Initial commit

This commit is contained in:
Alex committed 2026-11-02 22:54:14 +08:00
commit 57fc3525eb
7311 files changed
+1291398

No files matched your search

@@ -0,0 +1,950 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from json import loads, JSONDecodeError
from os.path import sep
from pathlib import Path
from re import findall
from threading import Thread
from time import perf_counter, sleep
from DrissionRecord.tools import make_valid_name
from .._base.base import BasePage
from .._elements.chromium_element import run_js, make_chromium_eles
from .._elements.none_element import NoneElement
from .._elements.session_element import make_session_ele
from .._functions.cookies import CookiesList
from .._functions.elements import SessionElementsList, get_frame, ChromiumElementsList
from .._functions.locator import get_loc
from .._functions.settings import Settings as _S
from .._functions.tools import raise_error
from .._functions.web import location_in_viewport
from .._units.actions import Actions
from .._units.console import Console
from .._units.listener import Listener
from .._units.rect import TabRect
from .._units.screencast import Screencast
from .._units.scroller import PageScroller
from .._units.setter import ChromiumBaseSetter
from .._units.states import PageStates
from .._units.waiter import BaseWaiter
from ..errors import (ContextLostError, CDPError, PageDisconnectedError, ElementLostError, JavaScriptError,
BrowserConnectError, LocatorError)
__ERROR__ = 'error'
class ChromiumBase(BasePage):
def __init__(self, browser, target_id=None):
super().__init__()
self._browser = browser
self._is_loading = None
self._root_id = None # object id
self._set = None
self._screencast = None
self._actions = None
self._states = None
self._has_alert = False
self._ready_state = None
self._rect = None
self._wait = None
self._scroll = None
self._console = None
self._upload_list = None
self._doc_got = False # 用于在LoadEventFired和FrameStoppedLoading间标记是否已获取doc
self._auto_handle_alert = None
self._load_end_time = 0
self._init_jss = []
self._disconnect_flag = False
self._type = 'ChromiumBase'
if not hasattr(self, '_listener'):
self._listener = None
self._d_set_runtime_settings()
self._connect_browser(target_id)
def __call__(self, locator, index=1, timeout=None):
return self.ele(locator, index, timeout)
def _d_set_runtime_settings(self):
pass
def _connect_browser(self, target_id=None):
self._is_reading = False
if not target_id:
if self.browser._ws_only:
tabs = self._run_cdp('Target.getTargets')['targetInfos']
_id = 'targetId'
else:
tabs = self.browser._driver.get(f'http://{self.browser.address}/json').json()
_id = 'id'
tabs = [(i[_id], i['url']) for i in tabs
if i['type'] in ('page', 'webview') and not i['url'].startswith('devtools://')]
dialog = None
if len(tabs) > 1:
for k, t in enumerate(tabs):
if t[1] == 'chrome://privacy-sandbox-dialog/notice':
dialog = k
elif not target_id:
target_id = t[0]
if target_id and dialog is not None:
break
if dialog is not None:
close_privacy_dialog(self, tabs[dialog][0])
else:
target_id = tabs[0][0]
self._driver_init(target_id)
if self._js_ready_state == 'complete' and self._ready_state is None:
self._get_document()
self._ready_state = 'complete'
def _driver_init(self, target_id):
self._is_loading = True
self._driver = self.browser._get_driver(target_id, self)
self._alert = Alert(self._auto_handle_alert)
self._driver.set_callback('Page.javascriptDialogOpening', self._on_alert_open, immediate=True)
self._driver.set_callback('Page.javascriptDialogClosed', self._on_alert_close)
self._driver.run('Page.enable')
self._driver.run('DOM.enable')
self._driver.run('Emulation.setFocusEmulationEnabled', enabled=True)
r = self._run_cdp('Page.getFrameTree')
for i in findall(r"'id': '(.*?)'", str(r)):
self.browser._frames[i] = self.tab_id
if not hasattr(self, '_frame_id'):
self._frame_id = r['frameTree']['frame']['id']
self._driver.set_callback('Page.frameStartedLoading', self._onFrameStartedLoading)
self._driver.set_callback('Page.frameNavigated', self._onFrameNavigated)
self._driver.set_callback('Page.domContentEventFired', self._onDomContentEventFired)
self._driver.set_callback('Page.loadEventFired', self._onLoadEventFired)
self._driver.set_callback('Page.frameStoppedLoading', self._onFrameStoppedLoading)
self._driver.set_callback('Page.frameAttached', self._onFrameAttached)
self._driver.set_callback('Page.frameDetached', self._onFrameDetached)
def _get_document(self, timeout=10):
if self._is_reading:
return
self._is_reading = True
timeout = max(timeout, 2)
end_time = perf_counter() + timeout
while perf_counter() < end_time:
try:
b_id = self._run_cdp('DOM.getDocument', _timeout=timeout)['root']['backendNodeId']
timeout = end_time - perf_counter()
timeout = 1 if timeout <= 1 else timeout
self._root_id = self._run_cdp('DOM.resolveNode', backendNodeId=b_id,
_timeout=timeout)['object']['objectId']
result = True
break
except PageDisconnectedError:
result = False
break
except:
timeout = end_time - perf_counter()
timeout = .5 if timeout <= 0 else timeout
sleep(.05)
else:
result = False
if result:
r = self._run_cdp('Page.getFrameTree', _ignore=PageDisconnectedError)
for i in findall(r"'id': '(.*?)'", str(r)):
self.browser._frames[i] = self.tab_id
self._is_loading = False
self._is_reading = False
return result
def _onFrameDetached(self, **kwargs):
self.browser._frames.pop(kwargs['frameId'], None)
def _onFrameAttached(self, **kwargs):
self.browser._frames[kwargs['frameId']] = self.tab_id
def _onFrameStartedLoading(self, **kwargs):
self.browser._frames[kwargs['frameId']] = self.tab_id
if kwargs['frameId'] == self._frame_id:
self._doc_got = False
self._ready_state = 'connecting'
self._is_loading = True
self._load_end_time = perf_counter() + self.timeouts.page_load
if self._load_mode == 'eager':
t = Thread(target=self._wait_to_stop)
t.daemon = True
t.start()
def _onFrameNavigated(self, **kwargs):
if kwargs['frame']['id'] == self._frame_id:
self._doc_got = False
self._ready_state = 'loading'
self._is_loading = True
if kwargs.get('type', None) == 'BackForwardCacheRestore':
self._get_document()
def _onDomContentEventFired(self, **kwargs):
if self._load_mode == 'eager':
self._run_cdp('Page.stopLoading')
if self._get_document(self._load_end_time - perf_counter() - .1):
self._doc_got = True
self._ready_state = 'interactive'
def _onLoadEventFired(self, **kwargs):
if self._doc_got is False and self._get_document(self._load_end_time - perf_counter() - .1):
self._doc_got = True
self._ready_state = 'complete'
def _onFrameStoppedLoading(self, **kwargs):
self.browser._frames[kwargs['frameId']] = self.tab_id
if kwargs['frameId'] == self._frame_id:
if self._doc_got is False:
self._get_document(self._load_end_time - perf_counter() - .1)
self._ready_state = 'complete'
def _onFileChooserOpened(self, **kwargs):
if self._upload_list:
if 'backendNodeId' not in kwargs:
raise RuntimeError(_S._lang.join(_S._lang.CANNOT_INPUT_FILE))
files = self._upload_list if kwargs['mode'] == 'selectMultiple' else self._upload_list[:1]
self._run_cdp('DOM.setFileInputFiles', files=files, backendNodeId=kwargs['backendNodeId'])
self.driver.set_callback('Page.fileChooserOpened', None)
self._run_cdp('Page.setInterceptFileChooserDialog', enabled=False)
self._upload_list = None
def _wait_to_stop(self):
end_time = perf_counter() + self.timeouts.page_load
while perf_counter() < end_time:
sleep(.02)
if self._ready_state in ('interactive', 'complete') and self._is_loading:
self.stop_loading()
# ----------挂件----------
@property
def wait(self):
if self._wait is None:
self._wait = BaseWaiter(self)
return self._wait
@property
def set(self):
if self._set is None:
self._set = ChromiumBaseSetter(self)
return self._set
@property
def screencast(self):
if self._screencast is None:
self._screencast = Screencast(self)
return self._screencast
@property
def actions(self):
if self._actions is None:
self._actions = Actions(self)
self.wait.doc_loaded()
return self._actions
@property
def listen(self):
if self._listener is None:
self._listener = Listener(self)
return self._listener
@property
def states(self):
if self._states is None:
self._states = PageStates(self)
return self._states
@property
def scroll(self):
self.wait.doc_loaded()
if self._scroll is None:
self._scroll = PageScroller(self)
return self._scroll
@property
def rect(self):
# self.wait.doc_loaded()
if self._rect is None:
self._rect = TabRect(self)
return self._rect
@property
def console(self):
if self._console is None:
self._console = Console(self)
return self._console
@property
def timeout(self):
return self._timeouts.base
@property
def timeouts(self):
return self._timeouts
# ----------挂件结束----------
@property
def browser(self):
return self._browser
@property
def driver(self):
if self._driver is None:
raise BrowserConnectError(_S._lang.BROWSER_DISCONNECTED)
return self._driver
@property
def title(self):
return self._run_cdp_loaded('Target.getTargetInfo', targetId=self._target_id)['targetInfo']['title']
@property
def url(self):
return self._run_cdp_loaded('Target.getTargetInfo', targetId=self._target_id)['targetInfo']['url']
@property
def _browser_url(self):
return self.url
@property
def html(self):
self.wait.doc_loaded()
return self._run_cdp('DOM.getOuterHTML', objectId=self._root_id)['outerHTML']
@property
def json(self):
try:
return loads(self('t:pre', timeout=.5).text)
except JSONDecodeError:
return None
@property
def tab_id(self):
return self.driver.id
@property
def _target_id(self):
return self.driver.id
@property
def active_ele(self):
return self._run_js_loaded('return document.activeElement;')
@property
def load_mode(self):
return self._load_mode
@property
