crawlrs 0.2.0

crawlrs - Enterprise-grade web scraping and crawling platform built with Rust
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
// Copyright (c) 2025-2026 Kirky.X🌠
// SPDX-License-Identifier: Apache-2.0

use crate::engines::engine_client::{
    EngineClient, ScrapeOptions, ScrapeRequest as EngineScrapeRequest,
};
use crate::search::{
    engine_trait::SearchEngine,
    error::SearchError,
    response::{Response, ResponseItem},
    types::{EngineHealth, SearchEngineType},
    SearchRequest,
};
use async_trait::async_trait;
use scraper::Html;
use std::collections::HashMap;
use std::sync::Arc;

use super::shared_utils::{escape_html_text, safe_parse_selector};

/// Sogou Search Engine implementation with EngineClient support
pub struct SogouSearchEngine {
    engine_client: Arc<EngineClient>,
}

impl SogouSearchEngine {
    pub fn new(engine_client: Arc<EngineClient>) -> Self {
        Self { engine_client }
    }

    /// 解析并补全搜狗搜索结果中的URL
    /// 处理相对路径和中转URL格式
    pub fn resolve_url(&self, url: &str) -> String {
        if url.is_empty() {
            return String::new();
        }

        // 处理直接完整URL (仅允许 http/https)
        if url.starts_with("http://") || url.starts_with("https://") {
            return Self::validated_absolute_url(url);
        }

        // 处理搜狗中转链接: /link?url=...
        if url.starts_with("/link?url=") {
            // 提取参数中的URL并解码,递归验证解码后的 URL
            return match Self::decode_sogou_redirect(url) {
                Some(decoded) => self.resolve_url(decoded.as_ref()),
                None => String::new(),
            };
        }

        // 处理其他相对路径 (仅搜狗域名)
        if url.starts_with("/") {
            return format!("https://www.sogou.com{}", url);
        }

        // 其他情况拒绝
        String::new()
    }

    /// 完整 URL 的协议校验:仅接受 http/https,其余返回空串
    fn validated_absolute_url(url: &str) -> String {
        if let Ok(parsed) = url::Url::parse(url) {
            // 只允许 http/https 协议
            if parsed.scheme() == "http" || parsed.scheme() == "https" {
                return url.to_string();
            }
        }
        String::new()
    }

    /// 去掉搜狗中转前缀并 URL 解码;解码失败返回 None
    fn decode_sogou_redirect(url: &str) -> Option<std::borrow::Cow<'_, str>> {
        let encoded_url = url.trim_start_matches("/link?url=");
        // URL解码
        urlencoding::decode(encoded_url).ok()
    }

    pub fn parse_search_results(
        &self,
        html_content: &str,
    ) -> Result<Vec<ResponseItem>, SearchError> {
        let document = Html::parse_document(html_content);
        let selectors = SogouSelectors::new();
        let mut results = Vec::with_capacity(10);

        for element in document.select(&selectors.result) {
            if let Some(item) = self.parse_result_element(element, &selectors) {
                results.push(item);
            }
        }

        Ok(results)
    }

    /// 解析单个结果容器:广告/推荐容器、缺标题、缺链接、空 URL 均跳过
    /// (`None`),成功时产出补齐标题/URL/摘要的结果项。
    fn parse_result_element(
        &self,
        element: scraper::ElementRef<'_>,
        selectors: &SogouSelectors,
    ) -> Option<ResponseItem> {
        // 过滤特殊内容容器(广告/推荐等),参考 SearXNG XPath:
        // `vrwrap and not(.//div[contains(@class, "special-wrap")])`
        if element.select(&selectors.special_wrap).next().is_some() {
            return None;
        }

        // 提取标题 - 获取纯文本并清理空白
        let title_node = element.select(&selectors.title).next()?;
        let raw_title = title_node.text().collect::<String>();
        let title = escape_html_text(raw_title.trim());

        if title.is_empty() {
            return None;
        }

        // 提取链接
        let url_node = element.select(&selectors.link).next()?;
        let raw_url = url_node.value().attr("href").unwrap_or("").to_string();

