A command is what tells Selenium what to do. Selenium commands come in three 'flavors': Actions, Accessors and Assertions. Each command call is one line in the test table of the form:
command target value
Actions are commands that generally manipulate the state of the application. They do things like "click this link" and "select that option". If an Action fails, or has an error, the execution of the current test is stopped.
Many Actions can be called with the "AndWait" suffix, e.g. "clickAndWait". This suffix tells Selenium that the action will cause the browser to make a call to the server, and that Selenium should wait for a new page to load.
Accessors examine the state of the application and store the results in variables, e.g. "storeTitle". They are also used to automatically generate Assertions.
Assertions are like Accessors, but they verify that the state of the application conforms to what is expected. Examples include "make sure the page title is X" and "verify that this checkbox is checked".
All Selenium Assertions can be used in 3 modes: "assert", "verify", and "waitFor". For example, you can "assertText", "verifyText" and "waitForText". When an "assert" fails, the test is aborted. When a "verify" fails, the test will continue execution, logging the failure. This allows a single "assert" to ensure that the application is on the correct page, followed by a bunch of "verify" assertions to test form field values, labels, etc.
"waitFor" commands wait for some condition to become true (which can be useful for testing Ajax applications). They will succeed immediately if the condition is already true. However, they will fail and halt the test if the condition does not become true within the current timeout setting (see the setTimeout action below).
Element Locators tell Selenium which HTML element a command refers to. Many commands require an Element Locator as the "target" attribute. Examples of Element Locators include "elementId" and "document.forms[0].element". These are described more clearly in the next section.
Patterns are used for various reasons, e.g. to specify the expected value of an input field, or identify a select option. Selenium supports various types of pattern, including regular-expressions, all of which are described in more detail below.
Defines an object that runs Selenium commands.
Element Locators tell Selenium which HTML element a command refers to. The format of a locator is:
locatorType = argument
We support the following strategies for locating elements:
The name may optionally be followed by one or more element-filters, separated from the name by whitespace. If the filterType is not specified, value is assumed.
Currently the css selector locator supports all css1, css2 and css3 selectors except namespace in css3, some pseudo classes(:nth-of-type, :nth-last-of-type, :first-of-type, :last-of-type, :only-of-type, :visited, :hover, :active, :focus, :indeterminate) and pseudo elements(::first-line, ::first-letter, ::selection, ::before, ::after).
Without an explicit locator prefix, Selenium uses the following default strategies:
Element filters can be used with a locator to refine a list of candidate elements. They are currently used only in the 'name' element-locator.
Filters look much like locators, ie.
filterType = argumentSupported element-filters are:
value= valuePattern
Matches elements based on their values. This is particularly useful for refining a list of similarly-named toggle-buttons.index= index
Selects a single element based on its position in the list (offset from zero).
Various Pattern syntaxes are available for matching string values:
If no pattern prefix is specified, Selenium assumes that it's a "glob" pattern.
For commands that return multiple values (such as verifySelectOptions), the string being matched is a comma-separated list of the return values, where both commas and backslashes in the values are backslash-escaped. When providing a pattern, the optional matching syntax (i.e. glob, regexp, etc.) is specified once, as usual, at the beginning of the pattern.
Arguments:
return inDocument.getElementById(locator);
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
By default, Selenium's overridden window.confirm() function will return true, as if the user had manually clicked OK; after running this command, the next call to confirm() will return false, as if the user had clicked Cancel. Selenium will then resume using the default behavior for future confirmations, automatically returning true (OK) unless/until you explicitly call this command for each confirmation.
Take note - every time a confirmation comes up, you must consume it with a corresponding getConfirmation, or else the next selenium operation will fail.
Undo the effect of calling chooseCancelOnNextConfirmation. Note that Selenium's overridden window.confirm() function will normally automatically return true, as if the user had manually clicked OK, so you shouldn't need to use this command unless for some reason you need to change your mind prior to the next confirmation. After any confirmation, Selenium will resume using the default behavior for future confirmations, automatically returning true (OK) unless/until you explicitly call chooseCancelOnNextConfirmation for each confirmation.
Take note - every time a confirmation comes up, you must consume it with a corresponding getConfirmation, or else the next selenium operation will fail.
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
This command can also be a useful workaround for bug SEL-339. In some cases, Selenium will be unable to intercept a call to window.open (if the call occurs during or before the "onLoad" event, for example). In those cases, you can force Selenium to notice the open window's name by using the Selenium openWindow command, using an empty (blank) url, like this: openWindow("", "myFunnyWindow").