def user_agent(self):
return self._run_cdp('Runtime.evaluate', expression='navigator.userAgent;')['result']['value']
@property
def upload_list(self):
return self._upload_list
@property
def session(self):
if self._session is None:
self._create_session()
return self._session
@property
def _js_ready_state(self):
try:
return self._run_cdp('Runtime.evaluate', expression='document.readyState;', _timeout=3)['result']['value']
except ContextLostError:
return None
except TimeoutError:
return 'timeout'
def run_cdp(self, cmd, **cmd_args):
r = self.driver.run(cmd, **cmd_args)
return r if __ERROR__ not in r else raise_error(r, self.browser, user=True)
def run_cdp_loaded(self, cmd, **cmd_args):
self.wait.doc_loaded()
r = self.driver.run(cmd, **cmd_args)
return r if __ERROR__ not in r else raise_error(r, self.browser, user=True)
def _run_cdp(self, cmd, **cmd_args):
ignore = cmd_args.pop('_ignore', None)
r = self.driver.run(cmd, **cmd_args)
return r if __ERROR__ not in r else raise_error(r, self.browser, ignore)
def _run_cdp_loaded(self, cmd, **cmd_args):
self.wait.doc_loaded()
return self._run_cdp(cmd, **cmd_args)
def run_js(self, script, *args, as_expr=False, timeout=None):
return self._run_js(script, *args, as_expr=as_expr, timeout=timeout)
def run_js_loaded(self, script, *args, as_expr=False, timeout=None):
self.wait.doc_loaded()
return self._run_js(script, *args, as_expr=as_expr, timeout=timeout)
def _run_js(self, script, *args, as_expr=False, timeout=None):
return run_js(self, script, as_expr, self.timeouts.script if timeout is None else timeout, args)
def _run_js_loaded(self, script, *args, as_expr=False, timeout=None):
self.wait.doc_loaded()
return run_js(self, script, as_expr, self.timeouts.script if timeout is None else timeout, args)
def run_async_js(self, script, *args, as_expr=False):
run_js(self, script, as_expr, 0, args)
def get(self, url, show_errmsg=False, retry=None, interval=None, timeout=None):
retry, interval, is_file = self._before_connect(url, retry, interval)
self._url_available = self._d_connect(self._url, times=retry, interval=interval,
show_errmsg=show_errmsg, timeout=timeout)
return self._url_available
def cookies(self, all_domains=False, all_info=False):
txt = 'Storage' if all_domains else 'Network'
cookies = self._run_cdp_loaded(f'{txt}.getCookies')['cookies']
if all_info:
r = cookies
else:
r = [{'name': cookie['name'], 'value': cookie['value'], 'domain': cookie['domain']} for cookie in cookies]
return CookiesList(r)
def ele(self, locator, index=1, timeout=None):
return self._ele(locator, timeout=timeout, index=index, method='ele()')
def eles(self, locator, timeout=None):
return self._ele(locator, timeout=timeout, index=None)
def s_ele(self, locator=None, index=1, timeout=None):
if timeout is None:
timeout = self.timeout
return (NoneElement(self, method='s_ele()', args={'locator': locator, 'index': index, 'timeout': timeout})
if locator and not self.wait.eles_loaded(locator, timeout=timeout)
else make_session_ele(self, locator, index=index, method='s_ele()'))
def s_eles(self, locator, timeout=None):
return (make_session_ele(self, locator, index=None)
if self.wait.eles_loaded(locator, timeout=timeout) else SessionElementsList())
def _find_elements(self, locator, timeout, index=1, relative=False, raise_err=None):
if isinstance(locator, (str, tuple)):
loc = get_loc(locator)[1]
elif locator._type in ('ChromiumElement', 'ChromiumFrame'):
return locator
else:
raise LocatorError(_S._lang.join(ALLOW_TYPE=_S._lang.ELE_OR_LOC, CURR_VAL=locator))
self.wait.doc_loaded()
end_time = perf_counter() + timeout
search_ids = []
timeout = .5 if timeout <= 0 else timeout
result = self.driver.run('DOM.performSearch', query=loc, _timeout=timeout, includeUserAgentShadowDOM=True)
if not result or __ERROR__ in result:
num = 0
else:
num = result['resultCount']
search_ids.append(result['searchId'])
while True:
if num > 0:
from_index = index_arg = 0
if index is None:
end_index = num
index_arg = None
elif index < 0:
from_index = index + num
end_index = from_index + 1
else:
from_index = index - 1
end_index = from_index + 1
if from_index <= num - 1:
nIds = self._driver.run('DOM.getSearchResults', searchId=result['searchId'],
fromIndex=from_index, toIndex=end_index)
if __ERROR__ not in nIds:
if nIds['nodeIds'][0] != 0:
r = make_chromium_eles(self, _ids=nIds['nodeIds'], index=index_arg,
is_obj_id=False, ele_only=True)
if r is not False:
break
elif nIds[__ERROR__] == 'connection disconnected':
raise PageDisconnectedError
if perf_counter() >= end_time:
return NoneElement(self) if index is not None else ChromiumElementsList(owner=self)
sleep(.01)
timeout = end_time - perf_counter()
timeout = .5 if timeout <= 0 else timeout
result = self.driver.run('DOM.performSearch', query=loc, _timeout=timeout, includeUserAgentShadowDOM=True)
if result and __ERROR__ not in result:
num = result['resultCount']
search_ids.append(result['searchId'])
elif result and result[__ERROR__] == 'connection disconnected':
raise PageDisconnectedError
for _id in search_ids:
self._driver.run('DOM.discardSearchResults', searchId=_id)
return r
def refresh(self, ignore_cache=False):
self._is_loading = True
self._run_cdp('Page.reload', ignoreCache=ignore_cache)
self.wait.load_start()
def forward(self, steps=1):
self._forward_or_back(steps)
def back(self, steps=1):
self._forward_or_back(-steps)
def _forward_or_back(self, steps):
if steps == 0:
return
history = self._run_cdp('Page.getNavigationHistory')
index = history['currentIndex']
history = history['entries']
direction = 1 if steps > 0 else -1
curr_url = history[index]['url']
nid = None
for num in range(abs(steps)):
for i in history[index::direction]:
index += direction
if i['url'] != curr_url:
nid = i['id']
curr_url = i['url']
break
if nid:
self._is_loading = True
self._run_cdp('Page.navigateToHistoryEntry', entryId=nid)
def stop_loading(self):
try:
self._run_cdp('Page.stopLoading')
end_time = perf_counter() + 5
while self._ready_state != 'complete' and perf_counter() < end_time:
sleep(.02)
except (PageDisconnectedError, CDPError):
pass
finally:
self._ready_state = 'complete'
def remove_ele(self, loc_or_ele):
if not loc_or_ele:
return
ele = self._ele(loc_or_ele, raise_err=False)
if ele:
self._run_cdp('DOM.removeNode', nodeId=ele._node_id, _ignore=ElementLostError)
def add_ele(self, html_or_info, insert_to=None, before=None):
if isinstance(html_or_info, str):
insert_to = self.ele(insert_to) if insert_to else self.ele('t:body')
args = [html_or_info, insert_to]
if before:
args.append(insert_to.ele(before))
js = '''
ele = document.createElement(null);
arguments[1].insertBefore(ele, arguments[2]);
ele.outerHTML = arguments[0];
return arguments[2].previousElementSibling;
'''
else:
js = '''
ele = document.createElement(null);
arguments[1].appendChild(ele);
ele.outerHTML = arguments[0];
return arguments[1].lastElementChild;
'''
elif isinstance(html_or_info, tuple):
args = [html_or_info[0], html_or_info[1]]
txt = ''
if insert_to:
args.append(self.ele(insert_to))
if before:
args.append(self.ele(before))
txt = '''
arguments[2].insertBefore(ele, arguments[3]);
'''
else:
txt = '''
arguments[2].appendChild(ele);
'''
js = f'''
ele = document.createElement(arguments[0]);
for(let k in arguments[1]){{
if(k=="innerHTML"){{ele.innerHTML=arguments[1][k]}}
else if(k=="innerText"){{ele.innerText=arguments[1][k]}}
else{{ele.setAttribute(k, arguments[1][k]);}}
}}
{txt}
return ele;
'''
else:
raise ValueError(_S._lang.join(_S._lang.INCORRECT_VAL_, 'html_or_info', ALLOW_TYPE='html, tuple',
TIP=_S._lang.NEW_ELE_INFO, CURR_VAL=html_or_info))
try:
ele = self._run_js(js, *args)
except JavaScriptError:
raise RuntimeError(_S._lang.join(_S._lang.DICT_TO_NEW_ELE))
return ele
def get_frame(self, loc_ind_ele, timeout=None):
return get_frame(self, loc_ind_ele=loc_ind_ele, timeout=timeout)
def get_frames(self, locator=None, timeout=None):
locator = locator or 'xpath://*[name()="iframe" or name()="frame"]'
frames = self._ele(locator, timeout=timeout, index=None, raise_err=False)
return ChromiumElementsList(self, frames)
def session_storage(self, item=None):
js = f'sessionStorage.getItem("{item}")' if item else 'sessionStorage'
return self._run_js_loaded(js, as_expr=True)
def local_storage(self, item=None):
js = f'localStorage.getItem("{item}")' if item else 'localStorage'
return self._run_js_loaded(js, as_expr=True)
def get_screenshot(self, path=None, name=None, as_bytes=None, as_base64=None,
full_page=False, left_top=None, right_bottom=None):
return self._get_screenshot(path=path, name=name, as_bytes=as_bytes, as_base64=as_base64,
full_page=full_page, left_top=left_top, right_bottom=right_bottom)
def add_init_js(self, script):
js_id = self._run_cdp('Page.addScriptToEvaluateOnNewDocument', source=script,
includeCommandLineAPI=True)['identifier']
self._init_jss.append(js_id)
return js_id
def remove_init_js(self, script_id=None):
if script_id is None:
for js_id in self._init_jss:
self._run_cdp('Page.removeScriptToEvaluateOnNewDocument', identifier=js_id)
self._init_jss.clear()
elif script_id in self._init_jss:
self._run_cdp('Page.removeScriptToEvaluateOnNewDocument', identifier=script_id)
self._init_jss.remove(script_id)
def clear_cache(self, session_storage=True, local_storage=True, cache=True, cookies=True):
if session_storage and local_storage and cache and cookies:
self._run_cdp_loaded("Storage.clearDataForOrigin", origin="*", storageTypes="all")
if session_storage or local_storage:
self._run_cdp_loaded('DOMStorage.enable')
i = self._run_cdp('Storage.getStorageKeyForFrame', frameId=self._frame_id)['storageKey']
if session_storage:
self._run_cdp('DOMStorage.clear', storageId={'storageKey': i, 'isLocalStorage': False})
