1 |
NAME |
2 |
AnyEvent::WebDriver - control browsers using the W3C WebDriver protocol |
3 |
|
4 |
SYNOPSIS |
5 |
# start geckodriver or any other w3c-compatible webdriver via the shell |
6 |
$ geckdriver -b myfirefox/firefox --log trace --port 4444 |
7 |
|
8 |
# then use it |
9 |
use AnyEvent::WebDriver; |
10 |
|
11 |
# create a new webdriver object |
12 |
my $wd = new AnyEvent::WebDriver; |
13 |
|
14 |
# create a new session with default capabilities. |
15 |
$wd->new_session ({}); |
16 |
|
17 |
$wd->navigate_to ("https://duckduckgo.com/html"); |
18 |
my $searchbox = $wd->find_element ("css selector" => 'input[type="text"]'); |
19 |
|
20 |
$wd->element_send_keys ($searchbox => "free software"); |
21 |
$wd->element_click ($wd->find_element ("css selector" => 'input[type="submit"]')); |
22 |
|
23 |
sleep 10; |
24 |
|
25 |
DESCRIPTION |
26 |
This module aims to implement the W3C WebDriver specification which is |
27 |
the standardised equivalent to the Selenium WebDriver API., which in |
28 |
turn aims at remotely controlling web browsers such as Firefox or |
29 |
Chromium. |
30 |
|
31 |
At the time of this writing, it was only available as a draft document, |
32 |
so changes will be expected. Also, only geckodriver did implement it, or |
33 |
at least, most of it. |
34 |
|
35 |
To make most of this module, or, in fact, to make any reasonable use of |
36 |
this module, you would need to refer tot he W3C WebDriver document, |
37 |
which can be found here <https://w3c.github.io/webdriver/>: |
38 |
|
39 |
https://w3c.github.io/webdriver/ |
40 |
|
41 |
CREATING WEBDRIVER OBJECTS |
42 |
new AnyEvent::WebDriver key => value... |
43 |
Create a new WebDriver object. Example for a remote WebDriver |
44 |
connection (the only type supported at the moment): |
45 |
|
46 |
my $wd = new AnyEvent::WebDriver host => "localhost", port => 4444; |
47 |
|
48 |
Supported keys are: |
49 |
|
50 |
endpoint => $string |
51 |
For remote connections, the endpoint to connect to (defaults to |
52 |
"http://localhost:4444"). |
53 |
|
54 |
proxy => $proxyspec |
55 |
The proxy to use (same as the "proxy" argument used by |
56 |
AnyEvent::HTTP). The default is "undef", which disables proxies. |
57 |
To use the system-provided proxy (e.g. "http_proxy" environment |
58 |
variable), specify a value of "default". |
59 |
|
60 |
autodelete => $boolean |
61 |
If true (the default), then automatically execute |
62 |
"delete_session" when the WebDriver object is destroyed with an |
63 |
active session. IF set to a false value, then the session will |
64 |
continue to exist. |
65 |
|
66 |
timeout => $seconds |
67 |
The HTTP timeout, in (fractional) seconds (default: 300, but |
68 |
this will likely drastically reduce). This timeout is reset on |
69 |
any activity, so it is not an overall request timeout. Also, |
70 |
individual requests might extend this timeout if they are known |
71 |
to take longer. |
72 |
|
73 |
SIMPLIFIED API |
74 |
This section documents the simplified API, which is really just a very |
75 |
thin wrapper around the WebDriver protocol commands. They all block |
76 |
(using AnyEvent condvars) the caller until the result is available, so |
77 |
must not be called from an event loop callback - see "EVENT BASED API" |
78 |
for an alternative. |
79 |
|
80 |
The method names are pretty much taken directly from the W3C WebDriver |
81 |
specification, e.g. the request documented in the "Get All Cookies" |
82 |
section is implemented via the "get_all_cookies" method. |
83 |
|
84 |
The order is the same as in the WebDriver draft at the time of this |
85 |
writing, and only minimal massaging is done to request parameters and |
86 |
results. |
87 |
|
88 |
SESSIONS |
89 |
