scraperapi-java-sdk

ScraperAPI — Java SDK 最佳实践

要求： Java 8+、Maven 或 Gradle、SCRAPERAPI_API_KEY 环境变量。

设置

Maven

<dependency>
  <groupId>com.scraperapi</groupId>
  <artifactId>sdk</artifactId>
  <version>1.2</version>
</dependency>

Gradle

implementation 'com.scraperapi:sdk:1.2'

实例化客户端

import com.scraperapi.ScraperApiClient;

ScraperApiClient client = new ScraperApiClient(System.getenv("SCRAPERAPI_API_KEY"));

切勿将 API 密钥硬编码。每次都从环境变量中读取。

基本用法

Java SDK 使用流式构建器模式。在 .get() 的结果上链式调用参数方法，然后调用 .result() 阻塞并获取 HTML。

// Simple GET — returns HTML as a String
String html = client.get("https://example.com/").result();

// With parameters — chain before .result()
String html = client.get("https://example.com/")
    .render(true)
    .result();

// Multiple parameters
String html = client.get("https://example.com/")
    .render(true)
    .countryCode("us")
    .result();

选择指南

情况	方案
单个 URL，同步处理	.get(url).<params>.result()
页面通过 JavaScript 加载内容	链式调用 .render(true)
网站屏蔽数据中心代理	链式调用 .premium(true)
最强反机器人保护	链式调用 .ultraPremium(true)
同一域名上的多步骤／分页流程	链式调用 .sessionNumber(n)
可能发生暂时性故障	链式调用 .retry(n)
20 个以上 URL 或批处理任务	使用 HttpClient 调用异步端点
受支持的平台（Amazon、Google 等）	结构化数据端点

参数参考

渲染

// Render JavaScript before returning HTML
// Use when: page is a React/Vue/Angular SPA, or initial scrape returns empty content
// Cost: +10 credits
String html = client.get("https://spa-site.com/").render(true).result();

// Wait for a DOM element before capturing (requires render)
String html = client.get("https://spa-site.com/")
    .render(true)
    .waitForSelector(".product-list")
    .result();

不要默认调用 .render(true)——先尝试不启用它。这会增加费用和延迟。

代理与地理定位

// Route through a country-specific proxy — no extra credit cost
String html = client.get("https://example.com/").countryCode("gb").result();

// Premium residential/mobile IPs — for sites that block datacenter proxies
// Cost: 10 credits (25 with render)
String html = client.get("https://hard-site.com/").premium(true).result();

// Ultra-premium — for the toughest anti-bot protections
// Cost: 30 credits (75 with render)
// Note: incompatible with custom headers
String html = client.get("https://hardest-site.com/").ultraPremium(true).result();

premium 和 ultraPremium 互斥——切勿同时链式调用。 升级顺序：标准（1 cr）→ 渲染（10 cr）→ premium（10 cr）→ ultraPremium（30 cr）。

会话（粘性代理）

// Reuse the same proxy IP across requests — useful for pagination and multi-step flows
// Sessions expire 15 minutes after last use; any integer is a valid ID
String page1 = client.get("https://example.com/page1").sessionNumber(42).result();
String page2 = client.get("https://example.com/page2").sessionNumber(42).result();

设备类型与自动解析

// Emulate a mobile browser user-agent
String html = client.get("https://example.com/").deviceType("mobile").result();

// Return structured JSON for supported sites (Amazon, Google, etc.)
String json = client.get("https://amazon.com/dp/B09V3KXJPB").autoparse(true).result();

重试

// Override the default retry count (default: 3)
// ScraperAPI retries failed requests for up to 70 seconds internally;
// .retry() controls how many times the SDK retries after a non-200 response
String html = client.get("https://flaky-site.com/").retry(5).result();

不要设置过短的超时时间——SDK 默认值经过校准，可留出 ScraperAPI 内部重试窗口（最长 70 秒）。将客户端超时设为 5 秒会导致误报失败。

升级阶梯

始终从最便宜的方案开始。只有在网站屏蔽了上一层级时才升级。

public static String scrapeWithEscalation(ScraperApiClient client, String url) throws Exception {
    // Try each tier in order — stop at the first success
    String[][] tiers = {
        {},                                  // 1 credit — standard
        {"render:true"},                     // 10 credits
        {"premium:true"},                    // 10 credits
        {"premium:true", "render:true"},     // 25 credits
        {"ultraPremium:true"},               // 30 credits
    };

