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
200impl ProxyType {
201    /// Get the canonical string representation of the proxy type.
202    pub fn as_str(&self) -> &'static str {
203        match self {
204            ProxyType::Residential => "residential",
205            ProxyType::ResidentialFast => "residential_fast",
206            ProxyType::ResidentialStatic => "residential_static",
207            ProxyType::Mobile => "mobile",
208            ProxyType::Isp => "isp",
209            ProxyType::ResidentialPremium => "residential_premium",
210            ProxyType::ResidentialCore => "residential_core",
211            ProxyType::ResidentialPlus => "residential_plus",
212        }
213    }
214}
215
216/// Send multiple return formats.
217#[derive(Debug, Deserialize, Serialize, Clone)]
218#[serde(untagged)]
219pub enum ReturnFormatHandling {
220    /// A single return item.
221    Single(ReturnFormat),
222    /// Multiple return formats.
223    Multi(std::collections::HashSet<ReturnFormat>),
224}
225
226impl Default for ReturnFormatHandling {
227    fn default() -> ReturnFormatHandling {
228        ReturnFormatHandling::Single(ReturnFormat::Raw)
229    }
230}
231
232#[derive(Debug, Default, Deserialize, Serialize, Clone)]
233pub struct EventTracker {
234    /// The responses received.
235    responses: Option<bool>,
236    ///The request sent.
237    requests: Option<bool>,
238}
239
240/// Structure representing request parameters.
241#[derive(Debug, Default, Deserialize, Serialize, Clone)]
242pub struct RequestParams {
243    #[serde(default)]
244    /// The URL to be crawled.
245    pub url: Option<String>,
246    #[serde(default)]
247    /// The type of request to be made.
248    pub request: Option<RequestType>,
249    #[serde(default)]
250    /// The maximum number of pages the crawler should visit.
251    pub limit: Option<u32>,
252    #[serde(default)]
253    /// The format in which the result should be returned.
254    pub return_format: Option<ReturnFormatHandling>,
255    /// The country code for request
256    pub country_code: Option<String>,
257    #[serde(default)]
258    /// Specifies whether to only visit the top-level domain.
259    pub tld: Option<bool>,
260    #[serde(default)]
261    /// The depth of the crawl.
262    pub depth: Option<u32>,
263    #[serde(default)]
264    /// Specifies whether the request should be cached.
265    pub cache: Option<bool>,
266    #[serde(default)]
267    /// Perform an infinite scroll on the page as new content arises. The request param also needs to be set to 'chrome' or 'smart'.
268    pub scroll: Option<u32>,
269    #[serde(default)]
270    /// The budget for various resources.
271    pub budget: Option<HashMap<String, u32>>,
272    #[serde(default)]
273    /// The blacklist routes to ignore. This can be a Regex string pattern.
274    pub blacklist: Option<Vec<String>>,
275    #[serde(default)]
276    /// The whitelist routes to only crawl. This can be a Regex string pattern and used with black_listing.
277    pub whitelist: Option<Vec<String>>,
278    #[serde(default)]
279    /// The locale to be used during the crawl.
280    pub locale: Option<String>,
281    #[serde(default)]
282    /// The cookies to be set for the request, formatted as a single string.
283    pub cookies: Option<String>,
284    #[serde(default)]
285    /// Specifies whether to use stealth techniques to avoid detection.
286    pub stealth: Option<bool>,
287    #[serde(default)]
288    /// The headers to be used for the request.
289    pub headers: Option<HashMap<String, String>>,
290    #[serde(default)]
291    /// Specifies whether anti-bot measures should be used.
292    pub anti_bot: Option<bool>,
293    #[serde(default)]
294    /// Specifies whether to send data via webhooks.
295    pub webhooks: Option<WebhookSettings>,
296    #[serde(default)]
297    /// Specifies whether to include metadata in the response.
298    pub metadata: Option<bool>,
299    #[serde(default)]
300    /// The dimensions of the viewport.
301    pub viewport: Option<Viewport>,
302    #[serde(default)]
303    /// The encoding to be used for the request.