if local_storage:
self._run_cdp('DOMStorage.clear', storageId={'storageKey': i, 'isLocalStorage': True})
self._run_cdp_loaded('DOMStorage.disable')
if cache:
self._run_cdp_loaded('Network.clearBrowserCache')
if cookies:
self._run_cdp_loaded('Network.clearBrowserCookies')
def disconnect(self):
if self._driver:
self._disconnect_flag = True
self._driver.stop()
self.browser._all_drivers.get(self._driver.id, set()).discard(self._driver)
self._disconnect_flag = False
def reconnect(self, wait=0):
t_id = self._target_id
self.disconnect()
sleep(wait)
self.browser.reconnect()
self._driver = self.browser._get_driver(t_id, self)
self._driver_init(t_id)
self._get_document()
def handle_alert(self, accept=True, send=None, timeout=None, next_one=False):
if not isinstance(accept, bool):
return self._handle_alert(accept=accept, send=send, timeout=timeout, next_one=next_one)
r = self._handle_alert(accept=accept, send=send, timeout=timeout, next_one=next_one)
while self._has_alert:
sleep(.0001)
return r
def _handle_alert(self, accept=True, send=None, timeout=None, next_one=False):
if next_one:
self._alert.handle_next = accept
self._alert.next_text = send
return
if timeout is None:
timeout = self.timeout
timeout = .1 if timeout <= 0 else timeout
end_time = perf_counter() + timeout
while not self._alert.activated and perf_counter() < end_time:
sleep(.01)
if not self._alert.activated:
return False
res_text = self._alert.text
if not isinstance(accept, bool):
return res_text
d = {'accept': accept, '_timeout': 0}
if self._alert.type == 'prompt' and send is not None:
d['promptText'] = send
self.driver.run('Page.handleJavaScriptDialog', **d)
return res_text
def _on_alert_open(self, **kwargs):
self._alert.activated = True
self._alert.text = kwargs['message']
self._alert.type = kwargs['type']
self._alert.defaultPrompt = kwargs.get('defaultPrompt', None)
self._alert.response_accept = None
self._alert.response_text = None
self._has_alert = True
if self._alert.auto is not None:
if self._alert.auto != 'close':
self._handle_alert(self._alert.auto)
elif _S.auto_handle_alert is not None:
self._handle_alert(_S.auto_handle_alert)
elif self._alert.handle_next is not None:
self._handle_alert(self._alert.handle_next, self._alert.next_text)
self._alert.handle_next = None
def _on_alert_close(self, **kwargs):
self._alert.activated = False
self._alert.text = None
self._alert.type = None
self._alert.defaultPrompt = None
self._alert.response_accept = kwargs.get('result')
self._alert.response_text = kwargs['userInput']
self._has_alert = False
def _wait_loaded(self, timeout=None):
if timeout is None:
timeout = self.timeouts.page_load
end_time = perf_counter() + timeout
while perf_counter() < end_time:
if self._ready_state == 'complete':
return True
elif self._load_mode == 'eager' and self._ready_state in ('interactive',
'complete') and not self._is_loading:
return True
sleep(.01)
try:
self.stop_loading()
except CDPError:
pass
return False
def _d_connect(self, to_url, times=0, interval=1, show_errmsg=False, timeout=None):
err = None
self._is_loading = True
timeout = timeout if timeout is not None else self.timeouts.page_load
for t in range(times + 1):
err = None
end_time = perf_counter() + timeout
try:
result = self._run_cdp('Page.navigate', frameId=self._frame_id, url=to_url, _timeout=timeout)
if 'errorText' in result:
err = ConnectionError(_S._lang.join(_S._lang.CONNECT_ERR, INFO=result['errorText']))
except TimeoutError:
err = TimeoutError(_S._lang.join(_S._lang.TIMEOUT_, _S._lang.PAGE_CONNECT, timeout))
if err:
if t < times:
sleep(interval)
if show_errmsg:
print(f'{_S._lang.RETRY}{t + 1} {to_url}')
end_time1 = end_time - perf_counter()
while self._ready_state not in ('loading', 'complete') and perf_counter() < end_time1: # 等待出错信息显示
sleep(.01)
self.stop_loading()
continue
if self._load_mode == 'none':
return True
yu = end_time - perf_counter()
ok = self._wait_loaded(1 if yu <= 0 else yu)
if not ok:
err = TimeoutError(_S._lang.join(_S._lang.TIMEOUT_, _S._lang.PAGE_CONNECT, timeout))
if t < times:
sleep(interval)
if show_errmsg:
print(f'{_S._lang.RETRY}{t + 1} {to_url}')
continue
if not err:
break
if err:
if show_errmsg:
raise err if err is not None else ConnectionError(_S._lang.join(_S._lang.CONNECT_ERR))
return False
return True
def _get_screenshot(self, path=None, name=None, as_bytes=None, as_base64=None,
full_page=False, left_top=None, right_bottom=None, ele=None):
if as_bytes:
if as_bytes is True:
pic_type = 'png'
else:
if as_bytes not in ('jpg', 'jpeg', 'png', 'webp'):
raise ValueError(_S._lang.join(_S._lang.INCORRECT_VAL_, 'as_bytes',
ALLOW_VAL='"jpg", "jpeg", "png", "webp"', CURR_VAL=as_bytes))
pic_type = 'jpeg' if as_bytes == 'jpg' else as_bytes
elif as_base64:
if as_base64 is True:
pic_type = 'png'
else:
if as_base64 not in ('jpg', 'jpeg', 'png', 'webp'):
raise ValueError(_S._lang.join(_S._lang.INCORRECT_VAL_, 'as_base64',
ALLOW_VAL='"jpg", "jpeg", "png", "webp"', CURR_VAL=as_base64))
pic_type = 'jpeg' if as_base64 == 'jpg' else as_base64
else:
path = str(path).rstrip('\\/') if path else '.'
if not path.endswith(('.jpg', '.jpeg', '.png', '.webp')):
if not name:
name = f'{self.title}.jpg'
elif not name.endswith(('.jpg', '.jpeg', '.png', '.webp')):
name = f'{name}.jpg'
path = f'{path}{sep}{make_valid_name(name)}'
path = Path(path)
pic_type = path.suffix.lower()
pic_type = 'jpeg' if pic_type == '.jpg' else pic_type[1:]
if full_page:
width, height = self.rect.size
if width == 0 or height == 0:
raise RuntimeError(_S._lang.join(_S._lang.ZERO_PAGE_SIZE))
vp = {'x': 0, 'y': 0, 'width': width, 'height': height, 'scale': 1}
args = {'format': pic_type, 'captureBeyondViewport': True, 'clip': vp}
else:
if left_top or right_bottom:
if not left_top:
left_top = (0, 0)
if not right_bottom:
right_bottom = self.rect.size
x, y = left_top
w = right_bottom[0] - x
h = right_bottom[1] - y
v = not (location_in_viewport(self, x, y) and
location_in_viewport(self, right_bottom[0], right_bottom[1]))
if v and (self._run_js('return document.body.scrollHeight > window.innerHeight;') and
not self._run_js('return document.body.scrollWidth > window.innerWidth;')):
x += 10
vp = {'x': x, 'y': y, 'width': w, 'height': h, 'scale': 1}
args = {'format': pic_type, 'captureBeyondViewport': v, 'clip': vp}
else:
args = {'format': pic_type}
if pic_type == 'jpeg':
args['quality'] = 100
png = self._run_cdp_loaded('Page.captureScreenshot', **args)['data']
if as_base64:
return png
from base64 import b64decode
png = b64decode(png)
if as_bytes:
return png
path.parent.mkdir(parents=True, exist_ok=True)
with open(path, 'wb') as f:
f.write(png)
return str(path.absolute())
class Timeout(object):
def __init__(self, base=None, page_load=None, script=None):
self.base = 10 if base is None else base
self.page_load = 30 if page_load is None else page_load
self.script = 30 if script is None else script
def __repr__(self):
return str({'base': self.base, 'page_load': self.page_load, 'script': self.script})
@property
def as_dict(self):
return {'base': self.base, 'page_load': self.page_load, 'script': self.script}
class Alert(object):
def __init__(self, auto=None):
self.activated = False
self.text = None
self.type = None
self.defaultPrompt = None
self.response_accept = None
self.response_text = None
self.handle_next = None
self.next_text = None
self.auto = auto
def close_privacy_dialog(page, tid):
"""关闭隐私声明弹窗
:param page: ChromiumBase对象
:param tid: tab id
:return: None
"""
try:
driver = page.browser._get_driver(tid)
driver.run('Runtime.enable')
driver.run('DOM.enable')
driver.run('DOM.getDocument')
sid = driver.run('DOM.performSearch', query='//*[name()="privacy-sandbox-notice-dialog-app"]',
includeUserAgentShadowDOM=True)['searchId']
r = driver.run('DOM.getSearchResults', searchId=sid, fromIndex=0, toIndex=1)['nodeIds'][0]
end_time = perf_counter() + 3
while perf_counter() < end_time:
try:
r = driver.run('DOM.describeNode', nodeId=r)['node']['shadowRoots'][0]['backendNodeId']
break
except KeyError:
pass
sleep(.05)
driver.run('DOM.discardSearchResults', searchId=sid)
r = driver.run('DOM.resolveNode', backendNodeId=r)['object']['objectId']
r = driver.run('Runtime.callFunctionOn', objectId=r,
functionDeclaration='function(){return this.getElementById("ackButton");}')['result']['objectId']
driver.run('Runtime.callFunctionOn', objectId=r, functionDeclaration='function(){return this.click();}')
driver.close()
except:
pass
@@ -0,0 +1,681 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from pathlib import Path
from typing import Union, Tuple, Any, Optional, Literal
from requests import Session
from .chromium_page import ChromiumPage
from .chromium_tab import ChromiumTab
from .mix_tab import MixTab
from .web_page import WebPage
from .._base.base import BasePage
from .._base.chromium import Chromium
from .._base.driver import Driver
from .._elements.chromium_element import ChromiumElement
from .._elements.session_element import SessionElement
from .._functions.cookies import CookiesList
from .._functions.elements import SessionElementsList, ChromiumElementsList
from .._pages.chromium_frame import ChromiumFrame
from .._units.actions import Actions
from .._units.console import Console
from .._units.listener import Listener
from .._units.rect import TabRect, FrameRect
from .._units.screencast import Screencast
from .._units.scroller import Scroller, PageScroller
from .._units.setter import ChromiumBaseSetter
from .._units.states import PageStates
from .._units.waiter import BaseWaiter
PIC_TYPE = Literal['jpg', 'jpeg', 'png', 'webp', True]
class ChromiumBase(BasePage):
"""标签页、Frame、Page基类"""
_tab: Union[ChromiumTab, MixTab, ChromiumFrame, ChromiumPage, WebPage] = ...