        // 搜狗中转链接 `/link?url=XXX` 里的 XXX 是内部 ID(非 URL 编码的真实 URL),
        // 无法直接解码。真实 URL 在同一 vrwrap 块的 `data-url` 属性里。
        // 参考 searxng/searx/engines/sogou.py 的 `_extract_url`。
        let resolved_url = if raw_url.starts_with("/link?url=") {
            element
                .select(&selectors.data_url)
                .next()
                .and_then(|n| n.value().attr("data-url"))
                .map(|s| s.to_string())
                .filter(|s| !s.is_empty())
                .unwrap_or_default()
        } else {
            self.resolve_url(&raw_url)
        };

        if resolved_url.is_empty() {
            return None;
        }

        // 提取摘要 - 按选择器优先级回退,找到第一个非空文本
        let description = element
            .select(&selectors.snippet)
            .next()
            .map(|node| escape_html_text(&node.text().collect::<String>()))
            .unwrap_or_default();

        Some(ResponseItem {
            title,
            url: resolved_url,
            description,
            engine: SearchEngineType::Sogou,
        })
    }
}

/// Sogou 结果解析所需的 CSS 选择器集合(构造失败即 panic,与原就地解析语义一致)
struct SogouSelectors {
    result: scraper::Selector,
    title: scraper::Selector,
    link: scraper::Selector,
    data_url: scraper::Selector,
    snippet: scraper::Selector,
    special_wrap: scraper::Selector,
}

impl SogouSelectors {
    fn new() -> Self {
        Self {
            result: safe_parse_selector(".vrwrap, .rb")
                .expect("Failed to parse Sogou result selector"),
            title: safe_parse_selector("h3").expect("Failed to parse Sogou title selector"),
            link: safe_parse_selector("h3 a").expect("Failed to parse Sogou link selector"),
            // 搜狗中转链接的真实 URL 存在 data-url 属性里(参考 searxng _extract_url)。
            data_url: safe_parse_selector("[data-url]")
                .expect("Failed to parse Sogou data-url selector"),
            // 摘要选择器(按优先级回退):text-layout > ft > 裸 p
            // 保持原 smart 端 parse_sogou_results 行为(等价)。
            snippet: safe_parse_selector("div.text-layout p, div.ft p, p")
                .expect("Failed to parse Sogou snippet selector"),
            special_wrap: safe_parse_selector("div.special-wrap").expect("special-wrap selector"),
        }
    }
}

#[async_trait]
impl SearchEngine for SogouSearchEngine {
    fn name(&self) -> &'static str {
        "Sogou"
    }
    fn engine_type(&self) -> SearchEngineType {
        SearchEngineType::Sogou
    }
    fn health(&self) -> EngineHealth {
        EngineHealth::Healthy
    }

    async fn search(&self, request: &SearchRequest) -> Result<Response<ResponseItem>, SearchError> {
        // 测试固定结果仅存在于测试构建(CWE-489:生产路径不得被环境变量旁路)
        #[cfg(test)]
        {
            if let Some(resp) = sogou_test_results_override(&request.query) {
                return Ok(resp);
            }
        }

        let engine_request = build_engine_request(request);

        let scrape_response = self
            .engine_client
            .scrape(&engine_request)
            .await
            .map_err(|e| SearchError::EngineClient("Sogou".to_string(), e.to_string()))?;

        if scrape_response.status_code < 200 || scrape_response.status_code >= 300 {
            return Err(SearchError::BadHttpStatus(
                "Sogou".to_string(),
                scrape_response.status_code,
            ));
        }

        let html_content = scrape_response.content;

        if is_antispider_page(&html_content) {
            return Err(SearchError::Captcha("Sogou".to_string()));
        }
        let items = self.parse_search_results(&html_content)?;
        let total_count = items.len() as u64;

        Ok(Response {
            items,
            total_results: Some(total_count),
            engine: SearchEngineType::Sogou,
        })
    }
}

/// offset/limit 推导 1-based 页码(limit 0 防除零取 1)
fn page_from_offset(offset: u32, limit: u32) -> u32 {
    if offset > 0 {
        (offset / limit.max(1)) + 1
    } else {
        1
    }
}

/// 构建请求头(参考 SearXNG + 完整浏览器指纹)
fn build_search_headers() -> HashMap<String, String> {
    let mut headers = HashMap::with_capacity(6);
    headers.insert(
        "Accept".to_string(),
        "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8".to_string(),
    );
    headers.insert(
        "Accept-Language".to_string(),
        "zh-CN,zh;q=0.9,en;q=0.8".to_string(),
    );
    headers.insert("DNT".to_string(), "1".to_string());
    headers.insert("Connection".to_string(), "keep-alive".to_string());
    headers.insert("Sec-GPC".to_string(), "1".to_string());
    headers.insert("Cache-Control".to_string(), "max-age=0".to_string());
    headers
}