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Arguments:
Option locators provide different ways of specifying options of an HTML Select element (e.g. for selecting a specific option, or for asserting that the selected option satisfies a specification). There are several forms of Select Option Locator.
If no option locator prefix is provided, the default behaviour is to match on label.
Arguments:
You may also use a DOM expression to identify the frame you want directly,
like this:
dom=frames["main"].frames["subframe"]
Arguments:
Window locators provide different ways of specifying the window object: by title, by internal JavaScript "name," or by JavaScript variable.
If no window locator prefix is provided, we'll try to guess what you mean like this:
1.) if windowID is null, (or the string "null") then it is assumed the user is referring to the original window instantiated by the browser).
2.) if the value of the "windowID" parameter is a JavaScript variable name in the current application window, then it is assumed that this variable contains the return value from a call to the JavaScript window.open() method.
3.) Otherwise, selenium looks in a hash it maintains that maps string names to window "names".
4.) If that fails, we'll try looping over all of the known windows to try to find the appropriate "title". Since "title" is not necessarily unique, this may have unexpected behavior.
If you're having trouble figuring out the name of a window that you want to manipulate, look at the Selenium log messages which identify the names of windows created via window.open (and therefore intercepted by Selenium). You will see messages like the following for each window as it is opened:
debug: window.open call intercepted; window ID (which you can use with selectWindow()) is "myNewWindow"
In some cases, Selenium will be unable to intercept a call to window.open (if the call occurs during or before the "onLoad" event, for example). (This is bug SEL-339.) In those cases, you can force Selenium to notice the open window's name by using the Selenium openWindow command, using an empty (blank) url, like this: openWindow("", "myFunnyWindow").
Arguments:
Arguments:
Arguments:
Setting this value to 0 means that we'll send a "mousemove" event to every single pixel in between the start location and the end location; that can be very slow, and may cause some browsers to force the JavaScript to timeout.
If the mouse speed is greater than the distance between the two dragged objects, we'll just send one "mousemove" at the start location and then one final one at the end location.
Arguments:
Arguments:
Actions that require waiting include "open" and the "waitFor*" actions.
The default timeout is 30 seconds.Arguments:
Arguments:
Arguments:
Can also be used to set the value of combo boxes, check boxes, etc. In these cases, value should be the value of the option selected, not the visible text.
Arguments:
This is a convenience method for calling keyDown, keyUp, keyPress for every character in the specified string; this is useful for dynamic UI widgets (like auto-completing combo boxes) that require explicit key events.
Unlike the simple "type" command, which forces the specified value into the page directly, this command may or may not have any visible effect, even in cases where typing keys would normally have a visible effect. For example, if you use "typeKeys" on a form element, you may or may not see the results of what you typed in the field.
In some cases, you may need to use the simple "type" command to set the value of the field and then the "typeKeys" command to send the keystroke events corresponding to what you just typed.
Arguments:
Arguments:
Arguments:
Note that, by default, the snippet will be run in the runner's test window, not in the window
of your application. To get the window of your application, you can use
the JavaScript snippet
selenium.browserbot.getCurrentWindow()
, and then
run your JavaScript in there
Arguments:
Selenium constantly keeps track of new pages and frames loading, and sets a "newPageLoaded" flag when it first notices a page load.
See waitForPageToLoad for more information.Arguments:
You can use this command instead of the "AndWait" suffixes, "clickAndWait", "selectAndWait", "typeAndWait" etc. (which are only available in the JS API).
Selenium constantly keeps track of new pages loading, and sets a "newPageLoaded" flag when it first notices a page load. Running any other Selenium command after turns the flag to false. Hence, if you want to wait for a page to load, you must wait immediately after a Selenium command that caused a page-load.
Arguments:
Arguments:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
See the select command for more information about option locators.
Arguments:
Related Assertions, automatically generated:
Getting an alert has the same effect as manually clicking OK. If an alert is generated but you do not consume it with getAlert, the next Selenium action will fail.
Under Selenium, JavaScript alerts will NOT pop up a visible alert dialog.
Selenium does NOT support JavaScript alerts that are generated in a page's onload() event handler. In this case a visible dialog WILL be generated and Selenium will hang until someone manually clicks OK.
Related Assertions, automatically generated:
If a given button has no ID, it will appear as "" in this array.
Related Assertions, automatically generated:
If a given field has no ID, it will appear as "" in this array.
Related Assertions, automatically generated:
If a given link has no ID, it will appear as "" in this array.
