spider_client/shapes/
request.rs

1use serde::{Deserialize, Serialize};
2use std::collections::HashMap;
3
4/// Structure representing the Chunking algorithm dictionary.
5#[derive(Debug, Deserialize, Serialize, Clone)]
6pub struct ChunkingAlgDict {
7    /// The chunking algorithm to use, defined as a specific type.
8    r#type: ChunkingType,
9    /// The amount to chunk by.
10    value: i32,
11}
12
13// The nested structures
14#[derive(Serialize, Deserialize, Debug, Clone, Default)]
15pub struct Timeout {
16    /// The seconds up to 60.
17    pub secs: u64,
18    /// The nanoseconds.
19    pub nanos: u32,
20}
21
22#[derive(Serialize, Deserialize, Debug, Clone)]
23pub struct IdleNetwork {
24    /// The timeout to wait until.
25    pub timeout: Timeout,
26}
27
28#[derive(Serialize, Deserialize, Debug, Clone)]
29#[serde(tag = "type", rename_all = "PascalCase")]
30pub enum WebAutomation {
31    Evaluate { code: String },
32    Click { selector: String },
33    Wait { duration: u64 },
34    WaitForNavigation,
35    WaitFor { selector: String },
36    WaitForAndClick { selector: String },
37    ScrollX { pixels: i32 },
38    ScrollY { pixels: i32 },
39    Fill { selector: String, value: String },
40    InfiniteScroll { times: u32 },
41}
42
43#[derive(Default, Serialize, Deserialize, Debug, Clone)]
44#[serde(tag = "type", rename_all = "PascalCase")]
45pub enum RedirectPolicy {
46    Loose,
47    #[default]
48    Strict,
49}
50
51pub type WebAutomationMap = std::collections::HashMap<String, Vec<WebAutomation>>;
52pub type ExecutionScriptsMap = std::collections::HashMap<String, String>;
53
54#[derive(Serialize, Deserialize, Debug, Clone)]
55pub struct Selector {
56    /// The timeout to wait until.
57    pub timeout: Timeout,
58    /// The selector to wait for.
59    pub selector: String,
60}
61
62#[derive(Serialize, Deserialize, Debug, Clone, Default)]
63pub struct Delay {
64    /// The timeout to wait until.
65    pub timeout: Timeout,
66}
67
68/// Default as true.
69fn default_some_true() -> Option<bool> {
70    Some(true)
71}
72
73#[derive(Serialize, Deserialize, Debug, Clone, Default)]
74pub struct WaitFor {
75    /// Wait until idle networks with a timeout of idleness.
76    pub idle_network: Option<IdleNetwork>,
77    /// Wait until a selector exist. Can determine if a selector exist after executing all js and network events.
78    pub selector: Option<Selector>,
79    /// Wait for the dom to update
80    pub dom: Option<Selector>,
81    /// Wait until a hard delay.
82    pub delay: Option<Delay>,
83    /// Wait until page navigation happen. Default is true.
84    #[serde(default = "default_some_true")]
85    pub page_navigations: Option<bool>,
86}
87
88/// Query request to get a document.
89#[derive(Serialize, Deserialize, Debug, Clone, Default)]
90pub struct QueryRequest {
91    /// The exact website url.
92    pub url: Option<String>,
93    /// The website domain.
94    pub domain: Option<String>,
95    /// The path of the resource.
96    pub pathname: Option<String>,
97}
98
99/// Enum representing different types of Chunking.
100#[derive(Default, Debug, Deserialize, Serialize, Clone)]
101#[serde(rename_all = "lowercase")]
102pub enum ChunkingType {
103    #[default]
104    /// By the word count.
105    ByWords,
106    /// By the line count.
107    ByLines,
108    /// By the char length.
109    ByCharacterLength,
110    /// By sentence.
111    BySentence,
112}
113
114#[derive(Default, Debug, Deserialize, Serialize, Clone)]
115/// View port handling for chrome.
116pub struct Viewport {
117    /// Device screen Width
118    pub width: u32,
119    /// Device screen size
120    pub height: u32,
121    /// Device scale factor
122    pub device_scale_factor: Option<f64>,
123    /// Emulating Mobile?
