|
| 1 | +#!/usr/bin/env python3 |
| 2 | + |
| 3 | +import argparse |
| 4 | +import glob |
| 5 | +import os.path |
| 6 | +import sys |
| 7 | +import gzip |
| 8 | +import csv |
| 9 | +from datetime import datetime |
| 10 | +from dateutil import parser as dateparser |
| 11 | + |
| 12 | + |
| 13 | +def get_files(prefix): |
| 14 | + files = [] |
| 15 | + files += glob.glob(prefix + '*.gz') |
| 16 | + files = list(map(lambda s: s.replace('\\', '/'), files)) |
| 17 | + files.sort() |
| 18 | + if os.path.isfile(prefix): |
| 19 | + files.append(prefix) |
| 20 | + return files |
| 21 | + |
| 22 | + |
| 23 | +def _text_max_ts(r, max_ts): |
| 24 | + if r: |
| 25 | + if int(r[1]) > max_ts: |
| 26 | + exit(0) |
| 27 | + return r |
| 28 | + else: |
| 29 | + return None |
| 30 | + |
| 31 | + |
| 32 | +def make_filters(min_time, max_time, seconds, convert_time): |
| 33 | + filters = [] |
| 34 | + |
| 35 | + if min_time: |
| 36 | + min_dt = dateparser.parse(min_time) |
| 37 | + min_ts = int(min_dt.timestamp()) |
| 38 | + filters.append(lambda r: r if r and int(r[1]) >= min_ts else None) |
| 39 | + if seconds: |
| 40 | + max_ts = min_ts + int(seconds) |
| 41 | + filters.append(lambda r: _text_max_ts(r, max_ts)) |
| 42 | + |
| 43 | + if max_time: |
| 44 | + max_dt = dateparser.parse(max_time) |
| 45 | + max_ts = int(max_dt.timestamp()) |
| 46 | + filters.append(lambda r: _text_max_ts(r, max_ts)) |
| 47 | + |
| 48 | + if convert_time: |
| 49 | + filters.append(lambda r: [r[0], datetime.fromtimestamp( |
| 50 | + int(r[1])).strftime('%Y-%m-%d %H:%M:%S')] + r[2:] if r else None) |
| 51 | + |
| 52 | + return filters |
| 53 | + |
| 54 | + |
| 55 | +def process_file(file, filters, writer): |
| 56 | + if file.endswith('.gz'): |
| 57 | + fp = gzip.open(file, 'rt', encoding='utf-8') |
| 58 | + else: |
| 59 | + fp = open(file, 'r', encoding='utf-8') |
| 60 | + |
| 61 | + reader = csv.reader(fp, delimiter=',') |
| 62 | + for row in reader: |
| 63 | + for filter in filters: |
| 64 | + row = filter(row) |
| 65 | + if not row: |
| 66 | + continue |
| 67 | + if row: |
| 68 | + writer.writerow(row) |
| 69 | + |
| 70 | + fp.close() |
| 71 | + |
| 72 | + |
| 73 | +if __name__ == "__main__": |
| 74 | + sys.stdin.reconfigure(encoding='utf-8') |
| 75 | + sys.stdout.reconfigure(encoding='utf-8') |
| 76 | + sys.stderr.reconfigure(encoding='utf-8') |
| 77 | + |
| 78 | + # Process program arguments... |
| 79 | + parser = argparse.ArgumentParser() |
| 80 | + parser.add_argument("--log", type=str, help="Prefix path to user actions log ('./user_actions.log' is default)") |
| 81 | + parser.add_argument('--after', type=str, help="Show records after given time.") |
| 82 | + parser.add_argument('--before', type=str, help="Show records before given time.") |
| 83 | + parser.add_argument('--seconds', type=str, help="If 'after' is set, then time window of given seconds is listed.") |
| 84 | + parser.add_argument('--convert-time', action='store_true', |
| 85 | + help="If set, unix ts will be converted to readable format.") |
| 86 | + args = parser.parse_args() |
| 87 | + |
| 88 | + writer = csv.writer(sys.stdout, delimiter=',') |
| 89 | + filters = make_filters(args.after, args.before, args.seconds or 0, args.convert_time) |
| 90 | + files = get_files(args.log or './user_actions.log') |
| 91 | + for file in files: |
| 92 | + process_file(file, filters, writer) |
0 commit comments