Last active
May 3, 2020 12:21
-
-
Save NeuroWhAI/08b548e7fe3e95f0046d50ea427b3c5c to your computer and use it in GitHub Desktop.
기상청 실시간 지진 감시 페이지 크롤러.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
using System; | |
using System.Collections.Generic; | |
using System.Linq; | |
using System.Text; | |
using System.Text.RegularExpressions; | |
using System.Threading.Tasks; | |
using System.Threading; | |
using System.Net.Http; | |
using System.Net; | |
using System.IO; | |
using System.Web; | |
namespace CSTest | |
{ | |
class Program | |
{ | |
const int HeadLength = 4; | |
const int MaxEqkStrLen = 60; | |
const int MaxEqkInfoLen = 120; | |
static string[] AreaNames = { "서울", "부산", "대구", "인천", "광주", "대전", "울산", "세종", "경기", "강원", "충북", "충남", "전북", "전남", "경북", "경남", "제주" }; | |
static bool StationUpdate = true; | |
static void Main(string[] args) | |
{ | |
using (var sw = new StreamWriter(new FileStream("log.txt", FileMode.Append))) | |
{ | |
sw.WriteLine(DateTime.UtcNow); | |
} | |
string prevBinTime = string.Empty; | |
double tide = 1000; | |
DateTime nextSyncTime = DateTime.MinValue; | |
while (true) | |
{ | |
try | |
{ | |
string binTime = DateTime.UtcNow.AddMilliseconds(-tide).ToString("yyyyMMddHHmmss"); | |
if (prevBinTime == binTime) | |
{ | |
continue; | |
} | |
prevBinTime = binTime; | |
Console.WriteLine(binTime); | |
string url = $"https://www.weather.go.kr/pews/data/{binTime}"; | |
byte[] bytes = null; | |
using (var client = new WebClient()) | |
{ | |
client.Headers.Add("user-agent", "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; .NET CLR 1.0.3705;)"); | |
bytes = client.DownloadData(url + ".b"); | |
// 시간 동기화. | |
if (DateTime.UtcNow >= nextSyncTime) | |
{ | |
nextSyncTime = DateTime.UtcNow + TimeSpan.FromSeconds(10.0); | |
string stStr = client.ResponseHeaders.Get("ST"); | |
if (!string.IsNullOrWhiteSpace(stStr) | |
&& double.TryParse(stStr, out double serverTime)) | |
{ | |
tide = DateTimeOffset.UtcNow.ToUnixTimeMilliseconds() - serverTime * 1000 + 1000; | |
Console.WriteLine("Sync: {0}", tide); | |
} | |
} | |
} | |
if (bytes != null && bytes.Length > MaxEqkStrLen) | |
{ | |
var headerBuff = new StringBuilder(); | |
for (int i = 0; i < HeadLength; ++i) | |
{ | |
headerBuff.Append(ByteToBinStr(bytes[i])); | |
} | |
string header = headerBuff.ToString(); | |
var bodyBuff = new StringBuilder(ByteToBinStr(bytes[0])); | |
for (int i = HeadLength; i < bytes.Length; ++i) | |
{ | |
bodyBuff.Append(ByteToBinStr(bytes[i])); | |
} | |
string body = bodyBuff.ToString(); | |
StationUpdate = (StationUpdate || (header[0] == '1')); | |
int phase = 0; | |
if (header[1] == '0') | |
{ | |
phase = 1; | |
} | |
else if (header[1] == '1' && header[2] == '0') | |
{ | |
phase = 2; | |
} | |
else if (header[2] == '1') | |
{ | |
phase = 3; | |
} | |
if (phase > 1) | |
{ | |
using (var client = new WebClient()) | |
{ | |
client.Headers.Add("user-agent", "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; .NET CLR 1.0.3705;)"); | |
if (!Directory.Exists("bin")) | |
{ | |
Directory.CreateDirectory("bin"); | |
} | |
client.DownloadFile(url + ".b", $"bin/{binTime}.b"); | |
} | |
var infoBytes = bytes.Skip(bytes.Length - MaxEqkStrLen).ToArray(); | |
HandleEqk(body, infoBytes); | |
} | |
if (StationUpdate) | |
{ | |
byte[] stnBytes = null; | |
using (var client = new WebClient()) | |
{ | |
client.Headers.Add("user-agent", "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; .NET CLR 1.0.3705;)"); | |
stnBytes = client.DownloadData(url + ".s"); | |
} | |
if (!Directory.Exists("bin")) | |
{ | |
Directory.CreateDirectory("bin"); | |
} | |
File.WriteAllBytes($"bin/{binTime}.s", stnBytes); | |
bodyBuff = new StringBuilder(); | |
for (int i = 0; i < stnBytes.Length; ++i) | |
{ | |
bodyBuff.Append(ByteToBinStr(stnBytes[i])); | |
} | |
HandleStn(bodyBuff.ToString(), body); | |
