Python PhantomJS.set_window_size方法代码示例

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_window_size [as 别名]
def generate_image(structure):
    image_path = os.path.join(mkdtemp(), 'okc.png')
    html_path = os.path.join(
        os.path.dirname(os.path.realpath(__file__)),
        'okc.html',
    )
    url = 'file://{}'.format(html_path)
    driver = PhantomJS(service_log_path=mkstemp()[1])
    driver.set_window_size(2000, 500)
    driver.get(url)
    driver.execute_script('setText({});'.format(json.dumps(structure)))

    if random() > 0.4:
        driver.execute_script('hideForm();')
    elif random() > 0.5:
        driver.execute_script('uncheckForm();')

    driver.set_window_size(*driver.execute_script('return getSize();'))
    driver.save_screenshot(image_path)

    # twitter's gonna make our beautiful screenshot a jpeg unless we make it
    # think that we're using transparency for a reason, so,,
    img = Image.open(image_path)
    origin = img.getpixel((0, 0))
    new_origin = origin[:3] + (254,)
    img.putpixel((0, 0), new_origin)
    img.save(image_path)

    subprocess.check_call(['optipng', '-quiet', image_path])

    return image_path

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_window_size [as 别名]
 def post(self):
     id = request.values['page']
     page = Page.objects.get_or_404(id=id)
     # html = requests.get(page.baseurl).text
     screenshot = None
     try:
         phantom = PhantomJS(desired_capabilities={'acceptSslCerts': True},
                             service_args=['--web-security=false',
                                           '--ssl-protocol=any',
                                           '--ignore-ssl-errors=true'], port=8888)
         phantom.set_window_size(1024, 768)
         phantom.get(page.baseurl)
         html = phantom.page_source
         screenshot = phantom.get_screenshot_as_png()
         phantom.close()
     except Exception as ex:
         html = "error when i snap your page ... %s" % ex
         snap = Snap(html, datetime.datetime.now(), screenshot).save()
         page.update(push__snaps=snap)
     snap = Snap(html, datetime.datetime.now(), screenshot).save()
     page.update(push__snaps=snap)
     return jsonify({'id': "%s" % snap.id})

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_window_size [as 别名]
def get_ratio_data():
    import socket
    import re
    import time
    import dryscrape
    import webkit_server
    from random import randint
    from fake_useragent import UserAgent
    from bs4 import BeautifulSoup
    from selenium.webdriver import PhantomJS


    from app.models import Company, Indicators
    from app.utils import cash_to_float, depercentize


    # Dict item with list: element attribute, attribute value to look for, optional transform function
    indicators = {'roe': {
                          'attribute': 'data-reactid',
                          'value': re.compile(".*RETURN_ON_EQUITY\.1$"),
                          'transform': depercentize,
                          },
                  'fcf': {
                          'attribute': 'data-reactid',
                          'value': re.compile(".*LEVERED_FREE_CASH_FLOW\.1$"),
                          'transform': cash_to_float,
                          },
                  }


    ua = UserAgent()


    #with open("10.csv", "r") as f:
    #with open("sp500-2.csv", "r") as f:
    with open("10_stocks", "r") as f:
        data = f.read()

    symbols = []
    for i in data.split("\n"):
        if i:

            symbols.append(i.split(",")[0])

    print("Iterate through symbols")
    ## dryscrape
    #session = dryscrape.Session()
    #session.set_header('User-Agent', ua.random)
    #session.set_timeout(5)

    for symbol in symbols:
        print("{} Fetching {}  :".format(time.strftime("%H:%M:%S"), symbol))

        import pdb; pdb.set_trace()
        #driver = MyWebDriver()
        driver = PhantomJS()
        driver.set_window_size(1120, 550)
        driver.get("http://finance.yahoo.com/quote/{}/key-statistics?p={}".format(symbol, symbol))

        ##try:
        ##    session = dryscrape.Session()
        ##except socket.error as e:
        ##    print("Failed to configure session {}".format(e))
        ##    continue

        ##session.set_header('User-Agent', ua.random)
        ##session.set_timeout(30)
        #try:
        #    #session.visit("http://finance.yahoo.com/quote/{}/key-statistics?p={}".format(symbol, symbol))
        #except Exception as e:
        #    print e, "try once more......"
        #    session.reset()
        #    time.sleep(5)
        #    session = dryscrape.Session()
        #    #session.set_header('User-Agent', ua.random)
        #    try:
        #        session.set_timeout(5)
        #        session.visit("http://finance.yahoo.com/quote/{}/key-statistics?p={}".format(symbol, symbol))
        #    except Exception as e:
        #        print e, "done trying..."
        #        session.reset()
        #        time.sleep(2)
        #        session = dryscrape.Session()
        #        continue


        #except socket.error as e:
        #    print("Failed to get {}, {} (1)".format(symbol, e))
        #    continue
        #except webkit_server.EndOfStreamError as e:
        #    print("Failed to get {}, {}, breaking (2)".format(symbol, e))
        #    continue
        #except webkit_server.InvalidResponseError as e:
        #    print("Failed to get {}, {}, breaking (3)".format(symbol, e))
        #    continue

        #response = session.body()
        #soup = BeautifulSoup(response, "lxml")

        with open("{}.out".format(symbol), "w") as f:
#.........这里部分代码省略.........