304    pub encoding: Option<String>,
305    #[serde(default)]
306    /// Specifies whether to include subdomains in the crawl.
307    pub subdomains: Option<bool>,
308    #[serde(default)]
309    /// The user agent string to be used for the request.
310    pub user_agent: Option<String>,
311    #[serde(default)]
312    /// Specifies whether the response data should be stored.
313    pub store_data: Option<bool>,
314    #[serde(default)]
315    /// Configuration settings for GPT (general purpose texture mappings).
316    pub gpt_config: Option<HashMap<String, String>>,
317    #[serde(default)]
318    /// Specifies whether to use fingerprinting protection.
319    pub fingerprint: Option<bool>,
320    #[serde(default)]
321    /// Specifies whether to perform the request without using storage.
322    pub storageless: Option<bool>,
323    #[serde(default)]
324    /// Specifies whether readability optimizations should be applied.
325    pub readability: Option<bool>,
326    #[serde(default)]
327    /// Specifies whether to use a proxy for the request. [Deprecated]: use the 'proxy' param instead.
328    pub proxy_enabled: Option<bool>,
329    #[serde(default)]
330    /// Specifies whether to respect the site's robots.txt file.
331    pub respect_robots: Option<bool>,
332    #[serde(default)]
333    /// CSS selector to be used to filter the content.
334    pub root_selector: Option<String>,
335    #[serde(default)]
336    /// Specifies whether to load all resources of the crawl target.
337    pub full_resources: Option<bool>,
338    #[serde(default)]
339    /// The text string to extract data from.
340    pub text: Option<String>,
341    #[serde(default)]
342    /// Specifies whether to use the sitemap links.
343    pub sitemap: Option<bool>,
344    #[serde(default)]
345    /// External domains to include the crawl.
346    pub external_domains: Option<Vec<String>>,
347    #[serde(default)]
348    /// 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`.
349    pub return_embeddings: Option<bool>,
350    #[serde(default)]
351    /// Returns the HTTP response headers.
352    pub return_headers: Option<bool>,
353    #[serde(default)]
354    /// Returns the link(s) found on the page that match the crawler query.
355    pub return_page_links: Option<bool>,
356    #[serde(default)]
357    /// Returns the HTTP response cookies.
358    pub return_cookies: Option<bool>,
359    #[serde(default)]
360    /// The timeout for the request, in seconds.
361    pub request_timeout: Option<u8>,
362    #[serde(default)]
363    /// Specifies whether to run the request in the background.
364    pub run_in_background: Option<bool>,
365    #[serde(default)]
366    /// Specifies whether to skip configuration checks.
367    pub skip_config_checks: Option<bool>,
368    #[serde(default)]
369    /// 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.
370    pub css_extraction_map: Option<CSSExtractionMap>,
371    #[serde(default)]
372    /// The chunking algorithm to use.
373    pub chunking_alg: Option<ChunkingAlgDict>,
374    #[serde(default)]
375    /// 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.
376    pub disable_intercept: Option<bool>,
377    #[serde(default)]
378    /// The wait for events on the page. You need to make your `request` `chrome` or `smart`.
379    pub wait_for: Option<WaitFor>,
380    #[serde(default)]
381    /// Perform custom Javascript tasks on a url or url path. You need to make your `request` `chrome` or `smart`
382    pub execution_scripts: Option<ExecutionScriptsMap>,
383    #[serde(default)]
384    /// Perform web automated tasks on a url or url path. You need to make your `request` `chrome` or `smart`
385    pub automation_scripts: Option<WebAutomationMap>,
386    #[serde(default)]
387    /// The redirect policy for HTTP request. Set the value to Loose to allow all.
388    pub redirect_policy: Option<RedirectPolicy>,
389    #[serde(default)]
390    /// 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.
391    pub event_tracker: Option<EventTracker>,
392    #[serde(default)]
393    /// The timeout to stop the crawl.
394    pub crawl_timeout: Option<Timeout>,
395    #[serde(default)]
396    /// Evaluates given script in every frame upon creation (before loading frame's scripts).