$wd->new_session ({ key => value... }) |
90 |
Try to connect to the WebDriver and initialize a new session with a |
91 |
"new session" command, passing the given key-value pairs as value |
92 |
(e.g. "capabilities"). |
93 |
|
94 |
No session-dependent methods must be called before this function |
95 |
returns successfully, and only one session can be created per |
96 |
WebDriver object. |
97 |
|
98 |
On success, "$wd->{sid}" is set to the session ID, and |
99 |
"$wd->{capabilities}" is set to the returned capabilities. |
100 |
|
101 |
my $wd = new AnyEvent::Selenium endpoint => "http://localhost:4545"; |
102 |
|
103 |
$wd->new_session ({ |
104 |
capabilities => { |
105 |
pageLoadStrategy => "normal", |
106 |
}. |
107 |
}); |
108 |
|
109 |
$wd->delete_session |
110 |
Deletes the session - the WebDriver object must not be used after |
111 |
this call. |
112 |
|
113 |
$timeouts = $wd->get_timeouts |
114 |
Get the current timeouts, e.g.: |
115 |
|
116 |
my $timeouts = $wd->get_timeouts; |
117 |
=> { implicit => 0, pageLoad => 300000, script => 30000 } |
118 |
|
119 |
$wd->set_timeouts ($timeouts) |
120 |
Sets one or more timeouts, e.g.: |
121 |
|
122 |
$wd->set_timeouts ({ script => 60000 }); |
123 |
|
124 |
NAVIGATION |
125 |
$wd->navigate_to ($url) |
126 |
Navigates to the specified URL. |
127 |
|
128 |
$url = $wd->get_current_url |
129 |
Queries the current page URL as set by "navigate_to". |
130 |
|
131 |
$wd->back |
132 |
The equivalent of pressing "back" in the browser. |
133 |
|
134 |
$wd->forward |
135 |
The equivalent of pressing "forward" in the browser. |
136 |
|
137 |
$wd->refresh |
138 |
The equivalent of pressing "refresh" in the browser. |
139 |
|
140 |
$title = $wd->get_title |
141 |
Returns the current document title. |
142 |
|
143 |
COMMAND CONTEXTS |
144 |
$handle = $wd->get_window_handle |
145 |
Returns the current window handle. |
146 |
|
147 |
$wd->close_window |
148 |
Closes the current browsing context. |
149 |
|
150 |
$wd->switch_to_window ($handle) |
151 |
Changes the current browsing context to the given window. |
152 |
|
153 |
$handles = $wd->get_window_handles |
154 |
Return the current window handles as an array-ref of handle IDs. |
155 |
|
156 |
$handles = $wd->switch_to_frame ($frame) |
157 |
Switch to the given frame identified by $frame, which must be either |
158 |
"undef" to go back to the top-level browsing context, an integer to |
159 |
select the nth subframe, or an element object (as e.g. returned by |
160 |
the "element_object" method. |
161 |
|
162 |
$handles = $wd->switch_to_parent_frame |
163 |
Switch to the parent frame. |
164 |
|
165 |
$rect = $wd->get_window_rect |
166 |
Return the current window rect, e.g.: |
167 |
|
168 |
$rect = $wd->get_window_rect |
169 |
=> { height => 1040, width => 540, x => 0, y => 0 } |
170 |
|
171 |
$wd->set_window_rect ($rect) |
172 |
Sets the window rect. |
173 |
|
174 |
$wd->maximize_window |
175 |
$wd->minimize_window |
176 |
$wd->fullscreen_window |
177 |
Changes the window size by either maximising, minimising or making |
178 |
it fullscreen. In my experience, this will timeout if no window |
179 |
manager is running. |
180 |
|
181 |
ELEMENT RETRIEVAL |
182 |
$element_id = $wd->find_element ($location_strategy, $selector) |
183 |
Finds the first element specified by the given selector and returns |
184 |
its web element ID (the strong, not the object from the protocol). |
185 |
Raises an error when no element was found. |
186 |
|
187 |
$element = $wd->find_element ("css selector" => "body a"); |
188 |
$element = $wd->find_element ("link text" => "Click Here For Porn"); |
189 |
$element = $wd->find_element ("partial link text" => "orn"); |
190 |