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
By default, the confirm function will return true, having the same effect as manually clicking OK. This can be changed by prior execution of the chooseCancelOnNextConfirmation command.
If an confirmation is generated but you do not consume it with getConfirmation, the next Selenium action will fail.
NOTE: under Selenium, JavaScript confirmations will NOT pop up a visible dialog.
NOTE: Selenium does NOT support JavaScript confirmations that are generated in a page's onload() event handler. In this case a visible dialog WILL be generated and Selenium will hang until you manually click OK.
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Specifically, if the cursor/selection has been cleared by JavaScript, this command will tend to return the position of the last location of the cursor, even though the cursor is now gone from the page. This is filed as SEL-243.
This method will fail if the specified element isn't an input element or textarea, or there is no cursor in the element.Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Note that, by default, the snippet will run in the context of the "selenium"
object itself, so
this
will refer to the Selenium object. Use
window
to
refer to the window of your application, e.g.
window.document.getElementById('foo')
If you need to use
a locator to refer to a single element in your application page, you can
use
this.browserbot.findElement("id=foo")
where "id=foo" is your locator.
Arguments:
Related Assertions, automatically generated:
This is useful because of JavaScript preprocessing. It is used to generate commands like assertExpression and waitForExpression.
Arguments:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Successful handling of the prompt requires prior execution of the answerOnNextPrompt command. If a prompt is generated but you do not get/verify it, the next Selenium action will fail.
NOTE: under Selenium, JavaScript prompts will NOT pop up a visible dialog.
NOTE: Selenium does NOT support JavaScript prompts that are generated in a page's onload() event handler. In this case a visible dialog WILL be generated and Selenium will hang until someone manually clicks OK.
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
This is useful in proxy injection mode, where this code runs in every browser frame and window, and sometimes the selenium server needs to identify the "current" frame. In this case, when the test calls selectFrame, this routine is called for each frame to figure out which one has been selected. The selected frame will return true, while all others will return false.
Arguments:
Related Assertions, automatically generated:
This is useful in proxy injection mode, where this code runs in every browser frame and window, and sometimes the selenium server needs to identify the "current" window. In this case, when the test calls selectWindow, this routine is called for each window to figure out which one has been selected. The selected window will return true, while all others will return false.
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
This function never throws an exception
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
This function never throws an exception
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
This function never throws an exception
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
Arguments:
Related Assertions, automatically generated:
All Selenium command parameters can be constructed using both simple variable substitution as well as full javascript. Both of these mechanisms can access previously stored variables, but do so using different syntax.
The commands store, storeValue and storeText can be used to store a variable value for later access. Internally, these variables are stored in a map called "storedVars", with values keyed by the variable name. These commands are documented in the command reference.
Variable substitution
Variable substitution provides a simple way to include a previously stored variable in a command parameter. This is a simple mechanism, by which the variable to substitute is indicated by ${variableName}. Multiple variables can be substituted, and intermixed with static text.
Example:
store Mr title storeValue nameField surname store ${title} ${surname} fullname type textElement Full name is: ${fullname} Javascript evaluation
Javascript evaluation provides the full power of javascript in constructing a command parameter. To use this mechanism, the entire parameter value must be prefixed by 'javascript{' with a trailing '}'. The text inside the braces is evaluated as a javascript expression, and can access previously stored variables using the storedVars map detailed above. Note that variable substitution cannot be combined with javascript evaluation.
Example:
store javascript{'merchant' + (new Date()).getTime()} merchantId type textElement javascript{storedVars['merchantId'].toUpperCase()}
HTML automatically normalizes whitespace within elements, ignoring leading/trailing spaces and converting extra spaces, tabs and newlines into a single space. When Selenium reads text out of the page, it attempts to duplicate this behavior, so you can ignore all the tabs and newlines in your HTML and do assertions based on how the text looks in the browser when rendered. We do this by replacing all non-visible whitespace (including the non-breaking space " ") with a single space. All visible newlines (<br>, <p>, and <pre>formatted newlines) should be preserved.
We use the same normalization logic on the text of HTML Selenese test case tables. This has a number of advantages. First, you don't need to look at the HTML source of the page to figure out what your assertions should be; " " symbols are invisible to the end user, and so you shouldn't have to worry about them when writing Selenese tests. (You don't need to put " " markers in your test case to assertText on a field that contains " ".) You may also put extra newlines and spaces in your Selenese <td> tags; since we use the same normalization logic on the test case as we do on the text, we can ensure that assertions and the extracted text will match exactly.
