Capability traits - SeleniumBase for Rust

Note

Methods are listed as name(arguments), called on a BaseCase as sb.name(...).await?. The argument names, the selector syntax and the return types are explained in Reading the API reference.

Capability traits#

seleniumbase_rs::{BrowserApi, ElementApi, AssertionApi, ScreenshotApi} are four small async_trait traits (defined in src/api/traits.rs). Both engines implement all four: BaseCase (WebDriver) and sb_cdp::Page (pure CDP). A helper written over the traits runs on either engine. None of them names an engine type; to keep an element handle, use BaseCase::find_element (a thirtyfour::WebElement) or Page::locator(..). Every method returns seleniumbase_rs::Result<T>. Selector strings are the same on both engines: CSS by default, XPath for //..., link text for link=... (see the Selectors guide).

use seleniumbase_rs::{AssertionApi, BrowserApi, ElementApi};async fn sign_in<E>(sb: &mut E) -> Result<(), seleniumbase_rs::SeleniumBaseError>where    E: BrowserApi + ElementApi + AssertionApi,{    sb.open("https://example.com/login").await?;    sb.type_text("#user", "alice").await?;    sb.click("#submit").await?;    sb.assert_element("#dashboard").await}

BrowserApi#

Method	`BaseCase`	`Page`
`open(&mut self, url: &str) -> Result<()>`	Navigates the active window; a URL without a scheme gets `https://`; a file path or an unsupported scheme (`ftp:`, `javascript:`, `mailto:`) is refused with `InvalidConfig` (use `file://` for a local file).	`Page::goto(url)`.
`quit(&mut self) -> Result<()>`	Closes the browser session.	Closes this tab only. To close the browser call `Browser::close`.
`refresh(&self) -> Result<()>`	Reloads the page.	`Page::reload()`.
`go_back(&self) -> Result<()>`	Browser back.	`Page::back()`.
`go_forward(&self) -> Result<()>`	Browser forward.	`Page::forward()`.
`get_title(&mut self) -> Result<String>`	The page title.	`Page::title()`.
`get_url(&mut self) -> Result<String>`	The current URL.	`Page::url()`.

ElementApi#

Method	`BaseCase`	`Page`
`click(&mut self, css: &str) -> Result<()>`	Clicks the element.	`locator(css).click()`.
`double_click(&mut self, css: &str) -> Result<()>`	Double-clicks it.	`locator(css).double_click()`.
`type_text(&mut self, css: &str, text: &str) -> Result<()>`	Waits for the element to be visible, clears it and types.	Clears the field, then types one key at a time.
`get_text(&mut self, css: &str) -> Result<String>`	The element’s visible text, after waiting for it to be present.	`locator(css).text()`.
`get_attribute(&mut self, css: &str, attr: &str) -> Result<Option<String>>`	The attribute value, `None` when absent.	`locator(css).attribute(attr)`.

AssertionApi#

Method	`BaseCase`	`Page`
`assert_title(&mut self, expected: &str) -> Result<()>`	Fails with `AssertionFailed` unless the title equals `expected` exactly.	`expect().to_have_title(expected)`, which retries until the timeout.
`assert_element(&self, css: &str) -> Result<()>`	Fails unless the element is present in the DOM (visible or not).	`locator(css).expect().to_exist()`.
`assert_text(&mut self, expected: &str, css: &str) -> Result<()>`	Fails unless the element’s text contains `expected`.	`locator(css).expect().to_contain_text(expected)`.
`assert_no_js_errors(&self) -> Result<()>`	Fails if the browser log holds a `SEVERE` entry (a few known CDN sources are ignored).	Fails if the document raised an uncaught script error; the message names the first.

Note

The trait method assert_text(expected, css) takes the text first, like BaseCase::assert_text(text, selector) and Python’s assert_text.

ScreenshotApi#

Method	Behaviour
`save_screenshot(&self, filename: &str) -> Result<PathBuf>`	Saves a PNG into the logs directory and returns its path. `filename` must be a plain file name (no directory part, no `..`). The same for both engines.
`screenshot_as_png(&self) -> Result<Vec<u8>>`	The current page as PNG bytes.