    // Practical implementation — explicit tier cascade
    String[] attempts = { "standard", "render", "premium", "premiumRender", "ultraPremium" };
    for (String tier : attempts) {
        try {
            var req = client.get(url);
            switch (tier) {
                case "render":       req = req.render(true); break;
                case "premium":      req = req.premium(true); break;
                case "premiumRender": req = req.premium(true).render(true); break;
                case "ultraPremium": req = req.ultraPremium(true); break;
            }
            String html = req.result();
            if (html != null && html.toLowerCase().contains("<html")) return html;
        } catch (Exception e) {
            // Log and try next tier
        }
    }
    return null;
}

异步任务（用于批处理）

.result() 会阻塞调用线程。对于 20 个以上的 URL，请通过 REST 端点提交异步任务，并发收集结果。

import java.net.http.*;
import java.net.URI;
import com.fasterxml.jackson.databind.ObjectMapper;

private static final String API_KEY = System.getenv("SCRAPERAPI_API_KEY");
private static final HttpClient HTTP = HttpClient.newHttpClient();
private static final ObjectMapper JSON = new ObjectMapper();

public static Map<String, Object> submitJob(String url) throws Exception {
    String body = JSON.writeValueAsString(Map.of("apiKey", API_KEY, "url", url));
    HttpRequest req = HttpRequest.newBuilder()
        .uri(URI.create("https://async.scraperapi.com/jobs"))
        .POST(HttpRequest.BodyPublishers.ofString(body))
        .header("Content-Type", "application/json")
        .build();
    HttpResponse<String> resp = HTTP.send(req, HttpResponse.BodyHandlers.ofString());
    return JSON.readValue(resp.body(), Map.class); // {id, statusUrl}
}

public static String pollJob(Map<String, Object> job, int maxWaitSec) throws Exception {
    long deadline = System.currentTimeMillis() + maxWaitSec * 1000L;
    while (System.currentTimeMillis() < deadline) {
        HttpRequest req = HttpRequest.newBuilder()
            .uri(URI.create((String) job.get("statusUrl")))
            .GET().build();
        Map<String, Object> data = JSON.readValue(
            HTTP.send(req, HttpResponse.BodyHandlers.ofString()).body(), Map.class);
        if ("finished".equals(data.get("status")))
            return ((Map<?, ?>) data.get("response")).get("body").toString();
        if ("failed".equals(data.get("status")))
            throw new RuntimeException("Job " + job.get("id") + " failed");
        Thread.sleep(5_000);
    }
    throw new RuntimeException("Job " + job.get("id") + " timed out");
}

结构化数据端点

对于受支持的平台，请使用结构化端点，而非抓取原始 HTML。

public static String structuredGet(String vertical, Map<String, String> params) throws Exception {
    StringBuilder query = new StringBuilder("api_key=" + API_KEY);
    params.forEach((k, v) -> query.append("&").append(k).append("=").append(v));
    URI uri = URI.create("https://api.scraperapi.com/structured/" + vertical + "?" + query);
    HttpRequest req = HttpRequest.newBuilder().uri(uri).GET().build();
    HttpResponse<String> resp = HTTP.send(req, HttpResponse.BodyHandlers.ofString());
    if (resp.statusCode() != 200)
        throw new RuntimeException("Error " + resp.statusCode());
    return resp.body();
}

// Google SERP
String serp = structuredGet("google/search", Map.of("query", "java web scraping"));

// Amazon product
String product = structuredGet("amazon/product", Map.of("asin", "B09V3KXJPB"));

// Walmart search
String items = structuredGet("walmart/search", Map.of("query", "standing desk", "tld", "com"));

错误处理

public static String safeScrape(ScraperApiClient client, String url) {
    try {
        return client.get(url).retry(3).result();
    } catch (Exception e) {
        String msg = e.getMessage() != null ? e.getMessage() : "";
        if (msg.contains("401")) throw new RuntimeException("Invalid API key — check SCRAPERAPI_API_KEY", e);
        if (msg.contains("403")) throw new RuntimeException("Blocked or out of credits — try premium/ultraPremium", e);
        if (msg.contains("429")) throw new RuntimeException("Rate limit — reduce concurrency or use async", e);
        if (msg.contains("500") || msg.contains("503"))
            throw new RuntimeException("Transient error — retry with backoff", e);
        throw new RuntimeException("Scrape failed: " + url, e);
    }
}

状态代码：200 表示成功，401 表示密钥无效，403 表示被屏蔽／无可用额度，404 表示未找到目标，429 表示触发速率限制，500／503 表示暂时性故障（不扣费——可安全地采用退避策略重试）。

额度费用参考

请求类型	额度
标准 .result()	1
.render(true)	10
.premium(true)	10
.premium(true).render(true)	25
.ultraPremium(true)	30
.ultraPremium(true).render(true)	75

文档

• Java SDK 入门
• SDK 方法参考
• JavaScript 渲染
• 高级代理池
• 自动解析／JSON 响应
• 回调（异步）
• 控制面板与额度