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_window_size [as 别名]
class Client(object):
    """Client HTTP pour tester fonctionnellement Strass

    Adapteur du pilote Selenium, avec une interface inspirée de Nightwatch.js,
    et quelques paramètres spécifiques à Strass."""

    def __init__(self):
        self.driver = PhantomJS()
        self.driver.set_window_size(1120, 550)

    def __del__(self):
        self.driver.quit()

    def get(self, query=None):
        server = os.environ.get('STRASS_TEST_SERVER', 'http://localhost:8000')
        url = server + (query or '/')
        self.driver.get(url)
        return self

    def find(self, selector):
        return self.driver.find_element_by_css_selector(selector)

    def click(self, selector):
        self.find(selector).click()
        return self

    def fill(self, selector, value):
        if isinstance(value, datetime.date):
            self.fill(selector + ' input.day', str(value.day))
            self.fill(selector + ' input.month', str(value.month))
            self.fill(selector + ' input.year', str(value.year))
        else:
            control = self.find(selector)
            try:
                control.clear()
            except selexc.InvalidElementStateException:
                # On doit tenter de nettoyer un input[type=file]. On zap.
                pass
            control.send_keys(value)
        return self

    def select(self, selector, value):
        Select(self.find(selector)).select_by_value(value)
        return self

    def submit(self, selector='#document button[type=submit]'):
        return self.click(selector)

    def close(self):
        self.driver.close()
        if self.driver.window_handles:
            self.driver.switch_to.window(self.driver.window_handles[0])
        self.driver.set_window_size(1120, 550)
        return self

    def screenshot(self, filename):
        self.driver.get_screenshot_as_file(filename)
        sys.stderr.write("Capture d'écran enregistrée dans %r\n" % (filename,))
        return self

    def save(self, filename):
        with open(filename, 'w') as fo:
            fo.write(self.driver.page_source)
        sys.stderr.write("HTML enregistré dans %r\n" % (filename,))
        return self

    def __getattr__(self, name):
        return getattr(self.driver, name)

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_window_size [as 别名]
class LegacySensCritique(object):

    CHANGEPAGE_TIMEOUT = 20

    '''
    Interact with SensCritique website
    '''

    def __init__(self, login, password, userAgent=LINUX_USER_AGENT):
        '''
        Constructor

        :param login:
        :param password:
        '''

        self.login = login
        self.password = password

        dcap = dict(DesiredCapabilities.PHANTOMJS)
        dcap["phantomjs.page.settings.userAgent"] = (
            userAgent
        )
        self.driver = PhantomJS(desired_capabilities=dcap)
        self.driver.set_window_size(1366, 768)

    def sign_in(self):
        '''
        Sign-in to SensCritique using the given login details

        :rtype: bool
        :Return: true if login succeeded, false otherwise
        '''

        self.to(HomePage())

        self.page.alreadySuscribed().click()

        self.page.loginField().send_keys(self.login)
        self.page.passwordField().send_keys(self.password)

        self.page.submitLoginButton().click()

        #TODO changing page so wait or something
        currentUser = self.page.username(self.CHANGEPAGE_TIMEOUT)

        if currentUser is not None:
            self._currentUsername = currentUser.value()
            logging.warn("Logged in with user " + self._currentUsername)

            return True
        else:
            if self.page.loginError() is not None:
                logging.error("Couldn't login : " + self.page.loginError().value())

            return False

    def is_logged_in(self):
        return self.page is not None and self.page.username() is not None

    def retrieveListById(self, listId):
        self.to(ListCollectionPage(self._currentUsername))

        for l in self.page.lists():
            if listId in l.url():
                return self.createSCListFromListModule(l)

        return None

    def retrieveListByTitle(self, title):
        self.to(ListCollectionPage(self._currentUsername))

        for l in self.page.lists():
            if l.title() == title:
                return self.createSCListFromListModule(l)

        return None

    def retrieveMoviesFromList(self, l : SCList):
        self.to(ListPage(l))

        for movie in self.page.movies():
            yield movie

    def createList(self, l : SCList):
        self.to(ListCollectionPage(self._currentUsername))

        self.page.create_list_button().click()

        self.page.new_list_title().send_keys(l.title())
        self.page.film_type_radio().click()
        self.page.classic_list_radio().click()
        self.page.public_list_radio().click()

        self.page.confirm_create_list_button().click()

        # Change the current page as we are now on the list page
        self.page = ListPage(l)
        self.page._driver = self.driver  # TODO: fixme, we don't want to use self.to(page) as it would reload the page

#.........这里部分代码省略.........

