Python PhantomJS.close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 close [as 别名]
    def _init_robot(self, id):
        robot = WDriver()
        logging.debug("initialize")
        self.robots.update({str(id): robot})
        logging.debug("get facebook.com")
        robot.get('http://fb.com')

        logging.debug("login")
        robot.find_element_by_name('email').send_keys('[email protected]')
        robot.find_element_by_name('pass').send_keys('2855930022040')
        robot.find_element_by_name('pass').send_keys(Keys.RETURN)

        for index in range(len(self.remain_ids)):
            self.lock.acquire()
            user_id = self.remain_ids.pop()
            self.lock.release()
            try:
                self.get_name_for_id(robot, user_id)
            except:
                logging.debug("error while updating record with id=%s" % str(user_id))
                self.error_ids.add(user_id)
            else:
                self.done_ids.add(user_id)
        robot.close()
        return

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import close [as 别名]
def main(argv=sys.argv[1:]):
    parser = argparse.ArgumentParser()
    parser.add_argument('--url', default='http://127.0.0.1:8000/static/index.html')
    args = parser.parse_args(argv)

    url = args.url

    browser = WebDriver()
    browser.get(url)
    tags = browser.find_elements_by_css_selector('li')
    for tag in tags:
        print(tag.text)
    browser.close()

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import close [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 close [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 close [as 别名]
                extra_value = any(['% of students' in r for r in r_list])
                if r_list[0] in ('White', 'All students', 'Not low-income'):
                    continue
                df.loc[ind, 'indicator'] = indicator.replace('Equity', '')
                df.loc[ind, 'subject'] = subject
                df.loc[ind, 'category'] = r_list[0]
                df.loc[ind, 'school_score'] = r_list[1] if not extra_value else r_list[2]
                df.loc[ind, 'state_average'] = r_list[2] if not extra_value else r_list[3]
                ind += 1

    df['school_score'] = df['school_score'].str.replace(r'[^0-9]', '').astype('int')
    df['state_average'] = df['state_average'].str.replace(r'[^0-9]', '').astype('int')
    df['test_diff'] = df['school_score'] - df['state_average']
    subject_diffs = df.groupby('subject')['test_diff'].mean()

    output_df.loc[output_ind, 'school'] = school_name
    output_df.loc[output_ind, 'url'] = url
    output_df.loc[output_ind, 'students_per_grade'] = students_per_grade
    output_df.loc[output_ind, 'teachers_to_student'] = t_to_s_school
    output_df.loc[output_ind, 'counselors_to_student'] = c_to_s_school
    if 'Reading' in subject_diffs.index:
        output_df.loc[output_ind, 'reading'] = subject_diffs['Reading']
    if 'Math' in subject_diffs.index:
        output_df.loc[output_ind, 'math'] = subject_diffs['Math']
    if 'Science' in subject_diffs.index:
        output_df.loc[output_ind, 'science'] = subject_diffs['Science']
    output_ind += 1
    print time() - t1

wd.close()

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import close [as 别名]
class WeixinPhantomjs(Base):
    all_uids = {docs['uid'] for docs in in_collection.find({}, {'uid': 1}) if 'uid' in docs}

    def __init__(self):
        self.start_page = START_PAGE
        self.end_page = END_PAGE
        self.weixin_url = REFER_FIRST

        # self.driver = Firefox()
        if hasattr(config, 'PHANTOMJS_PATH'):
            self.driver = PhantomJS(executable_path=getattr(config, 'PHANTOMJS_PATH'))
        else:
            self.driver = PhantomJS()

    def open_weixin_browser(self, word):
        try:
            self.driver.get(self.weixin_url)
            self.driver.set_page_load_timeout(3)

            self.driver.find_element_by_id('upquery').send_keys(word)
            self.driver.find_element_by_class_name('swz').click()
            time.sleep(3)

            urls_uids = self.extract_urls_uids(word=word)
            Article(urls_uids=urls_uids, word=word).extract()
        except Exception as e:
            storage_word.append([word, 0])
            self.logger.info('Open weixin error: type <{}>, mag <{}>'.format(e.__class__, e))
            self.close_browser()
            return True
        return False

