diff --git a/2-9.py b/2-9.py index e69de29..e3c5a88 100644 --- a/2-9.py +++ b/2-9.py @@ -0,0 +1,74 @@ +import re +from pyspark import SparkContext +import matplotlib.pyplot as plt + +# 初始化 SparkContext +sc = SparkContext.getOrCreate() + +# 日志匹配的正则表达式 +LOG_PATTERN = re.compile( + r'^(\S+) (\S+) (\S+) \[([\w:/]+\s[+-]\d{4})\] "(\S+) (\S+)\s*(\S*)\s?" (\d{3}) (\S+)' +) + +# 解析日志的函数 +def parse_log_line(line): + match = LOG_PATTERN.match(line) + if not match: + return None + + content_size_str = match.group(9) + content_size = int(content_size_str) if content_size_str.isdigit() else 0 + + return { + 'ip': match.group(1), + 'user_identity': match.group(2), + 'user_id': match.group(3), + 'timestamp': match.group(4), + 'method': match.group(5), + 'endpoint': match.group(6), + 'protocol': match.group(7), + 'status_code': int(match.group(8)), + 'content_size': content_size + } + +# 提取小时 +def extract_hour(log): + timestamp = log['timestamp'] + hour = timestamp.split(":")[1] # 从时间戳中提取小时 + return hour + +if __name__ == "__main__": + # 加载日志文件 + logFile = "hdfs://master:9000/user/root/apache.access.log.PROJECT" + raw_logs = sc.textFile(logFile) + + # 解析并过滤有效日志 + access_logs = raw_logs.map(parse_log_line).filter(lambda x: x is not None).cache() + + # 只保留 404 响应代码的日志 + bad_records = access_logs.filter(lambda log: log['status_code'] == 404).cache() + + # 提取每小时的 404 错误记录 + hourly_404_counts = bad_records.map(lambda log: (extract_hour(log), 1)) \ + .reduceByKey(lambda a, b: a + b) \ + .sortByKey() # 按小时排序 + + # 将结果转换为列表 + hourly_404_counts_list = hourly_404_counts.collect() + + # 提取小时和对应的 404 次数 + hours = [hour for hour, count in hourly_404_counts_list] + counts = [count for hour, count in hourly_404_counts_list] + + # 使用 matplotlib 绘制条形图 + plt.figure(figsize=(10, 6)) + plt.bar(hours, counts, color='blue') + plt.title('每小时404响应代码数量') + plt.xlabel('小时') + plt.ylabel('404响应次数') + plt.xticks(rotation=45) # 将小时标签旋转 45 度 + plt.tight_layout() + plt.show() + + # 停止 Spark + sc.stop()