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_window_size [as 别名]
def get_ratio_data():
    import re
    import time
    from bs4 import BeautifulSoup

    from app.models import Company, Indicators
    from app.utils import cash_to_float, depercentize


    # Dict item with list: element attribute, attribute value to look for, optional transform function
    indicators = {'roe': {
                          'attribute': 'data-reactid',
                          'value': re.compile(".*RETURN_ON_EQUITY\.1$"),
                          'transform': depercentize,
                          },
                  'fcf': {
                          'attribute': 'data-reactid',
                          'value': re.compile(".*LEVERED_FREE_CASH_FLOW\.1$"),
                          'transform': cash_to_float,
                          },
                  'ev2ebitda': {
                          'attribute': 'data-reactid',
                          'value': re.compile(".*ENTERPRISE_VALUE_TO_EBITDA\.1$"),
                          },
                  }


    companies = Company.query.with_entities(Company.symbol).all()
    symbols = [company[0] for company in companies]

    print("Iterate through symbols")

    for symbol in symbols:
        print("{} Fetching {}  :".format(time.strftime("%H:%M:%S"), symbol))

        #driver = MyWebDriver()
        retry_current = 0
        retry_limit = 5
        while retry_current < retry_limit:
            try:
                driver = PhantomJS()
            except URLError:
                time.sleep(retry_current**2)
            retry_current += 1

        driver.set_window_size(1120, 550)
        driver.get("http://finance.yahoo.com/quote/{}/key-statistics?p={}".format(symbol, symbol))
        try:
            #WebDriverWait(driver, 10).until(EC.title_contains("AAPL Key Statistics | Apple Inc. Stock - Yahoo Finance"))
            #element = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.CSS_SELECTOR, "td[reactid]")))
            #element = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.XPATH, "//td[@data-reactid[ends-with(., 'RETURN_ON_EQUITY.1')]]")))

            # these two seem to work...
            element = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.XPATH, "//*[substring(@data-reactid, string-length(@data-reactid) - string-length('RETURN_ON_EQUITY.1') +1) = 'RETURN_ON_EQUITY.1']")))
            #element = WebDriverWait(driver, 10).until(EC.presence_of_element_located((By.XPATH, "//*[contains(@data-reactid,'RETURN_ON_EQUITY.1')]")))

            #"//input[@id[ends-with(.,'register')]]"
        except TimeoutException as e:
            print  "Caught", e
            print driver.title
            continue

        #time.sleep(5)

        #with open("{}.out".format(symbol), "w") as f:
        #    f.write(driver.page_source.encode('utf-8'))

        soup = BeautifulSoup(driver.page_source, "lxml")

        d = {'symbol': symbol}
        for indicator in indicators.keys():
            curr_ind = indicators[indicator]
            s = soup.find_all(attrs={curr_ind['attribute']: curr_ind['value']})
            print indicator, s

            for element in s:
                if curr_ind.has_key('transform'):
                    f = curr_ind['transform']
                    #print(f(element.text))
                    d[indicator] = f(element.text)
                else:
                    #print(element.text)
                    d[indicator] = element.text

        try:
            db.session.add(Indicators.from_json(d))
            db.session.commit()
        except (IntegrityError, UnmappedInstanceError) as e:
            print "Caught", e
            db.session.rollback()

        print "indicators", d

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_window_size [as 别名]
import os

from selenium.common.exceptions import NoSuchElementException
from selenium.webdriver import PhantomJS

ckan_user = os.environ['ckan_user']
ckan_password = os.environ['ckan_password']
ckan_api = os.environ['ckan_api']
ckan_url = 'http://127.0.0.1:5000'
headers = {
    'Authorization': 'token {0}'.format(ckan_api),
    'content-type': 'application/json'
}

phantom = PhantomJS(service_args=['--proxy-type=none', '--load-images=false'])
phantom.set_window_size(1024, 768)


def login_or_pass(driver):
    try:
        driver.find_element_by_class_name('username')
    except NoSuchElementException:
        driver.get(ckan_url)
        driver.find_element_by_link_text('Log in').click()
        driver.find_element_by_name('login').send_keys(ckan_user)
        driver.find_element_by_name('password').send_keys(ckan_password)
        driver.find_element_by_xpath(
            '//*[@id="wb-auto-1"]/div/div/div/form/div[4]/button').click()
        driver.find_element_by_class_name('username')
    else:
        return driver