124    pub emulating_mobile: bool,
125    /// Use landscape mode instead of portrait.
126    pub is_landscape: bool,
127    /// Touch screen device?
128    pub has_touch: bool,
129}
130
131// Define the CSSSelector struct
132#[derive(Debug, Clone, Default, Deserialize, Serialize)]
133pub struct CSSSelector {
134    /// The name of the selector group
135    pub name: String,
136    /// A vector of CSS selectors
137    pub selectors: Vec<String>,
138}
139
140// Define the CSSExtractionMap type
141pub type CSSExtractionMap = HashMap<String, Vec<CSSSelector>>;
142
143/// Represents the settings for a webhook configuration
144#[derive(Debug, Default, Deserialize, Serialize, Clone)]
145pub struct WebhookSettings {
146    /// The destination where the webhook information will be sent
147    destination: String,
148    /// Trigger an action when all credits are depleted
149    on_credits_depleted: bool,
150    /// Trigger an action when half of the credits are depleted
151    on_credits_half_depleted: bool,
152    /// Trigger an action on a website status update event
153    on_website_status: bool,
154    /// Send information about a new page find (such as links and bytes)
155    on_find: bool,
156    /// Handle the metadata of a found page
157    on_find_metadata: bool,
158}
159
160/// Proxy pool selection for outbound request routing.
161/// Choose a pool based on your use case (e.g., stealth, speed, or stability).
162///
163/// - 'residential'         → cost-effective entry-level residential pool
164/// - 'residential_fast'    → faster residential pool for higher throughput
165/// - 'residential_static'  → static residential IPs, rotated daily
166/// - 'residential_premium' → low-latency premium IPs
167/// - 'residential_core'    → balanced plan (quality vs. cost)
168/// - 'residential_plus'    → largest and highest quality core pool
169/// - 'mobile'              → 4G/5G mobile proxies for maximum evasion
170/// - 'isp'                 → ISP-grade datacenters
171#[derive(Debug, Clone, Copy, Default, PartialEq, Eq, Serialize, Deserialize)]
172pub enum ProxyType {
173    /// Cost-effective entry-level residential pool.
174    #[serde(rename = "residential")]
175    Residential,
176    /// Higher-throughput residential pool for better performance.
177    #[serde(rename = "residential_fast")]
178    ResidentialFast,
179    /// Static residential IPs, rotated daily for session persistence.
180    #[serde(rename = "residential_static")]
181    ResidentialStatic,
182    /// 4G / 5G mobile proxies for maximum stealth and evasion.
183    #[serde(rename = "mobile")]
184    Mobile,
185    /// ISP-grade residential routing (alias: `datacenter`).
186    #[serde(rename = "isp", alias = "datacenter")]
187    #[default]
188    Isp,
189    /// Premium low-latency residential proxy pool.
190    #[serde(rename = "residential_premium")]
191    ResidentialPremium,
192    /// Core residential plan optimized for balance between cost and quality.
193    #[serde(rename = "residential_core")]
194    ResidentialCore,
195    /// Extended core residential pool with the largest, highest-quality IPs.
196    #[serde(rename = "residential_plus")]
197    ResidentialPlus,
198}
199
200/// List of proxies.
201pub const PROXY_TYPE_LIST: [ProxyType; 10] = [
202    ProxyType::ResidentialStatic,
203    ProxyType::Residential,
204    ProxyType::Isp,
205    ProxyType::Mobile,
206    ProxyType::ResidentialPremium,
207    ProxyType::ResidentialPlus,
208    ProxyType::ResidentialCore,
209    ProxyType::ResidentialFast,
210    ProxyType::ResidentialStatic,
211    ProxyType::Residential,
212];
213
214impl ProxyType {
215    /// Get the canonical string representation of the proxy type.
216    pub fn as_str(&self) -> &'static str {
217        match self {
218            ProxyType::Residential => "residential",
219            ProxyType::ResidentialFast => "residential_fast",
220            ProxyType::ResidentialStatic => "residential_static",
221            ProxyType::Mobile => "mobile",
222            ProxyType::Isp => "isp",
223            ProxyType::ResidentialPremium => "residential_premium",
224            ProxyType::ResidentialCore => "residential_core",
225            ProxyType::ResidentialPlus => "residential_plus",
226        }
227    }
228}
229
230/// Send multiple return formats.