_browser: Chromium = ...
_driver: Optional[Driver] = ...
_frame_id: str = ...
_is_reading: bool = ...
_is_timeout: bool = ...
_timeouts: Timeout = ...
_first_run: bool = ...
_is_loading: Optional[bool] = ...
_load_mode: str = ...
_scroll: Optional[Scroller] = ...
_url: str = ...
_root_id: Optional[str] = ...
_upload_list: Optional[list] = ...
_wait: Optional[BaseWaiter] = ...
_set: Optional[ChromiumBaseSetter] = ...
_screencast: Optional[Screencast] = ...
_actions: Optional[Actions] = ...
_listener: Optional[Listener] = ...
_states: Optional[PageStates] = ...
_alert: Alert = ...
_has_alert: bool = ...
_auto_handle_alert: Optional[bool] = ...
_doc_got: bool = ...
_load_end_time: float = ...
_init_jss: list = ...
_ready_state: Optional[str] = ...
_rect: Optional[TabRect] = ...
_console: Optional[Console] = ...
_disconnect_flag: bool = ...
_type: str = ...
def __init__(self,
browser: Chromium,
tab_id: str = None):
"""
:param browser: Chromium
:param tab_id: 要控制的tab id,不指定默认为激活的标签页
"""
...
def __call__(self,
locator: Union[Tuple[str, str], str, ChromiumElement],
index: int = 1,
timeout: float = None) -> ChromiumElement:
"""在内部查找元素
例:ele = page('@id=ele_id')
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个元素,从1开始,可传入负数获取倒数第几个
:param timeout: 超时时间(秒)
:return: ChromiumElement对象
"""
...
def _d_set_runtime_settings(self) -> None: ...
def _connect_browser(self, target_id: str = None) -> None:
"""连接浏览器,在第一次时运行
:param target_id: 要控制的target id,不指定默认为激活的标签页
:return: None
"""
...
def _driver_init(self, target_id: str) -> None:
"""新建页面、页面刷新后要进行的cdp参数初始化
:param target_id: 要跳转到的target id
:return: None
"""
...
def _get_document(self, timeout: float = 10) -> bool:
"""获取页面文档
:param timeout: 超时时间(秒)
:return: 是否获取成功
"""
...
def _onFrameDetached(self, **kwargs) -> None: ...
def _onFrameAttached(self, **kwargs) -> None: ...
def _onFrameStartedLoading(self, **kwargs):
"""页面开始加载时执行"""
...
def _onFrameNavigated(self, **kwargs):
"""页面跳转时执行"""
...
def _onDomContentEventFired(self, **kwargs):
"""在页面刷新、变化后重新读取页面内容"""
...
def _onLoadEventFired(self, **kwargs):
"""在页面刷新、变化后重新读取页面内容"""
...
def _onFrameStoppedLoading(self, **kwargs):
"""页面加载完成后执行"""
...
def _onFileChooserOpened(self, **kwargs):
"""文件选择框打开时执行"""
...
def _wait_to_stop(self):
"""eager策略超时时使页面停止加载"""
...
@property
def wait(self) -> BaseWaiter:
"""返回用于等待的对象"""
...
@property
def set(self) -> ChromiumBaseSetter:
"""返回用于设置的对象"""
...
@property
def screencast(self) -> Screencast:
"""返回用于录屏的对象"""
...
@property
def actions(self) -> Actions:
"""返回用于执行动作链的对象"""
...
@property
def listen(self) -> Listener:
"""返回用于聆听数据包的对象"""
...
@property
def states(self) -> PageStates:
"""返回用于获取状态信息的对象"""
...
@property
def scroll(self) -> PageScroller:
"""返回用于滚动滚动条的对象"""
...
@property
def rect(self) -> Union[TabRect, FrameRect]:
"""返回获取窗口坐标和大小的对象"""
...
@property
def console(self) -> Console:
"""返回获取控制台信息的对象"""
...
@property
def timeout(self) -> float:
"""返回timeout设置"""
...
@property
def timeouts(self) -> Timeout:
"""返回timeouts设置"""
...
@property
def browser(self) -> Chromium:
"""返回浏览器对象"""
...
@property
def driver(self) -> Driver:
"""返回用于控制浏览器的Driver对象"""
...
@property
def title(self) -> str:
"""返回当前页面title"""
...
@property
def url(self) -> str:
"""返回当前页面url"""
...
@property
def _browser_url(self) -> str:
"""用于被MixTab覆盖"""
...
@property
def html(self) -> str:
"""返回当前页面html文本"""
...
@property
def json(self) -> Union[dict, None]:
"""当返回内容是json格式时,返回对应的字典,非json格式时返回None"""
...
@property
def tab_id(self) -> str:
"""返回当前标签页id"""
...
@property
def _target_id(self) -> str:
"""返回当前标签页id"""
...
@property
def active_ele(self) -> ChromiumElement:
"""返回当前焦点所在元素"""
...
@property
def load_mode(self) -> Literal['none', 'normal', 'eager']:
"""返回页面加载策略,有3种:'none'、'normal'、'eager'"""
...
@property
def user_agent(self) -> str:
"""返回user agent"""
...
@property
def upload_list(self) -> list:
"""返回等待上传文件列表"""
...
@property
def session(self) -> Session:
"""返回用于转换模式或download的Session对象"""
...
@property
def _js_ready_state(self) -> str:
"""返回js获取的ready state信息"""
...
def run_cdp(self, cmd: str, **cmd_args) -> dict:
"""执行Chrome DevTools Protocol语句
:param cmd: 协议项目
:param cmd_args: 参数
:return: 执行的结果
"""
...
def run_cdp_loaded(self, cmd: str, **cmd_args) -> dict:
"""执行Chrome DevTools Protocol语句,执行前等待页面加载完毕
:param cmd: 协议项目
:param cmd_args: 参数
:return: 执行的结果
"""
...
def _run_cdp(self, cmd: str, **cmd_args) -> dict:
"""执行Chrome DevTools Protocol语句
:param cmd: 协议项目
:param cmd_args: 参数
:return: 执行的结果
"""
...
def _run_cdp_loaded(self, cmd: str, **cmd_args) -> dict:
"""执行Chrome DevTools Protocol语句,执行前等待页面加载完毕
:param cmd: 协议项目
:param cmd_args: 参数
:return: 执行的结果
"""
...
def run_js(self, script: Union[str, Path], *args, as_expr: bool = False, timeout: float = None) -> Any:
"""运行javascript代码
:param script: js文本或js文件路径
:param args: 参数,按顺序在js文本中对应arguments[0]、arguments[1]...
:param as_expr: 是否作为表达式运行,为True时args无效
:param timeout: js超时时间(秒),为None则使用页面timeouts.script设置
:return: 运行的结果
"""
...
def run_js_loaded(self, script: Union[str, Path], *args, as_expr: bool = False, timeout: float = None) -> Any:
"""运行javascript代码,执行前等待页面加载完毕
:param script: js文本或js文件路径
:param args: 参数,按顺序在js文本中对应arguments[0]、arguments[1]...
:param as_expr: 是否作为表达式运行,为True时args无效
:param timeout: js超时时间(秒),为None则使用页面timeouts.script属性值
:return: 运行的结果
"""
...
def _run_js(self, script: Union[str, Path], *args, as_expr: bool = False, timeout: float = None) -> Any:
"""运行javascript代码
:param script: js文本或js文件路径
:param args: 参数,按顺序在js文本中对应arguments[0]、arguments[1]...
:param as_expr: 是否作为表达式运行,为True时args无效
:param timeout: js超时时间(秒),为None则使用页面timeouts.script设置
:return: 运行的结果
"""
...
def _run_js_loaded(self, script: Union[str, Path], *args, as_expr: bool = False, timeout: float = None) -> Any:
"""运行javascript代码,执行前等待页面加载完毕
:param script: js文本或js文件路径
:param args: 参数,按顺序在js文本中对应arguments[0]、arguments[1]...
