get-pybrowser 0.2.0

Selenium based, user friendly Browser Automation API

pybrowser

For documentation, refer here

Here is a quick example,

from pybrowser import Browser 
    with Browser(browser_name=Browser.CHROME) as b:
        b.goto("https://www.google.com/")
        b.input("name:=q").enter("news")
        b.button("name:=btnK").click()
        screenshot_path = b.take_screenshot()
        print(b.html().elements.links())

Release Notes:

0.2.0

• Changes

  • Changes in underlying Browser.requests_session object,

    • requests_session.result has been removed. There is just requests_session.response which is a blocking call (in case asynch flag is set). Also note, requests_session.is_request_done is still available to see if request is complete of not. There are no changes to other properties and are blocking in case of asynchronous call.

    • requests_session.content() method with options for bytes and text has been changed to properties, just like in underlying requests. That is now you call, requests_session.content for bytes and requests_session.text for text.

    • In context of above changes, Browser object also has below changes,

      • Browser.get and Browser.post now returns requests_session object (used to be requests_session.response).
      • Browser.content() has been changed to properties. That is now you call, Browser.content for bytes and Browser.text for text.

• New features

  • Support for remote url. Please note this requires Selenium Grid to be setup explicitly. Once done use the URL here.

  • Flags for Opera browser (Browser.OPERA). Webdriver executable needs to be present in driver_path. Please note EDGE and SAFARI are also supported the same way. That is, webdriver isn't automatically downloaded, instead path needs to be provided.

0.1.0

• Changes

  Please note some key changes with regards to properties changing to methods.

  • Methods which were properties in initial release are now method calls. Below is the impacted list.

    In Browser,

    • Browser.refresh()
    • Browser.back()
    • Browser.forward()
    • Browser.maximize_window()
    • Browser.minimize_window()
    • Browser.fullscreen_window()
    • Browser.delete_all_cookies()
    • Browser.close()

    In Action (used by all elements),

    • Action.refresh()
    • Action.highlight()
    • Action.double_click()
    • Action.move_to_element()

    Specific elements also,

    • Checkbox.check()
    • Checkbox.uncheck()
    • Radio.select()
    • Radio.unselect()
    • Input.clear()

  • Browser class __init__ method now has more options,

    • firefox_binary_path
    • firefox_profile_path
    • http_proxy
    • driver_path

  • Select element has below method changes,

    Select.options(get="text") method has been split to multiple properties to keep it simple,

    • Select.options_text
    • Select.options_value
    • Select.options_element

    Similar change has been done for Select.all_selected_options(get="text"),

    • Select.all_selected_options_text
    • Select.all_selected_options_value
    • Select.all_selected_options_element

  • File element changes,

    Below enhancements have been made to download feature,

    • Added more parameters -

      • unzip - Set this flag to unzip file. Default is False
      • del_zipfile - Set this flag to delete zip file after it has been unzipped. Default is False
      • add_to_ospath - Set this flag to add directory to PATH. Default is False

    • New properties -

      As you might already know, download happens in background (asynchronous) by default and can of course be changed with asynch. To check if download was successful, below properties are available,

      • is_download_complete - True or False
      • downloaded_files - list of downloaded files

  • HTML links method has below method changes,

    This is the one you would invoke via Browser as Browser.html().elements.link().

    • Added more parameters -

      • images - You can filter out images. Default is False that means to include images, you will need to set this to True
      • name_length_limit - This limits the length of name of the url. Default is 60

    • Change in return type. Before this change, return type was a list of types (name, url). This has been changed to list of named_tuples of form ('Link', ['name', 'url'])

  • Changes in some env options.

    Removed below options as they don't have sense. Either you provide complete driver download url or provide version so that the API tries to download,

    • CHROME_FILENAME
    • IE_FILENAME
    • CHROMEDRIVER_DEFAULT_VERSION

    Added below for Firefox support,

    • FIREFOX_HOME_URL
    • FIREFOX_DOWNLOAD_URL
    • FIREFOXDRIVER_VERSION

  • Refactoring (non functional) in code and tests. Lot more testing needed still :-\

• New features

  • Support for Firefox !

0.0.1

• Very first release !