$element = $wd->find_element ("tag name" => "input"); |
191 |
$element = $wd->find_element ("xpath" => '//input[@type="text"]'); |
192 |
=> e.g. "decddca8-5986-4e1d-8c93-efe952505a5f" |
193 |
|
194 |
$element_ids = $wd->find_elements ($location_strategy, $selector) |
195 |
As above, but returns an arrayref of all found element IDs. |
196 |
|
197 |
$element_id = $wd->find_element_from_element ($element_id, |
198 |
$location_strategy, $selector) |
199 |
Like "find_element", but looks only inside the specified $element. |
200 |
|
201 |
$element_ids = $wd->find_elements_from_element ($element_id, |
202 |
$location_strategy, $selector) |
203 |
Like "find_elements", but looks only inside the specified $element. |
204 |
|
205 |
my $head = $wd->find_element ("tag name" => "head"); |
206 |
my $links = $wd->find_elements_from_element ($head, "tag name", "link"); |
207 |
|
208 |
$element_id = $wd->get_active_element |
209 |
Returns the active element. |
210 |
|
211 |
ELEMENT STATE |
212 |
$bool = $wd->is_element_selected |
213 |
Returns whether the given input or option element is selected or |
214 |
not. |
215 |
|
216 |
$string = $wd->get_element_attribute ($element_id, $name) |
217 |
Returns the value of the given attribute. |
218 |
|
219 |
$string = $wd->get_element_property ($element_id, $name) |
220 |
Returns the value of the given property. |
221 |
|
222 |
$string = $wd->get_element_css_value ($element_id, $name) |
223 |
Returns the value of the given CSS value. |
224 |
|
225 |
$string = $wd->get_element_text ($element_id) |
226 |
Returns the (rendered) text content of the given element. |
227 |
|
228 |
$string = $wd->get_element_tag_name ($element_id) |
229 |
Returns the tag of the given element. |
230 |
|
231 |
$rect = $wd->get_element_rect ($element_id) |
232 |
Returns the element rect(angle) of the given element. |
233 |
|
234 |
$bool = $wd->is_element_enabled |
235 |
Returns whether the element is enabled or not. |
236 |
|
237 |
ELEMENT INTERACTION |
238 |
$wd->element_click ($element_id) |
239 |
Clicks the given element. |
240 |
|
241 |
$wd->element_clear ($element_id) |
242 |
Clear the contents of the given element. |
243 |
|
244 |
$wd->element_send_keys ($element_id, $text) |
245 |
Sends the given text as key events to the given element. |
246 |
|
247 |
DOCUMENT HANDLING |
248 |
$source = $wd->get_page_source |
249 |
Returns the (HTML/XML) page source of the current document. |
250 |
|
251 |
$results = $wd->execute_script ($javascript, $args) |
252 |
Synchronously execute the given script with given arguments and |
253 |
return its results ($args can be "undef" if no arguments are |
254 |
wanted/needed). |
255 |
|
256 |
$ten = $wd->execute_script ("return arguments[0]+arguments[1]", [3, 7]); |
257 |
|
258 |
$results = $wd->execute_async_script ($javascript, $args) |
259 |
Similar to "execute_script", but doesn't wait for script to return, |
260 |
but instead waits for the script to call its last argument, which is |
261 |
added to $args automatically. |
262 |
|
263 |
$twenty = $wd->execute_async_script ("arguments[0](20)", undef); |
264 |
|
265 |
COOKIES |
266 |
$cookies = $wd->get_all_cookies |
267 |
Returns all cookies, as an arrayref of hashrefs. |
268 |
|
269 |
# google surely sets a lot of cookies without my consent |
270 |
$wd->navigate_to ("http://google.com"); |
271 |
use Data::Dump; |
272 |
ddx $wd->get_all_cookies; |
273 |
|
274 |
$cookie = $wd->get_named_cookie ($name) |
275 |
Returns a single cookie as a hashref. |
276 |
|
277 |
$wd->add_cookie ($cookie) |
278 |
Adds the given cookie hashref. |
279 |
|
280 |
$wd->delete_cookie ($name) |
281 |
Delete the named cookie. |
282 |
|
283 |
$wd->delete_all_cookies |
284 |
Delete all cookies. |
285 |