    def get_total_pages_to_word(self):
        pages = []
        page_id_css = 'pagebar_container'

        try:
            e = self.driver.find_element_by_id(page_id_css)
            for _p in e.text.split():
                _p = _p.strip()

                if not _p.isdigit():
                    return DEFAULT_PAGES if DEFAULT_PAGES <= pages[-1] else pages[-1]
                else:
                    pages.append(int(_p))
            return 1
        except (NoSuchElementException, NoSuchWindowException, TypeError, IndexError):
            pass
        return 1

    def extract_urls_uids(self, word):
        urls_uids = []
        timestamp = [_t.get_attribute('t') for _t in self.driver.find_elements_by_css_selector('div.s-p')]
        urls_tits = [(t.get_attribute('href'), self.trim(t.text))
                     for t in self.driver.find_elements_by_css_selector('h4 a')]

        if len(urls_tits) != len(timestamp):
            return urls_uids

        for index, url_tit in enumerate(urls_tits):
            try:
                uid = self.md5(timestamp[index] + url_tit[1] + word)

                if uid not in self.__class__.all_uids:
                    self.__class__.all_uids.add(uid)
                    urls_uids.append({'url': url_tit[0], 'uid': uid})
            except (TypeError, IndexError):
                pass
        return urls_uids

    @property
    def is_forbidden(self):
        css_id = 'seccodeForm'

        try:
            if self.driver.find_element_by_id(css_id):
                return True
        except NoSuchElementException:
            pass
        return False

    def appear_element(self, by):
        try:
            # Have `click` function to specified element
            tag = WebDriverWait(self.driver, 20).until(lambda driver: driver.find_element_by_id(by))
            tag.click()
            return True
        except (TimeoutException, NoSuchWindowException, NoSuchElementException):
            pass
        return False

    def crawl_single(self, word=None, go=0):
        is_go = True
        go_page = int(go)
        next_page_css = 'sogou_page_%s'

        is_break = self.open_weixin_browser(word)
        pages = self.get_total_pages_to_word()

        for page in range(self.start_page + 1, (pages or self.end_page) + 1):
#.........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 close [as 别名]
class WeixinPhantomjs(Base):
    def __init__(self):
        self.start_page = START_PAGE
        self.end_page = END_PAGE
        self.weixin_url = REFER_FIRST

        # self.driver = Firefox()
        if hasattr(config, 'PHANTOMJS_PATH'):
            self.driver = PhantomJS(executable_path=getattr(config, 'PHANTOMJS_PATH'))
        else:
            self.driver = PhantomJS()

        self.client = MongoClient(HOST, PORT)
        self.collection = self.client[DB][COLLECTION]
        self.all_uids = self.uids

    def open_weixin_browser(self, word):
        try:
            self.driver.get(self.weixin_url)
            self.driver.set_page_load_timeout(3)

            self.driver.find_element_by_id('upquery').send_keys(word)
            self.driver.find_element_by_class_name('swz').click()
            time.sleep(3)

            urls_uids = self.extract_urls_uids(word=word)
            Article(urls_uids=urls_uids, word=word).extract()
        except Exception as e:
            storage_word.append([word, 0])
            self.logger.info('Open weixin error: type <{}>, mag <{}>'.format(e.__class__, e))
            self.close_browser()
            return True
        return False

    def get_total_pages_to_word(self):
        pages = []
        page_id_css = 'pagebar_container'

        try:
            e = self.driver.find_element_by_id(page_id_css)
            for _p in e.text.split():
                _p = _p.strip()

                if not _p.isdigit():
                    return DEFAULT_PAGES if DEFAULT_PAGES <= pages[-1] else pages[-1]
                else:
                    pages.append(int(_p))
            return 1
        except (NoSuchElementException, NoSuchWindowException, TypeError, IndexError):
            pass
        return 1

    def get_query_words(self, word):
        query_words = []

        for docs in self.collection.find({}, {'rel': 1, 'conp': 1}).sort([('_id', 1)]):
            w = docs['conp']

            if w not in query_words:
                query_words.append(w)

            for item in docs['rel']:
                if item not in query_words:
                    query_words.append(item)

        self.client.close()

        return self.query_index(query_words, word)