This creates a bit of a problem on those rare occasions when you really want/need to insert extra whitespace in your test case. For example, you may need to type text in a field like this: "foo ". But if you simply write <td>foo </td> in your Selenese test case, we'll replace your extra spaces with just one space.
This problem has a simple workaround. We've defined a variable in Selenese, ${space}, whose value is a single space. You can use ${space} to insert a space that won't be automatically trimmed, like this: <td>foo${space}${space}${space}</td>. We've also included a variable ${nbsp}, that you can use to insert a non-breaking space.
Note that XPaths do not normalize whitespace the way we do. If you need to write an XPath like //div[text()="hello world"] but the HTML of the link is really "hello world", you'll need to insert a real " " into your Selenese test case to get it to match, like this: //div[text()="hello${nbsp}world"].
It can be quite simple to extend Selenium, adding your own actions, assertions and locator-strategies. This is done with javascript by adding methods to the Selenium object prototype, and the PageBot object prototype. On startup, Selenium will automatically look through methods on these prototypes, using name patterns to recognise which ones are actions, assertions and locators.
The following examples try to give an indication of how Selenium can be extended with javascript.
Actions
All doFoo methods on the Selenium prototype are added as actions. For each action foo there is also an action fooAndWait registered. An action method can take up to 2 parameters, which will be passed the second and third column values in the test.
Example: Add a "typeRepeated" action to Selenium, which types the text twice into a text box.
Selenium.prototype.doTypeRepeated = function(locator, text) { // All locator-strategies are automatically handled by "findElement" var element = this.page().findElement(locator); // Create the text to type var valueToType = text + text; // Replace the element text with the new text this.page().replaceText(element, valueToType); };
Accessors/Assertions
All getFoo and isFoo methods on the Selenium prototype are added as accessors (storeFoo). For each accessor there is an assertFoo, verifyFoo and waitForFoo registered. An assert method can take up to 2 parameters, which will be passed the second and third column values in the test. You can also define your own assertions literally as simple "assert" methods, which will also auto-generate "verify" and "waitFor" commands.
Example: Add a valueRepeated assertion, that makes sure that the element value consists of the supplied text repeated. The 2 commands that would be available in tests would beassertValueRepeated and verifyValueRepeated.
Selenium.prototype.assertValueRepeated = function(locator, text) { // All locator-strategies are automatically handled by "findElement" var element = this.page().findElement(locator); // Create the text to verify var expectedValue = text + text; // Get the actual element value var actualValue = element.value; // Make sure the actual value matches the expected Assert.matches(expectedValue, actualValue); };
Automatic availability of storeFoo, assertFoo, assertNotFoo, waitForFoo and waitForNotFoo for every getFoo
All getFoo and isFoo methods on the Selenium prototype automatically result in the availability of storeFoo, assertFoo, assertNotFoo, verifyFoo, verifyNotFoo, waitForFoo, and waitForNotFoo commands.
Example, if you add a getTextLength() method, the following commands will automatically be available: storeTextLength, assertTextLength, assertNotTextLength, verifyTextLength, verifyNotTextLength, waitForTextLength, and waitForNotTextLength commands.
Selenium.prototype.getTextLength = function(locator, text) { return this.getText(locator).length; };Also note that the assertValueRepeated method described above could have been implemented using isValueRepeated, with the added benefit of also automatically getting assertNotValueRepeated, storeValueRepeated, waitForValueRepeated and waitForNotValueRepeated.
Locator Strategies
All locateElementByFoo methods on the PageBot prototype are added as locator-strategies. A locator strategy takes 2 parameters, the first being the locator string (minus the prefix), and the second being the document in which to search.
Example: Add a "valuerepeated=" locator, that finds the first element a value attribute equal to the the supplied value repeated.
// The "inDocument" is a the document you are searching. PageBot.prototype.locateElementByValueRepeated = function(text, inDocument) { // Create the text to search for var expectedValue = text + text; // Loop through all elements, looking for ones that have // a value === our expected value var allElements = inDocument.getElementsByTagName("*"); for (var i = 0; i < allElements.length; i++) { var testElement = allElements[i]; if (testElement.value && testElement.value === expectedValue) { return testElement; } } return null; };
user-extensions.js
By default, Selenium looks for a file called "user-extensions.js", and loads the javascript code found in that file. This file provides a convenient location for adding features to Selenium, without needing to modify the core Selenium sources.
In the standard distibution, this file does not exist. Users can create this file and place their extension code in this common location, removing the need to modify the Selenium sources, and hopefully assisting with the upgrade process.