|
286 |
ACTIONS |
287 |
$wd->perform_actions ($actions) |
288 |
Perform the given actions (an arrayref of action specifications |
289 |
simulating user activity). For further details, read the spec. |
290 |
|
291 |
An example to get you started: |
292 |
|
293 |
$wd->navigate_to ("https://duckduckgo.com/html"); |
294 |
$wd->set_timeouts ({ implicit => 10000 }); |
295 |
my $input = $wd->find_element ("css selector", 'input[type="text"]'); |
296 |
$wd->perform_actions ([ |
297 |
{ |
298 |
id => "myfatfinger", |
299 |
type => "pointer", |
300 |
pointerType => "touch", |
301 |
actions => [ |
302 |
{ type => "pointerMove", duration => 100, origin => $wd->element_object ($input), x => 40, y => 5 }, |
303 |
{ type => "pointerDown", button => 1 }, |
304 |
{ type => "pause", duration => 40 }, |
305 |
{ type => "pointerUp", button => 1 }, |
306 |
], |
307 |
}, |
308 |
{ |
309 |
id => "mykeyboard", |
310 |
type => "key", |
311 |
actions => [ |
312 |
{ type => "pause" }, |
313 |
{ type => "pause" }, |
314 |
{ type => "pause" }, |
315 |
{ type => "pause" }, |
316 |
{ type => "keyDown", value => "a" }, |
317 |
{ type => "pause", duration => 100 }, |
318 |
{ type => "keyUp", value => "a" }, |
319 |
{ type => "pause", duration => 100 }, |
320 |
{ type => "keyDown", value => "b" }, |
321 |
{ type => "pause", duration => 100 }, |
322 |
{ type => "keyUp", value => "b" }, |
323 |
{ type => "pause", duration => 2000 }, |
324 |
{ type => "keyDown", value => "\x{E007}" }, # enter |
325 |
{ type => "pause", duration => 100 }, |
326 |
{ type => "keyUp", value => "\x{E007}" }, # enter |
327 |
{ type => "pause", duration => 5000 }, |
328 |
], |
329 |
}, |
330 |
]); |
331 |
|
332 |
$wd->release_actions |
333 |
Release all keys and pointer buttons currently depressed. |
334 |
|
335 |
USER PROMPTS |
336 |
$wd->dismiss_alert |
337 |
Dismiss a simple dialog, if present. |
338 |
|
339 |
$wd->accept_alert |
340 |
Accept a simple dialog, if present. |
341 |
|
342 |
$text = $wd->get_alert_text |
343 |
Returns the text of any simple dialog. |
344 |
|
345 |
$text = $wd->send_alert_text |
346 |
Fills in the user prompt with the given text. |
347 |
|
348 |
SCREEN CAPTURE |
349 |
$wd->take_screenshot |
350 |
Create a screenshot, returning it as a PNG image in a "data:" URL. |
351 |
|
352 |
$wd->take_element_screenshot ($element_id) |
353 |
Accept a simple dialog, if present. |
354 |
|
355 |
HELPER METHODS |
356 |
$object = AnyEvent::WebDriver->element_object ($element_id) |
357 |
$object = $wd->element_object ($element_id) |
358 |
Encoding element IDs in data structures is done by representing them |
359 |
as an object with a special key and the element ID as value. This |
360 |
helper method does this for you. |
361 |
|
362 |
EVENT BASED API |
363 |
This module wouldn't be a good AnyEvent citizen if it didn't have a true |
364 |
event-based API. |
365 |
|
366 |
In fact, the simplified API, as documented above, is emulated via the |
367 |
event-based API and an "AUTOLOAD" function that automatically provides |
368 |
blocking wrappers around the callback-based API. |
369 |
|
370 |
Every method documented in the "SIMPLIFIED API" section has an |
371 |
equivalent event-based method that is formed by appending a underscore |
372 |
("_") to the method name, and appending a callback to the argument list |
373 |
(mnemonic: the underscore indicates the "the action is not yet finished" |
374 |
after the call returns). |
375 |
|
376 |
For example, instead of a blocking calls to "new_session", "navigate_to" |
377 |
and "back", you can make a callback-based ones: |
378 |
|
379 |
my $cv = AE::cv; |
380 |
|
381 |
$wd->new_session ({}, sub { |
382 |
my ($status, $value) = @_, |
383 |
|
384 |