/// 组装 EngineClient 请求:query 编码 + 分页参数
fn build_engine_request(request: &SearchRequest) -> EngineScrapeRequest {
    let base_url = "https://www.sogou.com/web";

    // 分页支持(参考 SearXNG sogou.py `page` 参数)
    let page = page_from_offset(request.offset, request.limit);

    let full_url = format!(
        "{}?query={}&page={}",
        base_url,
        urlencoding::encode(&request.query),
        page,
    );

    EngineScrapeRequest {
        url: full_url,
        options: ScrapeOptions {
            headers: build_search_headers(),
            timeout: std::time::Duration::from_secs(30),
            ..Default::default()
        },
    }
}

/// 反爬检测:Sogou 302 重定向到 antispider 页面
/// (参考 SearXNG sogou.py response:resp.next_request.url starts with antispider)
/// 由于 EngineClient 自动跟随重定向,最终 content 会是 antispider 页面内容
fn is_antispider_page(html_content: &str) -> bool {
    html_content.contains("www.sogou.com/antispider")
        || html_content.contains("antispider") && html_content.contains("验证")
}

/// 仅测试构建可用的固定结果旁路:`SOGOU_TEST_RESULTS=true` 时返回确定性结果。
///
/// 生产构建中该函数与 `search()` 内的调用点一并被 `#[cfg(test)]` 剔除,
/// 环境变量无法影响线上搜索行为(CWE-489)。
#[cfg(test)]
fn sogou_test_results_override(query: &str) -> Option<Response<ResponseItem>> {
    if std::env::var("SOGOU_TEST_RESULTS").unwrap_or_default() != "true" {
        return None;
    }
    let escaped_query = html_escape::encode_text(query);
    Some(Response {
        items: vec![
            ResponseItem {
                title: format!("Test Result 1 for {}", escaped_query),
                url: "https://sogou.com/1".to_string(),
                description: "Test description 1".to_string(),
                engine: SearchEngineType::Sogou,
            },
            ResponseItem {
                title: format!("Test Result 2 for {}", escaped_query),
                url: "https://sogou.com/2".to_string(),
                description: "Test description 2".to_string(),
                engine: SearchEngineType::Sogou,
            },
        ],
        total_results: Some(2),
        engine: SearchEngineType::Sogou,
    })
}

#[cfg(test)]
mod tests {
    use super::*;

    // 创建测试用 SogouSearchEngine
    fn make_engine() -> SogouSearchEngine {
        SogouSearchEngine::new(Arc::new(EngineClient::new()))
    }

    // safe_parse_selector 已统一到 shared_utils,测试见 shared_utils::tests

    // resolve_url 测试

    #[test]
    fn test_resolve_url_empty() {
        let engine = make_engine();
        assert_eq!(engine.resolve_url(""), "");
    }

    #[test]
    fn test_resolve_url_https() {
        let engine = make_engine();
        let url = "https://example.com/page";
        assert_eq!(engine.resolve_url(url), url);
    }

    #[test]
    fn test_resolve_url_http() {
        let engine = make_engine();
        let url = "http://example.com/page";
        assert_eq!(engine.resolve_url(url), url);
    }

    #[test]
    fn test_resolve_url_invalid_scheme_rejected() {
        let engine = make_engine();
        // ftp:// 开头的 URL 虽然以 http 开头匹配但会解析失败,返回空
        assert_eq!(engine.resolve_url("ftp://example.com"), "");
    }

    #[test]
    fn test_resolve_url_malformed_http_rejected() {
        let engine = make_engine();
        // "http://" 开头但无法解析的 URL 应返回空
        assert_eq!(engine.resolve_url("http://"), "");
    }

    #[test]
    fn test_resolve_url_sogou_link_redirect() {
        let engine = make_engine();
        // 搜狗中转链接,包含完整 URL
        let url = "/link?url=https%3A%2F%2Fwww.example.com%2Fpage";
        let resolved = engine.resolve_url(url);
        assert_eq!(resolved, "https://www.example.com/page");
    }

    #[test]
    fn test_resolve_url_relative_path() {
        let engine = make_engine();
        let url = "/web?query=test";
        let resolved = engine.resolve_url(url);
        assert_eq!(resolved, "https://www.sogou.com/web?query=test");
    }

    #[test]
    fn test_resolve_url_non_http_non_relative_rejected() {
        let engine = make_engine();
        // 既不是 http/https 开头,也不是 / 开头,应返回空
        assert_eq!(engine.resolve_url("just-some-text"), "");
    }