231#[derive(Debug, Deserialize, Serialize, Clone)]
232#[serde(untagged)]
233pub enum ReturnFormatHandling {
234    /// A single return item.
235    Single(ReturnFormat),
236    /// Multiple return formats.
237    Multi(std::collections::HashSet<ReturnFormat>),
238}
239
240impl Default for ReturnFormatHandling {
241    fn default() -> ReturnFormatHandling {
242        ReturnFormatHandling::Single(ReturnFormat::Raw)
243    }
244}
245
246#[derive(Debug, Default, Deserialize, Serialize, Clone)]
247pub struct EventTracker {
248    /// The responses received.
249    responses: Option<bool>,
250    ///The request sent.
251    requests: Option<bool>,
252}
253
254/// Structure representing request parameters.
255#[derive(Debug, Default, Deserialize, Serialize, Clone)]
256pub struct RequestParams {
257    #[serde(default)]
258    /// The URL to be crawled.
259    pub url: Option<String>,
260    #[serde(default)]
261    /// The type of request to be made.
262    pub request: Option<RequestType>,
263    #[serde(default)]
264    /// The maximum number of pages the crawler should visit.
265    pub limit: Option<u32>,
266    #[serde(default)]
267    /// The format in which the result should be returned.
268    pub return_format: Option<ReturnFormatHandling>,
269    /// The country code for request
270    pub country_code: Option<String>,
271    #[serde(default)]
272    /// Specifies whether to only visit the top-level domain.
273    pub tld: Option<bool>,
274    #[serde(default)]
275    /// The depth of the crawl.
276    pub depth: Option<u32>,
277    #[serde(default)]
278    /// Specifies whether the request should be cached.
279    pub cache: Option<bool>,
280    #[serde(default)]
281    /// Perform an infinite scroll on the page as new content arises. The request param also needs to be set to 'chrome' or 'smart'.
282    pub scroll: Option<u32>,
283    #[serde(default)]
284    /// The budget for various resources.
285    pub budget: Option<HashMap<String, u32>>,
286    #[serde(default)]
287    /// The blacklist routes to ignore. This can be a Regex string pattern.
288    pub blacklist: Option<Vec<String>>,
289    #[serde(default)]
290    /// The whitelist routes to only crawl. This can be a Regex string pattern and used with black_listing.
291    pub whitelist: Option<Vec<String>>,
292    #[serde(default)]
293    /// The locale to be used during the crawl.
294    pub locale: Option<String>,
295    #[serde(default)]
296    /// The cookies to be set for the request, formatted as a single string.
297    pub cookies: Option<String>,
298    #[serde(default)]
299    /// Specifies whether to use stealth techniques to avoid detection.
300    pub stealth: Option<bool>,
301    #[serde(default)]
302    /// The headers to be used for the request.
303    pub headers: Option<HashMap<String, String>>,
304    #[serde(default)]
305    /// Specifies whether anti-bot measures should be used.
306    pub anti_bot: Option<bool>,
307    #[serde(default)]
308    /// Specifies whether to send data via webhooks.
309    pub webhooks: Option<WebhookSettings>,
310    #[serde(default)]
311    /// Specifies whether to include metadata in the response.
312    pub metadata: Option<bool>,
313    #[serde(default)]
314    /// The dimensions of the viewport.
315    pub viewport: Option<Viewport>,
316    #[serde(default)]
317    /// The encoding to be used for the request.
318    pub encoding: Option<String>,
319    #[serde(default)]
320    /// Specifies whether to include subdomains in the crawl.
321    pub subdomains: Option<bool>,
322    #[serde(default)]
323    /// The user agent string to be used for the request.
324    pub user_agent: Option<String>,
325    #[serde(default)]
326    /// Specifies whether the response data should be stored.
327    pub store_data: Option<bool>,
328    #[serde(default)]
329    /// Configuration settings for GPT (general purpose texture mappings).