:param as_expr: 是否作为表达式运行,为True时args无效
:param timeout: js超时时间(秒),为None则使用页面timeouts.script属性值
:return: 运行的结果
"""
...
def run_async_js(self, script: Union[str, Path], *args, as_expr: bool = False) -> None:
"""以异步方式执行js代码或js文件路径
:param script: js文本
:param args: 参数,按顺序在js文本中对应arguments[0]、arguments[1]...
:param as_expr: 是否作为表达式运行,为True时args无效
:return: None
"""
...
def get(self, url: str, show_errmsg: bool = False, retry: int = None,
interval: float = None, timeout: float = None) -> Union[None, bool]:
"""访问url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间(秒),为None时使用页面对象timeouts.page_load属性值
:return: 目标url是否可用
"""
...
def cookies(self, all_domains: bool = False, all_info: bool = False) -> CookiesList:
"""返回cookies信息
:param all_domains: 是否返回所有域的cookies
:param all_info: 是否返回所有信息,为False时只返回name、value、domain
:return: cookies信息
"""
...
def ele(self,
locator: Union[Tuple[str, str], str, ChromiumElement, ChromiumFrame],
index: int = 1,
timeout: float = None) -> ChromiumElement:
"""获取一个符合条件的元素对象
:param locator: 定位符或元素对象
:param index: 获取第几个元素,从1开始,可传入负数获取倒数第几个
:param timeout: 查找超时时间(秒),默认与页面等待时间一致
:return: ChromiumElement对象
"""
...
def eles(self,
locator: Union[Tuple[str, str], str],
timeout: float = None) -> ChromiumElementsList:
"""获取所有符合条件的元素对象
:param locator: 定位符或元素对象
:param timeout: 查找超时时间(秒),默认与页面等待时间一致
:return: ChromiumElement对象组成的列表
"""
...
def s_ele(self,
locator: Union[Tuple[str, str], str] = None,
index: int = 1,
timeout: float = None) -> SessionElement:
"""查找一个符合条件的元素以SessionElement形式返回,处理复杂页面时效率很高
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: SessionElement对象或属性、文本
"""
...
def s_eles(self,
locator: Union[Tuple[str, str], str],
timeout: float = None) -> SessionElementsList:
"""查找所有符合条件的元素以SessionElement列表形式返回
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: SessionElement对象组成的列表
"""
...
def _find_elements(self,
locator: Union[Tuple[str, str], str, ChromiumElement, ChromiumFrame],
timeout: float,
index: Optional[int] = 1,
relative: bool = False,
raise_err: bool = None) -> Union[ChromiumElement, ChromiumFrame, ChromiumElementsList]:
"""执行元素查找
:param locator: 定位符或元素对象
:param timeout: 查找超时时间(秒)
:param index: 第几个结果,从1开始,可传入负数获取倒数第几个,为None返回所有
:param relative: MixTab用的表示是否相对定位的参数
:param raise_err: 找不到元素是是否抛出异常,为None时根据全局设置
:return: ChromiumElement对象或元素对象组成的列表
"""
...
def refresh(self, ignore_cache: bool = False) -> None:
"""刷新当前页面
:param ignore_cache: 是否忽略缓存
:return: None
"""
...
def forward(self, steps: int = 1) -> None:
"""在浏览历史中前进若干步
:param steps: 前进步数
:return: None
"""
...
def back(self, steps: int = 1) -> None:
"""在浏览历史中后退若干步
:param steps: 后退步数
:return: None
"""
...
def _forward_or_back(self, steps: int) -> None:
"""执行浏览器前进或后退,会跳过url相同的历史记录
:param steps: 步数
:return: None
"""
...
def stop_loading(self) -> None:
"""页面停止加载"""
...
def remove_ele(self, loc_or_ele: Union[ChromiumElement, ChromiumFrame, str, Tuple[str, str]]) -> None:
"""从页面上删除一个元素
:param loc_or_ele: 元素对象或定位符
:return: None
"""
...
def add_ele(self,
html_or_info: Union[str, Tuple[str, dict]],
insert_to: Union[ChromiumElement, str, Tuple[str, str], None] = None,
before: Union[ChromiumElement, str, Tuple[str, str], None] = None) -> Union[
ChromiumElement, ChromiumFrame]:
"""新建一个元素
:param html_or_info: 新元素的html文本或信息。信息格式为:(tag, {attr1: value, ...})
:param insert_to: 插入到哪个元素中,可接收元素对象和定位符,为None且为html添加到body,不为html不插入
:param before: 在哪个子节点前面插入,可接收对象和定位符,为None插入到父元素末尾
:return: 元素对象
"""
...
def get_frame(self,
loc_ind_ele: Union[str, int, tuple, ChromiumFrame, ChromiumElement],
timeout: float = None) -> ChromiumFrame:
"""获取页面中一个frame对象
:param loc_ind_ele: 定位符、iframe序号、ChromiumFrame对象,序号从1开始,可传入负数获取倒数第几个
:param timeout: 查找元素超时时间(秒)
:return: ChromiumFrame对象
"""
...
def get_frames(self,
locator: Union[str, tuple] = None,
timeout: float = None) -> ChromiumElementsList:
"""获取所有符合条件的frame对象
:param locator: 定位符,为None时返回所有
:param timeout: 查找超时时间(秒)
:return: ChromiumFrame对象组成的列表
"""
...
def session_storage(self, item: str = None) -> Union[str, dict, None]:
"""返回sessionStorage信息,不设置item则获取全部
:param item: 要获取的项,不设置则返回全部
:return: sessionStorage一个或所有项内容
"""
...
def local_storage(self, item: str = None) -> Union[str, dict, None]:
"""返回localStorage信息,不设置item则获取全部
:param item: 要获取的项目,不设置则返回全部
:return: localStorage一个或所有项内容
"""
...
def get_screenshot(self, path: [str, Path] = None, name: str = None, as_bytes: PIC_TYPE = None,
as_base64: PIC_TYPE = None, full_page: bool = False, left_top: Tuple[int, int] = None,
right_bottom: Tuple[int, int] = None) -> Union[str, bytes]:
"""对页面进行截图,可对整个网页、可见网页、指定范围截图。对可视范围外截图需要90以上版本浏览器支持
:param path: 保存路径
:param name: 完整文件名,后缀可选 'jpg','jpeg','png','webp'
:param as_bytes: 是否以字节形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数和as_base64参数无效
:param as_base64: 是否以base64字符串形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数无效
:param full_page: 是否整页截图,为True截取整个网页,为False截取可视窗口
:param left_top: 截取范围左上角坐标
:param right_bottom: 截取范围右下角角坐标
:return: 图片完整路径或字节文本
"""
...
def add_init_js(self, script: str) -> str:
"""添加初始化脚本,在页面加载任何脚本前执行
:param script: js文本
:return: 添加的脚本的id
"""
...
def remove_init_js(self, script_id: str = None) -> None:
"""删除初始化脚本,js_id传入None时删除所有
:param script_id: 脚本的id
:return: None
"""
...
def clear_cache(self, session_storage: bool = True, local_storage: bool = True, cache: bool = True,
cookies: bool = True) -> None:
"""清除缓存,可选要清除的项
:param session_storage: 是否清除sessionStorage
:param local_storage: 是否清除localStorage
:param cache: 是否清除cache
:param cookies: 是否清除cookies
:return: None
"""
...
def disconnect(self) -> None:
"""断开与页面的连接,不关闭页面"""
...
def reconnect(self, wait: float = 0) -> None:
"""断开与页面原来的页面,重新建立连接
:param wait: 断开后等待若干秒再连接
:return: None
"""
...
def handle_alert(self,
accept: Optional[bool] = True,
send: str = None,
timeout: float = None,
next_one: bool = False) -> Union[str, False]:
"""处理提示框,可以自动等待提示框出现
:param accept: True表示确认,False表示取消,为None不会按按钮但依然返回文本值
:param send: 处理prompt提示框时可输入文本
:param timeout: 等待提示框出现的超时时间(秒),为None则使用self.timeout属性的值
:param next_one: 是否处理下一个出现的提示框,为True时timeout参数无效
:return: 提示框内容文本,未等到提示框则返回False
"""
...
def _handle_alert(self,
accept: Optional[bool] = True,
send: str = None,
timeout: float = None,
next_one: bool = False) -> Union[str, False]:
"""处理提示框,可以自动等待提示框出现
:param accept: True表示确认,False表示取消,其它值不会按按钮但依然返回文本值
:param send: 处理prompt提示框时可输入文本
:param timeout: 等待提示框出现的超时时间(秒),为None则使用self.timeout属性的值
:param next_one: 是否处理下一个出现的提示框,为True时timeout参数无效
:return: 提示框内容文本,未等到提示框则返回False
"""
...
def _on_alert_open(self, **kwargs):
"""alert出现时触发的方法"""
...
def _on_alert_close(self, **kwargs):
"""alert关闭时触发的方法"""
...
def _wait_loaded(self, timeout: float = None) -> bool:
"""等待页面加载完成,超时触发停止加载
:param timeout: 超时时间(秒)
:return: 是否成功,超时返回False
"""
...
def _d_connect(self, to_url: str, times: int = 0, interval: float = 1, show_errmsg: bool = False,
timeout: float = None) -> Union[bool, None]:
"""尝试连接,重试若干次
:param to_url: 要访问的url
:param times: 重试次数
:param interval: 重试间隔(秒)
:param show_errmsg: 是否抛出异常
:param timeout: 连接超时时间(秒)
:return: 是否成功,返回None表示不确定
"""
...
def _get_screenshot(self, path: [str, Path] = None, name: str = None, as_bytes: PIC_TYPE = None,
as_base64: PIC_TYPE = None, full_page: bool = False, left_top: Tuple[float, float] = None,
right_bottom: Tuple[float, float] = None, ele: ChromiumElement = None) -> Union[str, bytes]:
"""对页面进行截图,可对整个网页、可见网页、指定范围截图。对可视范围外截图需要90以上版本浏览器支持
:param path: 保存路径
:param name: 完整文件名,后缀可选 'jpg','jpeg','png','webp'
:param as_bytes: 是否以字节形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数和as_base64参数无效
:param as_base64: 是否以base64字符串形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数无效
:param full_page: 是否整页截图,为True截取整个网页,为False截取可视窗口
:param left_top: 截取范围左上角坐标
:param right_bottom: 截取范围右下角角坐标
:param ele: 为异域iframe内元素截图设置
:return: 图片完整路径或字节文本
"""
...
class Timeout(object):
"""用于保存d模式timeout信息的类"""
base: float = ...
page_load: float = ...
script: float = ...
def __init__(self, base=None, page_load=None, script=None):
"""
:param base: 默认超时时间
:param page_load: 页面加载超时时间
:param script: js超时时间
"""
...