    // parse_search_results 测试

    #[test]
    fn test_parse_search_results_valid_html() {
        let engine = make_engine();
        let html = r#"
        <div class="vrwrap">
            <h3><a href="https://example.com/result1">Example Result 1</a></h3>
        </div>
        <div class="vrwrap">
            <h3><a href="https://example.com/result2">Example Result 2</a></h3>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert_eq!(results.len(), 2);
        assert_eq!(results[0].title, "Example Result 1");
        assert_eq!(results[0].url, "https://example.com/result1");
        assert_eq!(results[0].engine, SearchEngineType::Sogou);
        assert_eq!(results[1].title, "Example Result 2");
    }

    #[test]
    fn test_parse_search_results_rb_class() {
        let engine = make_engine();
        // .rb 也是一个结果容器选择器
        let html = r#"
        <div class="rb">
            <h3><a href="https://example.com/rb-result">RB Result</a></h3>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert_eq!(results.len(), 1);
        assert_eq!(results[0].title, "RB Result");
    }

    #[test]
    fn test_parse_search_results_empty_html() {
        let engine = make_engine();
        let results = engine.parse_search_results("").unwrap();
        assert!(results.is_empty());
    }

    #[test]
    fn test_parse_search_results_no_results() {
        let engine = make_engine();
        let html = "<html><body><p>No results here</p></body></html>";
        let results = engine.parse_search_results(html).unwrap();
        assert!(results.is_empty());
    }

    #[test]
    fn test_parse_search_results_missing_title_skipped() {
        let engine = make_engine();
        // 没有 h3 标题的 vrwrap 应被跳过
        let html = r#"
        <div class="vrwrap">
            <a href="https://example.com/no-title">No Title</a>
        </div>
        <div class="vrwrap">
            <h3><a href="https://example.com/with-title">With Title</a></h3>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert_eq!(results.len(), 1);
        assert_eq!(results[0].title, "With Title");
    }

    #[test]
    fn test_parse_search_results_empty_title_skipped() {
        let engine = make_engine();
        // 空标题应被跳过
        let html = r#"
        <div class="vrwrap">
            <h3>   </h3>
            <a href="https://example.com/empty-title">Empty Title</a>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert!(results.is_empty());
    }

    #[test]
    fn test_parse_search_results_sogou_redirect_link() {
        // 真实搜狗 HTML 结构:href 是 `/link?url=内部ID`(非 URL 编码的真实 URL),
        // 真实 URL 在同一 vrwrap 块的 `data-url` 属性里。
        // 参考 searxng/searx/engines/sogou.py 的 `_extract_url`。
        let engine = make_engine();
        let html = r#"
        <div class="vrwrap">
            <h3><a href="/link?url=hedJjaC291N0Yt4ikBP2zHNFN13gvLUJQ3i3ukpLokg.">Redirect Result</a></h3>
            <div class="img-layout" data-url="https://www.example.com/article"></div>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert_eq!(results.len(), 1);
        assert_eq!(results[0].url, "https://www.example.com/article");
        assert_eq!(results[0].title, "Redirect Result");
    }

    #[test]
    fn test_parse_search_results_sogou_redirect_link_without_data_url_skipped() {
        // 边界情况:`/link?url=` 中转链接但 vrwrap 块内没有 data-url 属性时,
        // 无法获取真实 URL,应跳过该结果(而非返回中转链接)。
        let engine = make_engine();
        let html = r#"
        <div class="vrwrap">
            <h3><a href="/link?url=hedJjaC291InternalIdNoDataUrl">No Data URL</a></h3>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert!(
            results.is_empty(),
            "result without data-url should be skipped"
        );
    }

    #[test]
    fn test_parse_search_results_mixed_redirect_and_direct_urls() {
        // 混合场景:同一页面既有中转链接(link?url=)又有直接 URL,
        // 中转链接走 data-url 提取,直接 URL 走 resolve_url。
        let engine = make_engine();
        let html = r#"
        <div class="vrwrap">
            <h3><a href="/link?url=internalId1">Redirect Result</a></h3>
            <div data-url="https://www.rust-lang.org/"></div>
        </div>
        <div class="vrwrap">
            <h3><a href="https://baike.sogou.com/v12345.htm">Direct URL Result</a></h3>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert_eq!(results.len(), 2);
        assert_eq!(results[0].url, "https://www.rust-lang.org/");
        assert_eq!(results[1].url, "https://baike.sogou.com/v12345.htm");
    }