330    pub gpt_config: Option<HashMap<String, String>>,
331    #[serde(default)]
332    /// Specifies whether to use fingerprinting protection.
333    pub fingerprint: Option<bool>,
334    #[serde(default)]
335    /// Specifies whether to perform the request without using storage.
336    pub storageless: Option<bool>,
337    #[serde(default)]
338    /// Specifies whether readability optimizations should be applied.
339    pub readability: Option<bool>,
340    #[serde(default)]
341    /// Specifies whether to use a proxy for the request. [Deprecated]: use the 'proxy' param instead.
342    pub proxy_enabled: Option<bool>,
343    #[serde(default)]
344    /// Specifies whether to respect the site's robots.txt file.
345    pub respect_robots: Option<bool>,
346    #[serde(default)]
347    /// CSS selector to be used to filter the content.
348    pub root_selector: Option<String>,
349    #[serde(default)]
350    /// Specifies whether to load all resources of the crawl target.
351    pub full_resources: Option<bool>,
352    #[serde(default)]
353    /// The text string to extract data from.
354    pub text: Option<String>,
355    #[serde(default)]
356    /// Specifies whether to use the sitemap links.
357    pub sitemap: Option<bool>,
358    #[serde(default)]
359    /// External domains to include the crawl.
360    pub external_domains: Option<Vec<String>>,
361    #[serde(default)]
362    /// Returns the OpenAI embeddings for the title and description. Other values, such as keywords, may also be included. Requires the `metadata` parameter to be set to `true`.
363    pub return_embeddings: Option<bool>,
364    #[serde(default)]
365    /// Returns the HTTP response headers.
366    pub return_headers: Option<bool>,
367    #[serde(default)]
368    /// Returns the link(s) found on the page that match the crawler query.
369    pub return_page_links: Option<bool>,
370    #[serde(default)]
371    /// Returns the HTTP response cookies.
372    pub return_cookies: Option<bool>,
373    #[serde(default)]
374    /// The timeout for the request, in seconds.
375    pub request_timeout: Option<u8>,
376    #[serde(default)]
377    /// Specifies whether to run the request in the background.
378    pub run_in_background: Option<bool>,
379    #[serde(default)]
380    /// Specifies whether to skip configuration checks.
381    pub skip_config_checks: Option<bool>,
382    #[serde(default)]
383    /// Use CSS query selectors to scrape contents from the web page. Set the paths and the CSS extraction object map to perform extractions per path or page.
384    pub css_extraction_map: Option<CSSExtractionMap>,
385    #[serde(default)]
386    /// The chunking algorithm to use.
387    pub chunking_alg: Option<ChunkingAlgDict>,
388    #[serde(default)]
389    /// Disable request interception when running 'request' as 'chrome' or 'smart'. This can help when the page uses 3rd party or external scripts to load content.
390    pub disable_intercept: Option<bool>,
391    #[serde(default)]
392    /// The wait for events on the page. You need to make your `request` `chrome` or `smart`.
393    pub wait_for: Option<WaitFor>,
394    #[serde(default)]
395    /// Perform custom Javascript tasks on a url or url path. You need to make your `request` `chrome` or `smart`
396    pub execution_scripts: Option<ExecutionScriptsMap>,
397    #[serde(default)]
398    /// Perform web automated tasks on a url or url path. You need to make your `request` `chrome` or `smart`
399    pub automation_scripts: Option<WebAutomationMap>,
400    #[serde(default)]
401    /// The redirect policy for HTTP request. Set the value to Loose to allow all.
402    pub redirect_policy: Option<RedirectPolicy>,
403    #[serde(default)]
404    /// Track the request sent and responses received for `chrome` or `smart`. The responses will track the bytes used and the requests will have the monotime sent.
405    pub event_tracker: Option<EventTracker>,
406    #[serde(default)]
407    /// The timeout to stop the crawl.
408    pub crawl_timeout: Option<Timeout>,
409    #[serde(default)]
410    /// Evaluates given script in every frame upon creation (before loading frame's scripts).