@property
def as_dict(self) -> dict:
"""以dict格式返回timeout设置"""
...
class Alert(object):
"""用于保存alert信息的类"""
activated: Optional[bool] = ...
text: Optional[str] = ...
type: Optional[str] = ...
defaultPrompt: Optional[str] = ...
response_accept: Optional[str] = ...
response_text: Optional[str] = ...
handle_next: Optional[bool] = ...
next_text: Optional[str] = ...
auto: Optional[bool] = ...
def __init__(self, auto: bool = None): ...
@@ -0,0 +1,457 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from copy import copy
from re import search, findall, DOTALL
from time import sleep, perf_counter
from .._elements.chromium_element import ChromiumElement
from .._functions.settings import Settings as _S
from .._pages.chromium_base import ChromiumBase
from .._units.listener import FrameListener
from .._units.rect import FrameRect
from .._units.scroller import FrameScroller
from .._units.setter import ChromiumFrameSetter
from .._units.states import FrameStates
from .._units.waiter import FrameWaiter
from ..errors import ContextLostError, ElementLostError, PageDisconnectedError, JavaScriptError
class ChromiumFrame(ChromiumBase):
_Frames = {}
def __new__(cls, owner, ele, info=None):
fid = info['node']['frameId'] if info else owner._run_cdp('DOM.describeNode',
backendNodeId=ele._backend_id)['node']['frameId']
if _S.singleton_tab_obj and fid in cls._Frames:
r = cls._Frames[fid]
while not hasattr(r, '_type') or r._type != 'ChromiumFrame':
sleep(.01)
return r
r = object.__new__(cls)
cls._Frames[fid] = r
return r
def __init__(self, owner, ele, info=None):
if _S.singleton_tab_obj and hasattr(self, '_created'):
return
self._created = True
self._tab = owner._tab
self._target_page = owner
self._backend_id = ele._backend_id
self._frame_ele = ele
self._reloading = False
try:
node = info['node'] if info else owner._run_cdp('DOM.describeNode', backendNodeId=ele._backend_id)['node']
self._frame_id = node['frameId']
if self._is_inner_frame():
self._is_diff_domain = False
self.doc_ele = ChromiumElement(self._target_page, backend_id=node['contentDocument']['backendNodeId'])
super().__init__(owner.browser, owner.driver.id)
else:
self._is_diff_domain = True
delattr(self, '_frame_id')
super().__init__(owner.browser, node['frameId'])
obj_id = super()._run_js('document;', as_expr=True)['objectId']
self.doc_ele = ChromiumElement(self, obj_id=obj_id)
except Exception as e:
ChromiumFrame._Frames.pop(self._frame_id, None)
raise e
self._type = 'ChromiumFrame'
def __call__(self, locator, index=1, timeout=None):
return self.ele(locator, index=index, timeout=timeout)
def __repr__(self):
attrs = [f"{k}='{v}'" for k, v in self._frame_ele.attrs.items()]
return f'<ChromiumFrame {self.frame_ele.tag} {" ".join(attrs)}>'
def __eq__(self, other):
return self._frame_id == getattr(other, '_frame_id', None)
def _d_set_runtime_settings(self):
if not hasattr(self, '_timeouts'):
self._timeouts = copy(self._target_page.timeouts)
self.retry_times = self._target_page.retry_times
self.retry_interval = self._target_page.retry_interval
self._download_path = self._target_page.download_path
self._auto_handle_alert = self._target_page._auto_handle_alert
self._load_mode = self._target_page._load_mode if not self._is_diff_domain else 'normal'
def _driver_init(self, target_id, is_init=True):
try:
super()._driver_init(target_id)
except:
if not self.browser._ws_only:
self.browser._driver.get(f'http://{self._browser.address}/json')
super()._driver_init(target_id)
self._driver.set_callback('Inspector.detached', self._onInspectorDetached, immediate=True)
self._driver.set_callback('Page.frameDetached', None)
self._driver.set_callback('Page.frameDetached', self._onFrameDetached, immediate=True)
def _reload(self):
self._is_loading = True
self._reloading = True
self._doc_got = False
self._driver.stop()
try:
self._frame_ele = ChromiumElement(self._target_page, backend_id=self._backend_id)
end_time = perf_counter() + 2
while perf_counter() < end_time:
node = self._target_page._run_cdp('DOM.describeNode', backendNodeId=self._frame_ele._backend_id)['node']
if 'frameId' in node:
break
sleep(.05)
else:
return
except (ElementLostError, PageDisconnectedError):
return
if self._is_inner_frame():
self._is_diff_domain = False
self.doc_ele = ChromiumElement(self._target_page, backend_id=node['contentDocument']['backendNodeId'])
self._frame_id = node['frameId']
if self._listener:
self._listener._to_target(self._target_page.tab_id, self._browser._ws_address, self)
super().__init__(self._browser, self._target_page.tab_id)
else:
self._is_diff_domain = True
if self._listener:
self._listener._to_target(node['frameId'], self._browser.address, self)
end_time = perf_counter() + self.timeouts.page_load
super().__init__(self._browser, node['frameId'])
timeout = end_time - perf_counter()
if timeout <= 0:
timeout = .5
self._wait_loaded(timeout)
self._is_loading = False
self._reloading = False
def _get_document(self, timeout=10):
if self._is_reading:
return
self._is_reading = True
try:
if self._is_diff_domain is False:
node = self._target_page._run_cdp('DOM.describeNode', backendNodeId=self._backend_id)['node']
self.doc_ele = ChromiumElement(self._target_page, backend_id=node['contentDocument']['backendNodeId'])
else:
timeout = max(timeout, 2)
b_id = self._run_cdp('DOM.getDocument', _timeout=timeout)['root']['backendNodeId']
self.doc_ele = ChromiumElement(self, backend_id=b_id)
self._root_id = self.doc_ele._obj_id
r = self._run_cdp('Page.getFrameTree', _ignore=PageDisconnectedError)
for i in findall(r"'id': '(.*?)'", str(r)):
self.browser._frames[i] = self.tab_id
return True
except:
return False
finally:
if not self._reloading: # 阻止reload时标识
self._is_loading = False
self._is_reading = False
def _onInspectorDetached(self, **kwargs):
# 异域转同域或退出
self._reload()
def _onFrameDetached(self, **kwargs):
# 同域变异域
self.browser._frames.pop(kwargs['frameId'], None)
ChromiumFrame._Frames.pop(kwargs['frameId'], None)
if kwargs['frameId'] == self._frame_id:
self._reload()
# ----------挂件----------
@property
def scroll(self):
self.wait.doc_loaded()
if self._scroll is None:
self._scroll = FrameScroller(self)
return self._scroll
@property
def set(self):
if self._set is None:
self._set = ChromiumFrameSetter(self)
return self._set
@property
def states(self):
if self._states is None:
self._states = FrameStates(self)
return self._states
@property
def wait(self):
if self._wait is None:
self._wait = FrameWaiter(self)
return self._wait
@property
def rect(self):
if self._rect is None:
self._rect = FrameRect(self)
return self._rect
@property
def listen(self):
if self._listener is None:
self._listener = FrameListener(self)
return self._listener
# ----------挂件结束----------
@property
def _obj_id(self):
return self.frame_ele._obj_id
@property
def _node_id(self):
return self.frame_ele._node_id
@property
def owner(self):
return self.frame_ele.owner
@property
def frame_ele(self):
return self._frame_ele
@property
def tag(self):
return self.frame_ele.tag
@property
def url(self):
try:
return self.doc_ele._run_js('return this.location.href;')
except JavaScriptError:
return None
@property
def html(self):
tag = self.tag
out_html = self._target_page._run_cdp('DOM.getOuterHTML', backendNodeId=self.frame_ele._backend_id)['outerHTML']
sign = search(rf'<{tag}.*?>', out_html, DOTALL).group(0)
return f'{sign}{self.inner_html}</{tag}>'
@property
def inner_html(self):
return self.doc_ele._run_js('return this.documentElement.outerHTML;')
@property
def link(self):
return self.frame_ele.link
@property
def title(self):
r = self._ele('t:title', raise_err=False)
return r.text if r else None
@property
def attrs(self):
return self.frame_ele.attrs
@property
def active_ele(self):
return self.doc_ele._run_js('return this.activeElement;')
@property
def xpath(self):
return self.frame_ele.xpath
@property
def css_path(self):
return self.frame_ele.css_path
@property
def tab(self):
return self._tab
@property
def tab_id(self):
return self.tab.tab_id
@property
def download_path(self):
return self._download_path
@property
def sr(self):
return self.frame_ele.sr
@property
def shadow_root(self):
return self.frame_ele.sr
@property
def child_count(self):
return int(self._ele('xpath:count(./*)'))
@property
def _js_ready_state(self):
if self._is_diff_domain:
return super()._js_ready_state
else:
try:
return self.doc_ele._run_js('return this.readyState;')
except ContextLostError:
try:
node = self._run_cdp('DOM.describeNode', backendNodeId=self.frame_ele._backend_id)['node']