397    pub evaluate_on_new_document: Option<Box<String>>,
398    #[serde(default)]
399    /// Runs the request using lite_mode:Lite mode reduces data transfer costs by 70%, with trade-offs in speed, accuracy,
400    /// geo-targeting, and reliability. It’s best suited for non-urgent data collection or when
401    /// targeting websites with minimal anti-bot protections.
402    pub lite_mode: Option<bool>,
403    #[serde(default)]
404    /// The proxy to use for request.
405    pub proxy: Option<ProxyType>,
406    #[serde(default)]
407    /// Use a remote proxy at ~70% reduced cost for file downloads.
408    /// This requires a user-supplied static IP proxy endpoint.
409    pub remote_proxy: Option<String>,
410    #[serde(default)]
411    /// Set the maximum number of credits to use per page.
412    /// Credits are measured in decimal units, where 10,000 credits equal one dollar (100 credits per penny).
413    /// Credit limiting only applies to request that are Javascript rendered using smart_mode or chrome for the 'request' type.
414    pub max_credits_per_page: Option<f64>,
415}
416
417/// The structure representing request parameters for a search request.
418#[derive(Debug, Default, Deserialize, Serialize, Clone)]
419pub struct SearchRequestParams {
420    /// The base request parameters.
421    #[serde(default, flatten)]
422    pub base: RequestParams,
423    // The search request.
424    pub search: String,
425    /// The search limit.
426    pub search_limit: Option<u32>,
427    // Fetch the page content. Defaults to true.
428    pub fetch_page_content: Option<bool>,
429    /// The search location of the request
430    pub location: Option<String>,
431    /// The country code of the request
432    pub country: Option<String>,
433    /// The language code of the request.
434    pub language: Option<String>,
435    /// The number of search results
436    pub num: Option<u32>,
437    /// The page of the search results.
438    pub page: Option<u32>,
439    #[serde(default)]
440    /// The websites limit if a list is sent from text or urls comma split. This helps automatic configuration of the system.
441    pub website_limit: Option<u32>,
442}
443
444/// Structure representing request parameters for transforming files.
445#[derive(Debug, Default, Deserialize, Serialize, Clone)]
446pub struct TransformParams {
447    #[serde(default)]
448    /// The format in which the result should be returned.
449    pub return_format: Option<ReturnFormat>,
450    #[serde(default)]
451    /// Specifies whether readability optimizations should be applied.
452    pub readability: Option<bool>,
453    #[serde(default)]
454    /// Clean the markdown or text for AI.
455    pub clean: Option<bool>,
456    #[serde(default)]
457    /// Clean the markdown or text for AI removing footers, navigation, and more.
458    pub clean_full: Option<bool>,
459    /// The data being transformed.
460    pub data: Vec<DataParam>,
461}
462
463#[derive(Serialize, Deserialize, Debug, Clone)]
464pub struct DataParam {
465    /// The HTML resource.
466    pub html: String,
467    /// The website url.
468    pub url: Option<String>,
469}
470
471/// the request type to perform
472#[derive(Debug, Default, Clone, Serialize, Deserialize, PartialEq, Eq)]
473#[serde(rename_all = "lowercase")]
474pub enum RequestType {
475    /// Default HTTP request
476    Http,
477    /// Chrome browser rendering
478    Chrome,
479    #[default]
480    /// Smart mode defaulting to HTTP and using Chrome when needed.
481    SmartMode,
482}
483
484/// Enum representing different return formats.
485#[derive(Default, Debug, Deserialize, Serialize, Clone, PartialEq, Eq, Hash)]
486#[serde(rename_all = "lowercase")]
487pub enum ReturnFormat {
488    #[default]
489    /// The default return format of the resource.
490    Raw,
491    /// Return the response as Markdown.
492    Markdown,
493    /// Return the response as Commonmark.
494    Commonmark,
495    /// Return the response as Html2text.
496    Html2text,
497    /// Return the response as Text.
498    Text,
499    /// Returns a screenshot as Base64Url
500    Screenshot,
501    /// Return the response as XML.
502    Xml,
503    /// Return the response as Bytes.
504    Bytes,
505}
spider_client/shapes/request.rs

spider_client/shapes/
request.rs