411    pub evaluate_on_new_document: Option<Box<String>>,
412    #[serde(default)]
413    /// Runs the request using lite_mode:Lite mode reduces data transfer costs by 70%, with trade-offs in speed, accuracy,
414    /// geo-targeting, and reliability. It’s best suited for non-urgent data collection or when
415    /// targeting websites with minimal anti-bot protections.
416    pub lite_mode: Option<bool>,
417    #[serde(default)]
418    /// The proxy to use for request.
419    pub proxy: Option<ProxyType>,
420    #[serde(default)]
421    /// Use a remote proxy at ~70% reduced cost for file downloads.
422    /// This requires a user-supplied static IP proxy endpoint.
423    pub remote_proxy: Option<String>,
424    #[serde(default)]
425    /// Set the maximum number of credits to use per page.
426    /// Credits are measured in decimal units, where 10,000 credits equal one dollar (100 credits per penny).
427    /// Credit limiting only applies to request that are Javascript rendered using smart_mode or chrome for the 'request' type.
428    pub max_credits_per_page: Option<f64>,
429}
430
431/// The structure representing request parameters for a search request.
432#[derive(Debug, Default, Deserialize, Serialize, Clone)]
433pub struct SearchRequestParams {
434    /// The base request parameters.
435    #[serde(default, flatten)]
436    pub base: RequestParams,
437    // The search request.
438    pub search: String,
439    /// The search limit.
440    pub search_limit: Option<u32>,
441    // Fetch the page content. Defaults to true.
442    pub fetch_page_content: Option<bool>,
443    /// The search location of the request
444    pub location: Option<String>,
445    /// The country code of the request
446    pub country: Option<String>,
447    /// The language code of the request.
448    pub language: Option<String>,
449    /// The number of search results
450    pub num: Option<u32>,
451    /// The page of the search results.
452    pub page: Option<u32>,
453    #[serde(default)]
454    /// The websites limit if a list is sent from text or urls comma split. This helps automatic configuration of the system.
455    pub website_limit: Option<u32>,
456}
457
458/// Structure representing request parameters for transforming files.
459#[derive(Debug, Default, Deserialize, Serialize, Clone)]
460pub struct TransformParams {
461    #[serde(default)]
462    /// The format in which the result should be returned.
463    pub return_format: Option<ReturnFormat>,
464    #[serde(default)]
465    /// Specifies whether readability optimizations should be applied.
466    pub readability: Option<bool>,
467    #[serde(default)]
468    /// Clean the markdown or text for AI.
469    pub clean: Option<bool>,
470    #[serde(default)]
471    /// Clean the markdown or text for AI removing footers, navigation, and more.
472    pub clean_full: Option<bool>,
473    /// The data being transformed.
474    pub data: Vec<DataParam>,
475}
476
477#[derive(Serialize, Deserialize, Debug, Clone)]
478pub struct DataParam {
479    /// The HTML resource.
480    pub html: String,
481    /// The website url.
482    pub url: Option<String>,
483}
484
485/// the request type to perform
486#[derive(Debug, Default, Clone, Serialize, Deserialize, PartialEq, Eq)]
487#[serde(rename_all = "lowercase")]
488pub enum RequestType {
489    /// Default HTTP request
490    Http,
491    /// Chrome browser rendering
492    Chrome,
493    #[default]
494    /// Smart mode defaulting to HTTP and using Chrome when needed.
495    SmartMode,
496}
497
498/// Enum representing different return formats.
499#[derive(Default, Debug, Deserialize, Serialize, Clone, PartialEq, Eq, Hash)]
500#[serde(rename_all = "lowercase")]
501pub enum ReturnFormat {
502    #[default]
503    /// The default return format of the resource.
504    Raw,
505    /// Return the response as Markdown.
506    Markdown,
507    /// Return the response as Commonmark.
508    Commonmark,
509    /// Return the response as Html2text.
510    Html2text,
511    /// Return the response as Text.
512    Text,
513    /// Returns a screenshot as Base64Url
514    Screenshot,
515    /// Return the response as XML.
516    Xml,
517    /// Return the response as Bytes.
518    Bytes,
519}
spider_client/shapes/request.rs

spider_client/shapes/
request.rs
