Browser¶

Basic usages¶

The browser is a part of oct.core module, and is instantiate by the GenericTransaction class in its __init__ method. The browser can be used as a stand-alone, and for advanced scripts it’s good to know how to use it. So how to use it ? First, you need to instantiate a new Browser object:

from oct.core.browser import Browser

br = Browser()

The Browser object takes two optional parameters :

sessions if you want to use custom session manager, default value : requests.Session()

base_url for setting up your links when parsing, default to empty string

Now you can use the browser to access urls :

response = br.open_url('http://localhost/index.html')
print(response.status_code)
response = br.open_url('http://localhost/other_page.html')
print(response.status_code)

This script opens two urls, and for each one display the status_code of the response object returned by the open_url method.

Since the return value is simply the return of the requests.get or of the requests.post method, you can access all properties of a basic requests.Response object. But we add one thing to it, an html property, containing an lxml.html object, representing the opened page.

The html property can be used for parsing or getting elements with the lxml syntax, since it’s a standard object from lxml.html parsing.

For example you can access all forms object by using :

response.html.forms

Or even use the xpath syntax

And can you check the render of the page ? Of course, you don’t need other imports, we’ve implemented an open_in_browser static method, calling the lxml.html.open_in_browser method. You can use it like this :

response = br.open_url('http://localhost/index.html')
br.open_in_browser(response)

This will open the page in your default system browser.

A last thing you need to know. Each time the .html property is filled, the browser make a call to the make_links_absolute method of lxml. If you want to avoid that, simply do not provide a base_url for your browser instance, it’s used only for this call

Form manipulation¶

Like we said in the previous part of this documentation, you can use all the lxml methods for parsing your page. But again, we have done a part of the job for you.

Let’s say that we have a simple html page like this at the index of your localhost favorite web server:

<!DOCTYPE html>
<html>

<head>
    <title> My test page </title>
</head>

<body>
    <div id="my_form_block">
        <form action="/action.py" method="post">
            <input type="text" name="firstname" />
        </form>
    </div>
</body>

</html>

A very simple page, but it’s just for the example.

Now let’s say that we want to get this form and submit it from the browser object :

from oct.core.browser import Browser

# instantiate the browser
br = Browser(base_url='http://localhost')

# open the url
br.open_url('http://localhost')

# now we getting the form, using css selector
br.get_form(selector='div#my_form_block > form')

# we now have two properties for handling the form
# br.form, containing the lxml for object
# br.form_data, a dict containing all fields and values
# let's just set the value and submit it
br.form_data['firstname'] = 'my name'

# and submit it
response = br.submit_form()

# and check the status code
print(response.status_code)

And yes, that’s it ! Simple, no ? Thanks to the awesome cssselector python library, getting your forms are now simpler (unless you know nothing about css selectors) but even if we don’t want or can not use it, we can still use the get_form method, and use the nr parameter. The nr param simply represent the position of the form in our page. Here, simple we only have one form, so let’s update our script :

from oct.core.browser import Browser

# instantiate the browser
br = Browser(base_url='http://localhost')

# open the url
br.open_url('http://localhost')

# now we getting the form, using css selector
br.get_form(nr=0)

# we now have two properties for handling the form
# br.form, containing the lxml for object
# br.form_data, a dict containing all fields and values
# let's just set the value and submit it
br.form_data['firstname'] = 'my name'

# and submit it
response = br.submit_form()

# and check the status code
print(response.status_code)

And here it is, same result !

For more information about form manipulation, please see the lxml documentation

More navigation¶

You can follow links inside the html page like this :

from oct.core.browser import Browser

# instantiate the browser
br = Browser(base_url='http://localhost')

# open the url
br.open_url('http://localhost')

# now we can follow any link using css selector or a regex
# the regex will look at the text or the href attribute of the link
response = br.follow_link('a.my_links', '.*this link.*')

# oooops wrong link ! (yeah i know, that's doesn't append in script by try to imagine)
# let's go back
response = br.back() # after this we will be again at the index page

And that’s it ! The follow_link method is pretty simple actually, it just finds a link by regex and / or css selector, and then opens the url contained in the href attribute of this link.

