def _get_job_rule(pool: happybase.ConnectionPool,
job_name) -> crawler.CrawlJobCore:
'''
获取 hbase 里的 crawl_job_core (爬取规则)
'''
with pool.connection() as conn:
try:
conn: happybase.Connection
table = conn.table(job_name)
row = table.row(rule_row_key, columns=[
rule_col,
])
rule = row[bytes(rule_col, encoding="utf-8")].decode("utf-8")
# _json_str = row.values
# print(rule)
common.print_info("get crawl rule: {}".format(rule))
crawl_job_core = crawler.CrawlJobCore.loads(rule)
# TODO 键 有点问题
return crawl_job_core
except Exception as e:
common.print_exception(e)
return None
pass
finally:
conn.close() # 关闭连接
def _save_results(pool: happybase.ConnectionPool, crawl_job_core, url,
result_list) -> bool:
'''
保存爬取结果到 hbase 里
如果 result_list 为空，不进行操作
'''
if not bool(result_list):
return False
core = crawl_job_core
with pool.connection() as conn:
try:
conn: happybase.Connection
table = conn.table(core.name)
row_key = url
table.put(row_key, {
results_col_pattern(i): ele
for i, ele in enumerate(result_list)
})
return True
except Exception as e:
common.print_exception(e)
return False
pass
finally:
conn.close() # 关闭连接
def _get_job_result(pool: happybase.ConnectionPool, crawl_job_name) -> list:
'''
获取爬虫结果
'''
with pool.connection() as conn:
try:
conn: happybase.Connection
table = conn.table(crawl_job_name)
result_list = []
for key, value in table.scan(include_timestamp=True):
tmp = {}
tmp['url'] = key.decode("utf-8")
# tmp['result']={ele.decode("utf-8"):value[ele].decode("utf-8") for ele in value}
tmp['result'] = {
ele.decode("utf-8"):
(value[ele][0].decode("utf-8"), value[ele][1])
for ele in value
}
result_list.append(tmp)
return result_list
except Exception as e:
common.print_exception(e)
return None
pass
finally:
conn.close() # 关闭连接
def run_task_fetcher(self):
'''
从 redis 中获取 任务
'''
while not self.end_flag:
try:
obj_tuple = QUEUE.get_wait(timeout=self.timeout)
# print(obj)
if obj_tuple is None:
# 取出为空，说明超时了
continue
_, content = obj_tuple
task_info = CrawlTaskJson.from_json_str(content)
# 如果该任务在 close_set 里，说明它被手动关闭了
if CLOSE_SET.is_member(task_info.job_name):
common.print_info(
"this crawl_job has been closed: {}".format(
task_info.job_name))
continue
# 判断是否为合法url
for url in task_info.urls:
assert common.urltools.check_url(url)
# 阻塞一段时间，防止其他节点抢不到
time.sleep(0.5)
self.add_urls(task_info.job_name, task_info.layer,
task_info.urls)
except Exception as e:
common.print_exception(e)
def work(self, tasks: common.LockedIterator):
'''
接收 crawl_tasks，并执行这些 task
'''
while not self.end_flag:
try:
self.activate()
for task in tasks:
task(self.driver)
except Exception as e:
common.print_exception(e)
continue
finally:
self.close()
def process(self):
while not self.quit:
#self.polling()
try:
self.polling()
found = self.queue.get(True, self.default_polling_interval)
found.process(self)
except queue.Empty:
found = None
except KeyboardInterrupt:
found = None
self.quit = True
except Exception as e:
print_exception()
def _get_job_list(pool: happybase.ConnectionPool) -> list:
'''
获取hbase中存的job名称list
'''
with pool.connection() as conn:
try:
conn: happybase.Connection
table_list = conn.tables()
return table_list
except Exception as e:
common.print_exception(e)
return None
pass
finally:
conn.close() # 关闭连接
def _remove_job(pool, crawl_job_name) -> bool:
'''
删除 job (删除job_name所对应的表)
要想删除 hbase 的表，应该先 disable 掉它
'''
with pool.connection() as conn:
try:
conn: happybase.Connection
conn.delete_table(crawl_job_name, disable=True)
return True
except Exception as e:
common.print_exception(e)
return False
pass
finally:
conn.close() # 关闭连接
def export_csv(service, docid, filename_template='%(title)s - %(sheet)s.csv'):
error_count = common.error_count # Note original error count
if docid and re.match("^https://docs.google.com/spreadsheets/d/", docid, re.IGNORECASE):
docid = docid.split("/")[5]
try:
for (doc, sheet), rows in itersheets(service, docid):
file_name = filename_template % {'title': doc, 'sheet': sheet}
file_path = os.path.join(common.args[TITLE_DIR], common.args[TITLE_PREFIX], file_name)
common.print_notice("Saving \"%s\" sheet to file: %s" % (common.colour_text(sheet), common.colour_text(file_name, common.COLOUR_GREEN)))
with open(file_name, 'wb') as fd:
write_csv(service, fd, rows)
except Exception as e:
common.print_exception(e)
return error_count == common.error_count
def _set_job_rule(pool: happybase.ConnectionPool, crawl_job_core) -> bool:
'''
改变规则
'''
core = crawl_job_core
with pool.connection() as conn:
try:
conn: happybase.Connection
table = conn.table(core.name)
table.put(rule_row_key, {
rule_col: core.dumps(),
})
return True
except Exception as e:
common.print_exception(e)
return False
pass
finally:
conn.close() # 关闭连接
def _save_job(pool: happybase.ConnectionPool, crawl_job_core)-> bool:
'''
存储 crawl_job_core (爬取规则) 到 hbase 里
'''
core = crawl_job_core
with pool.connection() as conn:
try:
conn: happybase.Connection
conn.create_table(name=core.name, families={
rule_col: dict(max_versions=rule_max_version),
results_family: dict(max_versions=results_max_version),
})
table = conn.table(core.name)
table.put(rule_row_key, {
rule_col: core.dumps()
})
return True
except Exception as e:
common.print_exception(e)
return False
pass
finally:
conn.close() # 关闭连接