doc = ChromiumElement(self._target_page, backend_id=node['contentDocument']['backendNodeId'])
return doc._run_js('return this.readyState;')
except:
return None
def refresh(self):
self.doc_ele._run_js('this.location.reload();')
def property(self, name):
return self.frame_ele.property(name)
def attr(self, name):
return self.frame_ele.attr(name)
def remove_attr(self, name):
self.frame_ele.remove_attr(name)
def style(self, style, pseudo_ele=''):
return self.frame_ele.style(style=style, pseudo_ele=pseudo_ele)
def run_js(self, script, *args, as_expr=False, timeout=None):
return self._run_js(script, *args, as_expr=as_expr, timeout=timeout)
def _run_js(self, script, *args, as_expr=False, timeout=None):
if script.startswith('this.scrollIntoView'):
return self.frame_ele._run_js(script, *args, as_expr=as_expr, timeout=timeout)
else:
return self.doc_ele._run_js(script, *args, as_expr=as_expr, timeout=timeout)
def parent(self, level_or_loc=1, index=1, timeout=0):
return self.frame_ele.parent(level_or_loc, index, timeout=timeout)
def prev(self, locator='', index=1, timeout=0, ele_only=True):
return self.frame_ele.prev(locator, index, timeout, ele_only=ele_only)
def next(self, locator='', index=1, timeout=0, ele_only=True):
return self.frame_ele.next(locator, index, timeout, ele_only=ele_only)
def before(self, locator='', index=1, timeout=None, ele_only=True):
return self.frame_ele.before(locator, index, timeout, ele_only=ele_only)
def after(self, locator='', index=1, timeout=None, ele_only=True):
return self.frame_ele.after(locator, index, timeout, ele_only=ele_only)
def prevs(self, locator='', timeout=0, ele_only=True):
return self.frame_ele.prevs(locator, timeout, ele_only=ele_only)
def nexts(self, locator='', timeout=0, ele_only=True):
return self.frame_ele.nexts(locator, timeout, ele_only=ele_only)
def befores(self, locator='', timeout=None, ele_only=True):
return self.frame_ele.befores(locator, timeout, ele_only=ele_only)
def afters(self, locator='', timeout=None, ele_only=True):
return self.frame_ele.afters(locator, timeout, ele_only=ele_only)
def get_screenshot(self, path=None, name=None, as_bytes=None, as_base64=None):
return self.frame_ele.get_screenshot(path=path, name=name, as_bytes=as_bytes, as_base64=as_base64)
def _get_screenshot(self, path=None, name=None, as_bytes: [bool, str] = None, as_base64: [bool, str] = None,
full_page=False, left_top=None, right_bottom=None, ele=None):
if not self._is_diff_domain:
return super().get_screenshot(path=path, name=name, as_bytes=as_bytes, as_base64=as_base64,
full_page=full_page, left_top=left_top, right_bottom=right_bottom)
if as_bytes:
if as_bytes is True:
pic_type = 'png'
else:
if as_bytes not in ('jpg', 'jpeg', 'png', 'webp'):
raise ValueError(_S._lang.join(_S._lang.INCORRECT_VAL_, 'as_bytes',
ALLOW_VAL='"jpg", "jpeg", "png", "webp"', CURR_VAL=as_bytes))
pic_type = 'jpeg' if as_bytes == 'jpg' else as_bytes
elif as_base64:
if as_base64 is True:
pic_type = 'png'
else:
if as_base64 not in ('jpg', 'jpeg', 'png', 'webp'):
raise ValueError(_S._lang.join(_S._lang.INCORRECT_VAL_, 'as_base64',
ALLOW_VAL='"jpg", "jpeg", "png", "webp"', CURR_VAL=as_base64))
pic_type = 'jpeg' if as_base64 == 'jpg' else as_base64
else:
path = str(path).rstrip('\\/') if path else '.'
if path and path.endswith(('.jpg', '.jpeg', '.png', '.webp')):
pic_type = path.rsplit('.', 1)[-1]
elif name and name.endswith(('.jpg', '.jpeg', '.png', '.webp')):
pic_type = name.rsplit('.', 1)[-1]
else:
pic_type = 'jpeg'
if pic_type == 'jpg':
pic_type = 'jpeg'
self.frame_ele.scroll.to_see(center=True)
self.scroll.to_see(ele, center=True)
cx, cy = ele.rect.viewport_location
w, h = ele.rect.size
img_data = f'data:image/{pic_type};base64,{self.frame_ele.get_screenshot(as_base64=True)}'
body = self.tab('t:body')
first_child = body('c::first-child')
if not isinstance(first_child, ChromiumElement):
first_child = first_child.frame_ele
js = f'''
img = document.createElement('img');
img.src = "{img_data}";
img.style.setProperty("z-index",9999999);
img.style.setProperty("position","fixed");
arguments[0].insertBefore(img, this);
return img;'''
new_ele = first_child._run_js(js, body)
new_ele.scroll.to_see(center=True)
top = int(self.frame_ele.style('border-top').split('px')[0])
left = int(self.frame_ele.style('border-left').split('px')[0])
r = self.tab._run_cdp('Page.getLayoutMetrics')['visualViewport']
sx = r['pageX']
sy = r['pageY']
r = self.tab.get_screenshot(path=path, name=name, as_bytes=as_bytes, as_base64=as_base64,
left_top=(cx + left + sx, cy + top + sy),
right_bottom=(cx + w + left + sx, cy + h + top + sy))
self.tab.remove_ele(new_ele)
return r
def _find_elements(self, locator, timeout, index=1, relative=False, raise_err=None):
if isinstance(locator, ChromiumElement):
return locator
self.wait.doc_loaded()
return self.doc_ele._ele(locator, index=index, timeout=timeout,
raise_err=raise_err) if index is not None else self.doc_ele.eles(locator, timeout)
def _is_inner_frame(self):
return self._frame_id in str(self._target_page._run_cdp('Page.getFrameTree')['frameTree'])
@@ -0,0 +1,458 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from pathlib import Path
from typing import Union, Tuple, List, Any, Optional, Literal
from .chromium_base import ChromiumBase
from .chromium_tab import ChromiumTab
from .mix_tab import MixTab
from .._elements.chromium_element import ChromiumElement, ShadowRoot
from .._functions.elements import ChromiumElementsList
from .._units.listener import FrameListener
from .._units.rect import FrameRect
from .._units.scroller import FrameScroller
from .._units.setter import ChromiumFrameSetter
from .._units.states import FrameStates
from .._units.waiter import FrameWaiter
class ChromiumFrame(ChromiumBase):
_Frames: dict = ...
_target_page: Union[ChromiumTab, ChromiumFrame] = ...
_tab: Union[MixTab, ChromiumTab] = ...
_set: ChromiumFrameSetter = ...
_frame_ele: ChromiumElement = ...
_backend_id: int = ...
_doc_ele: ChromiumElement = ...
_is_diff_domain: bool = ...
doc_ele: ChromiumElement = ...
_states: FrameStates = ...
_reloading: bool = ...
_rect: Optional[FrameRect] = ...
_listener: FrameListener = ...
def __init__(self,
owner: Union[ChromiumTab, ChromiumFrame],
ele: ChromiumElement,
info: dict = None):
"""
:param owner: frame所在的页面对象
:param ele: frame所在元素
:param info: frame所在元素信息
"""
...
def __call__(self,
locator: Union[Tuple[str, str], str],
index: int = 1,
timeout: float = None) -> ChromiumElement:
"""在内部查找元素
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 超时时间(秒)
:return: ChromiumElement对象或属性、文本
"""
...
def __repr__(self) -> str: ...
def __eq__(self, other: ChromiumFrame) -> bool: ...
def _d_set_runtime_settings(self) -> None:
"""重写设置浏览器运行参数方法"""
...
def _driver_init(self, target_id: str, is_init: bool = True) -> None:
"""避免出现服务器500错误
:param target_id: 要跳转到的target id
:return: None
"""
...
def _reload(self) -> None:
"""重新获取document"""
...
def _get_document(self, timeout: float = 10) -> bool:
"""刷新cdp使用的document数据
:param timeout: 超时时间(秒)
:return: 是否获取成功
"""
...
def _onFrameStoppedLoading(self, **kwargs): ...
def _onInspectorDetached(self, **kwargs): ...
@property
def scroll(self) -> FrameScroller:
"""返回用于滚动的对象"""
...
@property
def set(self) -> ChromiumFrameSetter:
"""返回用于设置的对象"""
...
@property
def states(self) -> FrameStates:
"""返回用于获取状态信息的对象"""
...
@property
def wait(self) -> FrameWaiter:
"""返回用于等待的对象"""
...
@property
def rect(self) -> FrameRect:
"""返回获取坐标和大小的对象"""
...
@property
def listen(self) -> FrameListener:
"""返回用于聆听数据包的对象"""
...
@property
def _obj_id(self) -> str: ...
@property
def _node_id(self) -> int: ...
@property
def owner(self) -> ChromiumBase:
"""返回所属页面对象"""
...
@property
def frame_ele(self) -> ChromiumElement:
"""返回总页面上的frame元素"""
...
@property
def tag(self) -> str:
"""返回元素tag"""
...
@property
def url(self) -> str:
"""返回frame当前访问的url"""
...
@property
def html(self) -> str:
"""返回元素outerHTML文本"""
...
@property
def inner_html(self) -> str:
"""返回元素innerHTML文本"""
...
@property
def link(self) -> str:
"""返回href或src绝对url"""
...
@property
def title(self) -> str:
"""返回页面title"""
...
@property
def attrs(self) -> dict:
"""返回frame元素所有attribute属性"""
...
@property
def active_ele(self) -> ChromiumElement:
"""返回当前焦点所在元素"""
...
@property
def xpath(self) -> str:
"""返回frame的xpath绝对路径"""
...
@property
def css_path(self) -> str:
"""返回frame的css selector绝对路径"""
...
@property
def tab(self) -> Union[ChromiumTab, MixTab]:
"""返回frame所在的tab对象"""
...
@property
def tab_id(self) -> str:
"""返回frame所在tab的id"""
...