die "error $value->{error}" if $status ne "200"; |
385 |
|
386 |
$wd->navigate_to_ ("http://www.nethype.de", sub { |
387 |
|
388 |
$wd->back_ (sub { |
389 |
print "all done\n"; |
390 |
$cv->send; |
391 |
}); |
392 |
|
393 |
}); |
394 |
}); |
395 |
|
396 |
$cv->recv; |
397 |
|
398 |
While the blocking methods "croak" on errors, the callback-based ones |
399 |
all pass two values to the callback, $status and $res, where $status is |
400 |
the HTTP status code (200 for successful requests, typically 4xx or 5xx |
401 |
for errors), and $res is the value of the "value" key in the JSON |
402 |
response object. |
403 |
|
404 |
Other than that, the underscore variants and the blocking variants are |
405 |
identical. |
406 |
|
407 |
LOW LEVEL API |
408 |
All the simplified API methods are very thin wrappers around WebDriver |
409 |
commands of the same name. They are all implemented in terms of the |
410 |
low-level methods ("req", "get", "post" and "delete"), which exists in |
411 |
blocking and callback-based variants ("req_", "get_", "post_" and |
412 |
"delete_"). |
413 |
|
414 |
Examples are after the function descriptions. |
415 |
|
416 |
$wd->req_ ($method, $uri, $body, $cb->($status, $value)) |
417 |
$value = $wd->req ($method, $uri, $body) |
418 |
Appends the $uri to the "endpoint/session/{sessionid}/" URL and |
419 |
makes a HTTP $method request ("GET", "POST" etc.). "POST" requests |
420 |
can provide a UTF-8-encoded JSON text as HTTP request body, or the |
421 |
empty string to indicate no body is used. |
422 |
|
423 |
For the callback version, the callback gets passed the HTTP status |
424 |
code (200 for every successful request), and the value of the |
425 |
"value" key in the JSON response object as second argument. |
426 |
|
427 |
$wd->get_ ($uri, $cb->($status, $value)) |
428 |
$value = $wd->get ($uri) |
429 |
Simply a call to "req_" with $method set to "GET" and an empty body. |
430 |
|
431 |
$wd->post_ ($uri, $data, $cb->($status, $value)) |
432 |
$value = $wd->post ($uri, $data) |
433 |
Simply a call to "req_" with $method set to "POST" - if $body is |
434 |
"undef", then an empty object is send, otherwise, $data must be a |
435 |
valid request object, which gets encoded into JSON for you. |
436 |
|
437 |
$wd->delete_ ($uri, $cb->($status, $value)) |
438 |
$value = $wd->delete ($uri) |
439 |
Simply a call to "req_" with $method set to "DELETE" and an empty |
440 |
body. |
441 |
|
442 |
Example: implement "get_all_cookies", which is a simple "GET" request |
443 |
without any parameters: |
444 |
|
445 |
$cookies = $wd->get ("cookie"); |
446 |
|
447 |
Example: implement "execute_script", which needs some parameters: |
448 |
|
449 |
$results = $wd->post ("execute/sync" => { script => "$javascript", args => [] }); |
450 |
|
451 |
Example: call "find_elements" to find all "IMG" elements, stripping the |
452 |
returned element objects to only return the element ID strings: |
453 |
|
454 |
my $elems = $wd->post (elements => { using => "css selector", value => "img" }); |
455 |
|
456 |
# yes, the W3C found an interesting way around the typelessness of JSON |
457 |
$_ = $_->{"element-6066-11e4-a52e-4f735466cecf"} |
458 |
for @$elems; |
459 |
|
460 |
HISTORY |
461 |
This module was unintentionally created (it started inside some quickly |
462 |
hacked-together script) simply because I couldn't get the existing |
463 |
"Selenium::Remote::Driver" module to work, ever, despite multiple |
464 |
attempts over the years and trying to report multiple bugs, which have |
465 |
been completely ignored. It's also not event-based, so, yeah... |
466 |
|
467 |
AUTHOR |
468 |
Marc Lehmann <schmorp@schmorp.de> |
469 |
http://anyevent.schmorp.de |
470 |
|