What about the navigation history ? Well it’s not a big deal, only a small history management, no next management for now. But it allow you to go back and see all pages opened previously. What appends actually when you go back ? It open the previous url in the history list property, and then delete the next page of it. So yeah, i know, pretty bad. But stay tuned, better history management is coming !

Module details¶

octbrowser.browser module¶

class octbrowser.browser.Browser(session=None, base_url='')¶

Bases: object

This class represent a minimal browser. Build on top of lxml awesome library it let you write script for accessing or testing website with python scripts

Parameters:	session – The session object to use. If set to None will use requests.Session base_url – The base url for the website, will append it for every link without a full url

add_header(name, value)¶

Allow you to add custom header, one by one. Specify existing name for update Headers will be used by every request

Parameters:	name (str) – the key of the header value (str) – the associated value
Returns:	None

back()¶

Go to the previous url in the history property

Returns:	the Response object

clean_session()¶

This function is called by the core of multi-mechanize. It cleans the session for avoiding cache or cookies errors, or giving false results based on cache

Returns:	None

del_header(key)¶

Try to delete the ‘key’ of headers property

Parameters:	key (mixed) – the key to delete
Returns:	None

follow_link(selector, url_regex=None)¶

Will access the first link found with the selector

Raise:: oct.core.exceptions.LinkNotFound

Parameters:	selector – a string representing a css selector url_regex – regex for finding the url, can represent the href attribute or the link content
Returns:	Response object

get_form(selector=None, nr=0, at_base=False)¶

Get the form selected by the selector and / or the nr param

Raise:

oct.core.exceptions.FormNotFoundException
oct.core.exceptions.NoUrlOpen

Parameters:	selector – A css-like selector for finding the form nr – the index of the form, if selector is set to None, it will search on the hole page at_base – must be set to true in case of form action is on the base_url page
Returns:	None

get_html_element(selector)¶

Return a html element as string. The element will be find using the selector param

Use this method for get single html elements, if you want to get a list of elements, please use get_html_elements

Parameters:	selector (str) – a string representing a css selector
Returns:	a string containing the element, if multiples elements are find, it will concat them
Return type:	str

get_html_elements(selector)¶

Return a list of lxml.html.HtmlElement matching the selector argument

Parameters:	selector (str) – a string representing a css selector
Returns:	a list of lxml.html.HtmlElement of finded elements
Return type:	list

get_ressource(selector, output_dir, source_attribute='src')¶

Get a specified ressource and write it to the output dir

Raise:: OSError

Parameters:	selector (str) – a string representing a css selector output_dir (str) – the directory where the ressources will be wright source_attribute (str) – the attribute to retreive the url needed for downloading the ressource
Returns:	True if ressources as been correctly downled, False in other case

get_select_values()¶

Get the available values of all select and select multiple fields in form

Returns:	a dict containing all values for each fields

history¶

Return the actual history

Returns:	the _history property
Return type:	list

static open_in_browser(response)¶

Provide a simple interface for lxml.html.open_in_browser function. Be careful, use this function only for debug purpose

Parameters:	response –
Returns:

open_url(url, data=None, back=False, **kwargs)¶

Open the given url

Parameters:	url – The url to access data – Data to send. If data is set, the browser will make a POST request back – tell if we actually accessing a page of the history
Returns:	The Response object from requests call

set_headers(headers)¶

Setter for headers property

Parameters:	headers (dict) – a dict containing all headers
Returns:	None

submit_form()¶

Submit the form filled with form_data property dict

Raise:: oct.core.exceptions.NoFormWaiting

Returns:	Response object after the submit

octbrowser.exceptions module¶

exception octbrowser.exceptions.FormNotFoundException¶

Bases: exceptions.Exception

Raised in case of FormNotFound with browser

exception octbrowser.exceptions.LinkNotFound¶

Bases: exceptions.Exception

Raised in case of link not found in current html document

exception octbrowser.exceptions.NoFormWaiting¶

Bases: exceptions.Exception

Raised in case of action required form if no form selected

exception octbrowser.exceptions.NoUrlOpen¶

Bases: exceptions.Exception

Raised in case of no url open but requested inside browser class

exception octbrowser.exceptions.OctGenericException¶

Bases: exceptions.Exception

Provide generic exception for reports