From a5444a864a8d4b848630e425d3bec1f9f0a68add Mon Sep 17 00:00:00 2001
From: "vy.boyko" <vy.boyko@tinkoff.ru>
Date: Wed, 16 Oct 2024 11:39:04 +0300
Subject: [PATCH] stats over logs

---
 .gitignore     |  4 +++
 pyproject.toml | 14 ++++++++
 src/app.py     | 92 ++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 110 insertions(+)
 create mode 100644 .gitignore
 create mode 100644 pyproject.toml
 create mode 100644 src/app.py

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..4595a6c
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,4 @@
+temp/**
+.idea
+.idea/**
+
diff --git a/pyproject.toml b/pyproject.toml
new file mode 100644
index 0000000..c0fdb13
--- /dev/null
+++ b/pyproject.toml
@@ -0,0 +1,14 @@
+[tool.poetry]
+name = "java-logs-analyzer"
+version = "0.1.0"
+description = ""
+authors = ["vy.boyko <vy.boyko@tinkoff.ru>"]
+readme = "README.md"
+
+[tool.poetry.dependencies]
+python = "^3.12"
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"
diff --git a/src/app.py b/src/app.py
new file mode 100644
index 0000000..821c4a3
--- /dev/null
+++ b/src/app.py
@@ -0,0 +1,92 @@
+import argparse
+import re
+from datetime import datetime
+import os
+
+parser = argparse.ArgumentParser(prog='java-logs-analyzer')
+parser.add_argument('-l', '--logs', help='Logs file')
+args = parser.parse_args()
+
+pattern = re.compile(r"""^(?P<day>\d{1,2})[-](?P<month>\d{2})[-](?P<year>\d{4})\s(?P<hour>\d{2})[:](?P<minute>\d{2})[:](?P<second>\d{2})[.]""", re.VERBOSE)
+stats_by_second = {}
+stats_by_minute = {}
+stats_by_hour = {}
+first_line = None
+last_line = None
+total_len = 0
+file_len = os.path.getsize(args.logs)
+
+
+def _add_stats(stats: dict, key: str, value: int) -> dict:
+    if key in stats:
+        return stats | { key: stats[key] + value }
+    else:
+        return stats | { key: value }
+
+
+def _calc_stats(stats: dict) -> (int, int, float):
+    avg = 0
+    min_val = 0
+    max_val = 0
+    count = len(stats.keys())
+    for key in stats.keys():
+        avg += stats[key]
+        min_val = min(min_val, stats[key]) if min_val > 0 else stats[key]
+        max_val = max(max_val, stats[key]) if max_val > 0 else stats[key]
+    return (min_val, max_val, avg / count if count > 0 else 0)
+
+def _print_stats(name: str, stats: dict) -> None:
+    (min_val, max_val, avg) = _calc_stats(stats)
+    print(f"{name}: MIN={min_val}, MAX={max_val}, AVG={avg}")
+
+def _parse(line: str) -> (bool, str, str, str, str, str, str):
+   match = pattern.match(line)
+   if match is not None:
+       day = match.group("day")
+       month = match.group("month")
+       year = match.group("year")
+       hour = match.group("hour")
+       minute = match.group("minute")
+       second = match.group("second")
+       return (True, year, month, day, hour, minute, second)
+   else:
+       return (False, None, None, None, None, None, None)
+
+
+with open(args.logs, 'r') as file:
+    read_len = 0
+    last_percent = -1
+    percent = 0
+    for line in file:
+        l = line.strip()
+        read_len += len(line)
+        percent = round(read_len / file_len * 100, 2)
+        cur_percent = round(percent / 10, 0)
+        if last_percent != cur_percent:
+            print(f"{percent}%", end='\r')
+            last_percent = cur_percent
+        (is_parsed, year, month, day, hour, minute, second) = _parse(l)
+        if is_parsed:
+            if first_line is None:
+                first_line = l
+            last_line = l
+            total_len += len(l)
+            key_hour = f"{year}-{month}-{day} {hour}"
+            key_minute = f"{year}-{month}-{day} {hour}:{minute}"
+            key_second = f"{year}-{month}-{day} {hour}:{minute}:{second}"
+            stats_by_hour = _add_stats(stats_by_hour, key_hour, 1)
+            stats_by_minute = _add_stats(stats_by_minute, key_minute, 1)
+            stats_by_second = _add_stats(stats_by_second, key_second, 1)
+
+
+_print_stats("Stats by hour", stats_by_hour)
+_print_stats("Stats by minute", stats_by_minute)
+_print_stats("Stats by second", stats_by_second)
+
+(fl_is_parsed, fl_year, fl_month, fl_day, fl_hour, fl_minute, fl_second) = _parse(first_line)
+(ll_is_parsed, ll_year, ll_month, ll_day, ll_hour, ll_minute, ll_second) = _parse(last_line)
+started = datetime(*map(int, [fl_year, fl_month, fl_day, fl_hour, fl_minute, fl_second]))
+finished = datetime(*map(int, [ll_year, ll_month, ll_day, ll_hour, ll_minute, ll_second]))
+delta = finished - started
+seconds = delta.seconds
+print(f"Average bytes per second: {total_len / seconds}")
\ No newline at end of file