} | |
else | |
{ | |
var mmiData = ParseMmi(body); | |
if (mmiData.Max() >= 2) | |
{ | |
Console.WriteLine("관측소 현재 최대, 최소 진도: {0}, {1}", mmiData.Max(), mmiData.Min()); | |
mmiData.RemoveAll((v) => v <= 1); | |
if (mmiData.Count > 0) | |
{ | |
Console.WriteLine("진도 목록: {0}", string.Join(", ", mmiData)); | |
} | |
if (!Directory.Exists("bin")) | |
{ | |
Directory.CreateDirectory("bin"); | |
} | |
File.WriteAllBytes($"bin/{binTime}.b", bytes); | |
} | |
} | |
} | |
} | |
catch (Exception e) | |
{ | |
Console.WriteLine(e.Message); | |
Console.WriteLine(e.StackTrace); | |
using (var sw = new StreamWriter(new FileStream("log.txt", FileMode.Append))) | |
{ | |
sw.WriteLine(DateTime.UtcNow); | |
sw.WriteLine(e.Message); | |
sw.WriteLine(e.StackTrace); | |
} | |
} | |
Thread.Sleep(100); | |
} | |
} | |
static string ByteToBinStr(byte val) | |
{ | |
return Convert.ToString(val, 2).PadLeft(8, '0'); | |
} | |
static void HandleEqk(string body, byte[] infoBytes) | |
{ | |
string data = body.Substring(body.Length - (MaxEqkStrLen * 8 + MaxEqkInfoLen)); | |
string eqkStr = WebUtility.UrlDecode(Encoding.UTF8.GetString(infoBytes)); | |
double origLat = 30 + (double)Convert.ToInt32(data.Substring(0, 10), 2) / 100; | |
double origLon = 124 + (double)Convert.ToInt32(data.Substring(10, 10), 2) / 100; | |
double eqkMag = (double)Convert.ToInt32(data.Substring(20, 7), 2) / 10; | |
double eqkDep = (double)Convert.ToInt32(data.Substring(27, 10), 2) / 10; | |
int eqkTime = Convert.ToInt32(data.Substring(37, 32), 2); | |
string eqkId = "20" + Convert.ToInt32(data.Substring(69, 26), 2); | |
int eqkMax = Convert.ToInt32(data.Substring(95, 4), 2); | |
string eqkMaxAreaStr = data.Substring(99, 17); | |
var eqkMaxArea = new List<string>(); | |
if (eqkMaxAreaStr != new string('1', eqkMaxAreaStr.Length)) | |
{ | |
for (int i = 0; i < eqkMaxAreaStr.Length; ++i) | |
{ | |
if (eqkMaxAreaStr[i] == '1') | |
{ | |
eqkMaxArea.Add(AreaNames[i]); | |
} | |
} | |
} | |
Console.WriteLine("위도: {0}", origLat); | |
Console.WriteLine("경도: {0}", origLon); | |
Console.WriteLine("규모: {0}", eqkMag); | |
Console.WriteLine("깊이: {0}", eqkDep); | |
Console.WriteLine("시각(UTC): {0}", DateTimeOffset.FromUnixTimeSeconds(eqkTime).AddHours(9.0)); | |
Console.WriteLine("ID: {0}", eqkId); | |
Console.WriteLine("문구: {0}", eqkStr); | |
Console.WriteLine("진도: {0}", eqkMax); | |
Console.WriteLine("지역: {0}", string.Join(", ", eqkMaxArea)); | |
} | |
static void HandleStn(string stnBody, string binBody) | |
{ | |
var stnLat = new List<double>(); | |
var stnLon = new List<double>(); | |
for (int i = 0; i + 20 <= stnBody.Length; i += 20) | |
{ | |
stnLat.Add(30 + (double)Convert.ToInt32(stnBody.Substring(i, 10), 2) / 100); | |
stnLon.Add(120 + (double)Convert.ToInt32(stnBody.Substring(i + 10, 10), 2) / 100); | |
} | |
Console.WriteLine("관측소 수: {0}", stnLat.Count); | |
if (stnLat.Count < 99) | |
{ | |
// 재시도. | |
return; | |
} | |
var mmiData = ParseMmi(binBody); | |
Console.WriteLine("관측소 현재 최대, 최소 진도: {0}, {1}", mmiData.Max(), mmiData.Min()); | |
// mmiData[i]가 stnLat[i], stnLon[i]에 위치한 관측소의 진도 정보임. | |
// 단, 목록의 수가 같지 않을 수 있음을 주의. (보통 mmiData의 수가 더 많음.) | |
mmiData.RemoveAll((v) => v <= 1); | |
if (mmiData.Count > 0) | |
{ | |
Console.WriteLine("진도 목록: {0}", string.Join(", ", mmiData)); | |
} | |
if (stnLat.Count >= 99) | |
{ | |
// 얻은 관측소 수가 충분하여 다음 업데이트 대기. | |
StationUpdate = false; | |
} | |
} | |
static List<int> ParseMmi(string data) | |
{ | |
var mmiData = new List<int>(); | |
string mmiBody = data.Split(new[] { "11111111" }, StringSplitOptions.None).First(); | |
for (int i = 8; i < mmiBody.Length; i += 4) | |
{ | |
mmiData.Add(Convert.ToInt32(mmiBody.Substring(i, 4), 2)); | |
} | |
return mmiData; | |
} | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
https://www.weather.go.kr/pews/pews.html