum(Files.c.file_size)])
.where(Files.c.resource_id == self.data['resource_id'])
.where(Files.c.file_path.like(file_path+file_name+'%')))
size = yield tx_pool.runQuery(query)
size = size[0]
if not size.sum_1:
size = 0
else:
size = long(size.sum_1)
query = (
Files.update()
.where(Files.c.resource_id == self.data['resource_id'])
.where(Files.c.file_path == file_path)
.where(Files.c.file_name == file_name)
.where(Files.c.file_isdir == True)
.values(file_size=size))
yield tx_pool.runOperation(query)
dirs.append('%s%s/' % (file_path, file_name))
dirs.pop(0)
log.msg("[%s] Recursively calculating folder sizes DONE" % self.data['resource_id'])
def db_busy_crawling_toggle(self):
resource_meta = yield tx_pool.runQuery(select([ResourceMeta]).where(ResourceMeta.c.id == self.data['resource_meta_id']))
resource_meta = resource_meta[0]
query = (ResourceMeta.update().where(ResourceMeta.c.id == self.data['resource_meta_id']).values(busy = 0 if resource_meta.busy else 1))
yield tx_pool.runOperation(query)