Files
pdga-rating/src/scrapers/player-http.js
T
Samuel Enocsson 6ac32457a9 feat: Add Pino structured logging, release-please CI/CD and Docker pipeline
Replace all console.log/error with Pino logger (info/warn/error/debug/fatal)
for structured JSON logging in production and pretty-print in development.
Remove redundant header dumps and consolidate rate-limit logging.

Add GitHub Actions workflow with release-please for automated semver releases
and Docker build/push to GHCR on new releases.
2026-02-21 15:56:57 +01:00

211 lines
6.5 KiB
JavaScript

const https = require('https');
const logger = require('../logger');
async function fetchPlayerDataHTTP(pdgaNumber) {
return new Promise((resolve, reject) => {
const options = {
hostname: 'www.pdga.com',
port: 443,
path: `/player/${pdgaNumber}`,
method: 'GET',
headers: {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'
},
timeout: 30000
};
const req = https.request(options, (res) => {
let data = '';
res.on('data', (chunk) => {
data += chunk;
});
res.on('end', () => {
if (res.statusCode === 200) {
resolve(data);
} else {
const rateLimitInfo = {
statusCode: res.statusCode,
headers: res.headers
};
logger.info(`PDGA Response Status for #${pdgaNumber}: ${res.statusCode}`);
logger.debug({
retryAfter: res.headers['retry-after'],
rateLimit: res.headers['x-ratelimit-limit'],
rateLimitRemaining: res.headers['x-ratelimit-remaining'],
rateLimitReset: res.headers['x-ratelimit-reset']
}, `Rate limit details for #${pdgaNumber}`);
const error = new Error(`HTTP ${res.statusCode}`);
error.rateLimitInfo = rateLimitInfo;
reject(error);
}
});
});
req.on('error', (error) => {
logger.error(`Request error for PDGA #${pdgaNumber}: ${error.code} ${error.message}`);
reject(error);
});
req.on('timeout', () => {
req.destroy();
reject(new Error('Request timeout'));
});
req.setTimeout(30000);
req.end();
});
}
function parsePlayerData(html, pdgaNumber) {
try {
const nameMatch = html.match(/<title>([^<]+?)\s*\|\s*Professional Disc Golf Association/i);
const name = nameMatch ? nameMatch[1].trim() : 'Unknown';
const ratingMatch = html.match(/Current Rating:[^>]*>\s*(\d+)/i);
const rating = ratingMatch ? parseInt(ratingMatch[1]) : 0;
const changeMatch = html.match(/Current Rating:[\s\S]*?([+\-]\d+)[\s\S]*?\(as of/i);
const ratingChange = changeMatch ? parseInt(changeMatch[1]) : null;
return {
pdgaNumber,
name: name.replace(/\s*#\d+$/, ''),
rating,
ratingChange,
predictedRating: null
};
} catch (error) {
logger.error(`Error parsing data for PDGA ${pdgaNumber}: ${error.message}`);
return {
pdgaNumber,
name: 'Error',
rating: 0,
ratingChange: null,
predictedRating: null
};
}
}
async function fetchRatingHistory(pdgaNumber) {
return new Promise((resolve, reject) => {
const options = {
hostname: 'www.pdga.com',
port: 443,
path: `/player/${pdgaNumber}/history`,
method: 'GET',
headers: {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'
},
timeout: 30000
};
logger.info(`Fetching rating history for PDGA #${pdgaNumber} from: https://www.pdga.com/player/${pdgaNumber}/history`);
const req = https.request(options, (res) => {
let data = '';
res.on('data', (chunk) => {
data += chunk;
});
res.on('end', () => {
if (res.statusCode === 200) {
logger.info(`Rating history request successful for PDGA #${pdgaNumber}`);
resolve(data);
} else {
logger.error(`Rating History Error for PDGA #${pdgaNumber}:`);
logger.error(`Status: ${res.statusCode}`);
logger.debug({
retryAfter: res.headers['retry-after'],
rateLimit: res.headers['x-ratelimit-limit'],
rateLimitRemaining: res.headers['x-ratelimit-remaining']
}, `Rate limit details for history #${pdgaNumber}`);
if (data.length > 0) {
logger.debug(`Partial response received (${data.length} bytes): ${data.substring(0, 200)}`);
}
const error = new Error(`HTTP ${res.statusCode} for rating history`);
error.statusCode = res.statusCode;
error.headers = res.headers;
reject(error);
}
});
});
req.on('error', (error) => {
logger.error({
code: error.code,
message: error.message,
errno: error.errno,
syscall: error.syscall
}, `Rating history request error for PDGA #${pdgaNumber}`);
if (error.code === 'ECONNRESET') {
logger.debug('Connection reset on rating history - likely rate limited by PDGA');
}
if (error.code === 'ECONNREFUSED') {
logger.debug('Connection refused - PDGA server may be blocking requests');
}
if (error.code === 'ETIMEDOUT') {
logger.debug('Request timed out - server may be overloaded');
}
reject(error);
});
req.on('timeout', () => {
logger.info(`Rating history request timeout for PDGA #${pdgaNumber} after 30s`);
req.destroy();
reject(new Error('Request timeout'));
});
req.setTimeout(30000);
req.end();
});
}
function parseRatingHistory(html) {
const history = [];
const rowMatches = html.match(/<tr[^>]*>[\s\S]*?<\/tr>/gi);
if (rowMatches) {
for (const row of rowMatches) {
if (row.includes('<th') || !row.includes('<td')) continue;
const cellMatches = row.match(/<td[^>]*>(.*?)<\/td>/gi);
if (cellMatches && cellMatches.length >= 2) {
const dateText = cellMatches[0].replace(/<[^>]*>/g, '').trim();
const ratingText = cellMatches[1].replace(/<[^>]*>/g, '').trim();
const dateMatch = dateText.match(/(\d{1,2})-([A-Za-z]{3})-(\d{4})/);
if (dateMatch && !isNaN(parseInt(ratingText))) {
const [, day, month, year] = dateMatch;
const monthMap = {
'Jan': 0, 'Feb': 1, 'Mar': 2, 'Apr': 3, 'May': 4, 'Jun': 5,
'Jul': 6, 'Aug': 7, 'Sep': 8, 'Oct': 9, 'Nov': 10, 'Dec': 11
};
const date = new Date(parseInt(year), monthMap[month], parseInt(day));
history.push({
date: date.toISOString().split('T')[0],
rating: parseInt(ratingText),
displayDate: dateText
});
}
}
}
}
return history.sort((a, b) => new Date(a.date) - new Date(b.date));
}
module.exports = { fetchPlayerDataHTTP, parsePlayerData, fetchRatingHistory, parseRatingHistory };