    @property
    def uids(self):
        return {docs['uid'] for docs in in_collection.find({}, {'uid': 1}) if 'uid' in docs}

    def extract_urls_uids(self, word):
        urls_uids = []
        timestamp = [_t.get_attribute('t') for _t in self.driver.find_elements_by_css_selector('div.s-p')]
        urls_tits = [(t.get_attribute('href'), self.trim(t.text))
                     for t in self.driver.find_elements_by_css_selector('h4 a')]

        if len(urls_tits) != len(timestamp):
            return urls_uids

        for index, url_tit in enumerate(urls_tits):
            try:
                uid = self.md5(timestamp[index] + url_tit[1] + word)

                if uid not in self.all_uids:
                    self.all_uids.add(uid)
                    urls_uids.append({'url': url_tit[0], 'uid': uid})
            except (TypeError, IndexError):
                pass
        return urls_uids

    @staticmethod
    def query_index(words, cut_word):
        temp_words = words[START_INDEX:END_INDEX]

        try:
            index = temp_words.index(cut_word)
            return temp_words[index:], index + START_INDEX
#.........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 close [as 别名]
def get_url_files(retail, invoice_doc_type, invoice_id, invoice_date, invoice_amount):
    retail_invoice_url = RETAIL_INVOICE_URL[retail]

    driver = PhantomJS()
    driver.get(retail_invoice_url)

    # 1 Set doc_type 'select'
    try:
        select_doc_type = Select(driver.find_element_by_name('txtTipoDte'))
        value = RETAIL_INVOICE_DOC_TYPES[retail][invoice_doc_type]['value']
        select_doc_type.select_by_value(value)
        # name = RETAIL_INVOICE_DOC_TYPES[retail][invoice_doc_type]['name']
        # select_doc_type.select_by_visible_text(name)
    except Exception:
        print 'ERROR: set doc_type select as Boleta'
        driver.save_screenshot('screen.png')
        return '', ''

    time.sleep(5)

    # 2 Get recaptcha img url
    try:
        recaptcha_img = driver.find_element_by_id('recaptcha_challenge_image')
        recaptcha_img_url = recaptcha_img.get_attribute('src')
    except Exception:
        print 'ERROR: get recaptcha image url'
        driver.save_screenshot('screen.png')
        return '', ''

    # 3 Solve recaptcha
    v = VisionApi()
    recaptcha_value = v.detect_text_from_url(recaptcha_img_url)

    if recaptcha_value is None:
        print 'ERROR: solving recaptcha image'
        driver.save_screenshot('screen.png')
        return '', ''

    # 4 Fill form
    script = u"""
        document.getElementsByName('txtFolio')[0].value = '{invoice_id}';
        document.getElementsByName('txtFechaEmision')[0].value = '{invoice_date}';
        document.getElementsByName('txtMontoTotal')[0].value = '{invoice_amount}';
        document.getElementsByName('recaptcha_response_field')[0].value = '{recaptcha_value}';
    """.format(
        invoice_id=invoice_id,
        invoice_date=invoice_date,
        invoice_amount=invoice_amount,
        recaptcha_value=recaptcha_value,
    )
    driver.execute_script(script)

    # 5 Submit form
    try:
        driver.find_element_by_name('frmDatos').submit()
    except Exception:
        print 'ERROR: submitting form'
        driver.save_screenshot('screen.png')
        return '', ''

    # 6 Get url files
    try:
        xml_a_tag = driver.find_element_by_xpath('//*[@id="Tabla_01"]/tbody/tr[1]/td[2]/p/a[2]')
        pdf_a_tag = driver.find_element_by_xpath('//*[@id="Tabla_01"]/tbody/tr[1]/td[2]/p/a[1]')

        xml_url = xml_a_tag.get_attribute('href')
        pdf_url = pdf_a_tag.get_attribute('href')
    except Exception:
        print 'ERROR: getting url files'
        driver.save_screenshot('screen.png')
        return '', ''

    # 8 Delete driver session
    driver.close()
    driver.quit()

    return xml_url, pdf_url