    #[test]
    fn test_parse_search_results_empty_url_skipped() {
        let engine = make_engine();
        // href 为空或 resolve_url 返回空时应跳过
        let html = r#"
        <div class="vrwrap">
            <h3><a>No href</a></h3>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert!(results.is_empty());
    }

    #[test]
    fn test_parse_search_results_description_is_empty() {
        let engine = make_engine();
        let html = r#"
        <div class="vrwrap">
            <h3><a href="https://example.com/test">Test</a></h3>
        </div>
        "#;
        let results = engine.parse_search_results(html).unwrap();
        assert_eq!(results.len(), 1);
        assert_eq!(results[0].description, "");
    }

    // SearchEngine trait 方法测试

    #[test]
    fn test_name() {
        let engine = make_engine();
        assert_eq!(engine.name(), "Sogou");
    }

    #[test]
    fn test_engine_type() {
        let engine = make_engine();
        assert_eq!(engine.engine_type(), SearchEngineType::Sogou);
    }

    #[test]
    fn test_health() {
        let engine = make_engine();
        assert_eq!(engine.health(), EngineHealth::Healthy);
    }

    // search test fallback path

    /// Mutex to serialize tests that mutate process-level environment
    /// variables (std::env::set_var is not thread-safe across tests).
    static ENV_LOCK: std::sync::Mutex<()> = std::sync::Mutex::new(());

    fn make_search_request(query: &str) -> SearchRequest {
        SearchRequest::new(query)
    }

    // ENV_LOCK must be held across .await because engine.search reads
    // SOGOU_TEST_RESULTS during execution; releasing it would let other tests
    // race-modify the env var. Single-threaded tokio runtime => no deadlock risk.
    #[allow(clippy::await_holding_lock)]
    #[tokio::test]
    async fn test_search_with_test_results_env() {
        let _lock = ENV_LOCK.lock().unwrap();
        std::env::set_var("SOGOU_TEST_RESULTS", "true");
        let engine = make_engine();
        let request = make_search_request("test query");
        let result = engine.search(&request).await;
        std::env::remove_var("SOGOU_TEST_RESULTS");

        assert!(result.is_ok());
        let response = result.unwrap();
        assert_eq!(response.items.len(), 2);
        assert!(response.items[0].title.contains("test query"));
        assert_eq!(response.engine, SearchEngineType::Sogou);
        assert_eq!(response.total_results, Some(2));
    }

    #[allow(clippy::await_holding_lock)] // ENV_LOCK serializes env var access; see above
    #[tokio::test]
    async fn test_search_fallback_returns_correct_urls_and_engine() {
        let _lock = ENV_LOCK.lock().unwrap();
        std::env::set_var("SOGOU_TEST_RESULTS", "true");

        let engine = make_engine();
        let request = make_search_request("url check");

        let response = engine.search(&request).await;

        std::env::remove_var("SOGOU_TEST_RESULTS");

        let response = response.expect("fallback should return Ok");
        assert_eq!(response.items.len(), 2);
        assert_eq!(response.items[0].url, "https://sogou.com/1");
        assert_eq!(response.items[1].url, "https://sogou.com/2");
        assert_eq!(response.items[0].engine, SearchEngineType::Sogou);
        assert_eq!(response.items[1].engine, SearchEngineType::Sogou);
        assert_eq!(response.items[0].description, "Test description 1");
        assert_eq!(response.items[1].description, "Test description 2");
    }

    #[allow(clippy::await_holding_lock)] // ENV_LOCK serializes env var access; see above
    #[tokio::test]
    async fn test_search_fallback_escapes_query_in_title() {
        let _lock = ENV_LOCK.lock().unwrap();
        std::env::set_var("SOGOU_TEST_RESULTS", "true");

        let engine = make_engine();
        // Query with HTML special characters that should be escaped
        let request = make_search_request("<script>alert(1)</script>");

        let response = engine.search(&request).await;

        std::env::remove_var("SOGOU_TEST_RESULTS");

        let response = response.expect("fallback should return Ok");
        // The title should contain the escaped query, not raw HTML
        assert!(response.items[0].title.contains("&lt;script&gt;"));
        assert!(!response.items[0].title.contains("<script>"));
        assert!(response.items[1].title.contains("&lt;script&gt;"));
        assert!(!response.items[1].title.contains("<script>"));
    }
}