@property
def download_path(self) -> str:
"""返回下载文件保存路径"""
...
@property
def sr(self) -> Union[None, ShadowRoot]:
"""返回iframe的shadow-root元素对象"""
...
@property
def shadow_root(self) -> Union[None, ShadowRoot]:
"""返回iframe的shadow-root元素对象"""
...
@property
def child_count(self) -> int:
"""返回直接子元素的个数"""
...
@property
def _js_ready_state(self) -> Literal['loading', 'interactive', 'complete']:
"""返回当前页面加载状态"""
...
def refresh(self) -> None:
"""刷新frame页面"""
...
def property(self, name: str) -> Union[str, None]:
"""返回frame元素一个property属性值
:param name: 属性名
:return: 属性值文本,没有该属性返回None
"""
...
def attr(self, name: str) -> Union[str, None]:
"""返回frame元素一个attribute属性值
:param name: 属性名
:return: 属性值文本,没有该属性返回None
"""
...
def remove_attr(self, name: str) -> None:
"""删除frame元素attribute属性
:param name: 属性名
:return: None
"""
...
def style(self, style: str, pseudo_ele: str = '') -> str:
"""返回frame元素样式属性值,可获取伪元素属性值
:param style: 样式属性名称
:param pseudo_ele: 伪元素名称(如有)
:return: 样式属性的值
"""
...
def run_js(self,
script: str,
*args,
as_expr: bool = False,
timeout: float = None) -> Any:
"""运行javascript代码
:param script: js文本
:param args: 参数,按顺序在js文本中对应arguments[0]、arguments[1]...
:param as_expr: 是否作为表达式运行,为True时args无效
:param timeout: js超时时间(秒),为None则使用页面timeouts.script设置
:return: 运行的结果
"""
...
def _run_js(self,
script: str,
*args,
as_expr: bool = False,
timeout: float = None) -> Any:
"""运行javascript代码
:param script: js文本
:param args: 参数,按顺序在js文本中对应arguments[0]、arguments[1]...
:param as_expr: 是否作为表达式运行,为True时args无效
:param timeout: js超时时间(秒),为None则使用页面timeouts.script设置
:return: 运行的结果
"""
...
def parent(self,
level_or_loc: Union[Tuple[str, str], str, int] = 1,
index: int = 1,
timeout: float = 0) -> ChromiumElement:
"""返回上面某一级父元素,可指定层数或用查询语法定位
:param level_or_loc: 第几级父元素,1开始,或定位符
:param index: 当level_or_loc传入定位符,使用此参数选择第几个结果,1开始
:param timeout: 查找超时时间(秒)
:return: 上级元素对象
"""
...
def prev(self,
locator: Union[Tuple[str, str], str, int] = '',
index: int = 1,
timeout: float = 0,
ele_only: bool = True) -> Union[ChromiumElement, str]:
"""返回当前元素前面一个符合条件的同级元素,可用查询语法筛选,可指定返回筛选结果的第几个
:param locator: 用于筛选的查询语法
:param index: 前面第几个查询结果,1开始
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 同级元素或节点
"""
...
def next(self,
locator: Union[Tuple[str, str], str, int] = '',
index: int = 1,
timeout: float = 0,
ele_only: bool = True) -> Union[ChromiumElement, str]:
"""返回当前元素后面一个符合条件的同级元素,可用查询语法筛选,可指定返回筛选结果的第几个
:param locator: 用于筛选的查询语法
:param index: 后面第几个查询结果,1开始
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 同级元素或节点
"""
...
def before(self,
locator: Union[Tuple[str, str], str, int] = '',
index: int = 1,
timeout: float = None,
ele_only: bool = True) -> Union[ChromiumElement, str]:
"""返回文档中当前元素前面符合条件的一个元素,可用查询语法筛选,可指定返回筛选结果的第几个
查找范围不限同级元素,而是整个DOM文档
:param locator: 用于筛选的查询语法
:param index: 前面第几个查询结果,1开始
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 本元素前面的某个元素或节点
"""
...
def after(self,
locator: Union[Tuple[str, str], str, int] = '',
index: int = 1,
timeout: float = None,
ele_only: bool = True) -> Union[ChromiumElement, str]:
"""返回文档中此当前元素后面符合条件的一个元素,可用查询语法筛选,可指定返回筛选结果的第几个
查找范围不限同级元素,而是整个DOM文档
:param locator: 用于筛选的查询语法
:param index: 后面第几个查询结果,1开始
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 本元素后面的某个元素或节点
"""
...
def prevs(self,
locator: Union[Tuple[str, str], str] = '',
timeout: float = 0,
ele_only: bool = True) -> List[Union[ChromiumElement, str]]:
"""返回当前元素前面符合条件的同级元素或节点组成的列表,可用查询语法筛选
:param locator: 用于筛选的查询语法
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 同级元素或节点文本组成的列表
"""
...
def nexts(self,
locator: Union[Tuple[str, str], str] = '',
timeout: float = 0,
ele_only: bool = True) -> List[Union[ChromiumElement, str]]:
"""返回当前元素后面符合条件的同级元素或节点组成的列表,可用查询语法筛选
:param locator: 用于筛选的查询语法
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 同级元素或节点文本组成的列表
"""
...
def befores(self,
locator: Union[Tuple[str, str], str] = '',
timeout: float = None,
ele_only: bool = True) -> List[Union[ChromiumElement, str]]:
"""返回文档中当前元素前面符合条件的元素或节点组成的列表,可用查询语法筛选
查找范围不限同级元素,而是整个DOM文档
:param locator: 用于筛选的查询语法
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 本元素前面的元素或节点组成的列表
"""
...
def afters(self,
locator: Union[Tuple[str, str], str] = '',
timeout: float = None,
ele_only: bool = True) -> List[Union[ChromiumElement, str]]:
"""返回文档中当前元素后面符合条件的元素或节点组成的列表,可用查询语法筛选
查找范围不限同级元素,而是整个DOM文档
:param locator: 用于筛选的查询语法
:param timeout: 查找节点的超时时间(秒)
:param ele_only: 是否只获取元素,为False时把文本、注释节点也纳入
:return: 本元素前面的元素或节点组成的列表
"""
...
def get_screenshot(self,
path: [str, Path] = None,
name: str = None,
as_bytes: [bool, str] = None,
as_base64: [bool, str] = None) -> Union[str, bytes]:
"""对页面进行截图,可对整个网页、可见网页、指定范围截图。对可视范围外截图需要90以上版本浏览器支持
:param path: 文件保存路径
:param name: 完整文件名,后缀可选 'jpg','jpeg','png','webp'
:param as_bytes: 是否以字节形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数和as_base64参数无效
:param as_base64: 是否以base64字符串形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数无效
:return: 图片完整路径或字节文本
"""
...
def _get_screenshot(self,
path: [str, Path] = None,
name: str = None,
as_bytes: [bool, str] = None,
as_base64: [bool, str] = None,
full_page: bool = False,
left_top: Tuple[int, int] = None,
right_bottom: Tuple[int, int] = None,
ele: ChromiumElement = None) -> Union[str, bytes]:
"""实现截图
:param path: 文件保存路径
:param name: 完整文件名,后缀可选 'jpg','jpeg','png','webp'
:param as_bytes: 是否以字节形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数和as_base64参数无效
:param as_base64: 是否以base64字符串形式返回图片,可选 'jpg','jpeg','png','webp',生效时path参数无效
:param full_page: 是否整页截图,为True截取整个网页,为False截取可视窗口
:param left_top: 截取范围左上角坐标
:param right_bottom: 截取范围右下角角坐标
:param ele: 为异域iframe内元素截图设置
:return: 图片完整路径或字节文本
"""
...
def _find_elements(self,
locator: Union[Tuple[str, str], str, ChromiumElement, ChromiumFrame],
timeout: float,
index: Optional[int] = 1,
relative: bool = False,
raise_err: bool = None) -> Union[ChromiumElement, ChromiumFrame, None, ChromiumElementsList]:
"""在frame内查找单个元素
:param locator: 定位符或元素对象
:param timeout: 查找超时时间(秒)
:param index: 第几个结果,从1开始,可传入负数获取倒数第几个,为None返回所有
:param relative: MixTab用的表示是否相对定位的参数
:param raise_err: 找不到元素是是否抛出异常,为None时根据全局设置
:return: ChromiumElement对象
"""
...
def _is_inner_frame(self) -> bool:
"""返回当前frame是否同域"""
...
@@ -0,0 +1,131 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from time import sleep
from .._base.chromium import Chromium
from .._functions.settings import Settings
from .._functions.web import save_page
from .._pages.chromium_base import ChromiumBase
from .._units.setter import ChromiumPageSetter
from .._units.waiter import ChromiumPageWaiter
class ChromiumPage(ChromiumBase):
_PAGES = {}
def __new__(cls, addr_or_opts=None, tab_id=None, timeout=None):
# 即将废弃timeout
browser = Chromium(addr_or_opts=addr_or_opts)
if browser.id in cls._PAGES:
r = cls._PAGES[browser.id]
while not hasattr(r, '_frame_id'):
sleep(.05)
return r
r = object.__new__(cls)
r._browser = browser
cls._PAGES[browser.id] = r
return r
def __init__(self, addr_or_opts=None, tab_id=None, timeout=None):
# 即将废弃timeout
if hasattr(self, '_created'):
return
self._created = True
self.tab = self
super().__init__(self.browser, tab_id)
self._type = 'ChromiumPage'
self.set.timeouts(base=timeout) # 即将废弃
self._tab = self
self._browser._dl_mgr._page_id = self.tab_id
def __repr__(self):
return f'<ChromiumPage browser_id={self.browser.id} tab_id={self.tab_id}>'
def _d_set_runtime_settings(self):
"""设置运行时用到的属性"""
self._timeouts = self.browser.timeouts
self._load_mode = self.browser._load_mode
self._download_path = self.browser.download_path
self.retry_times = self.browser.retry_times
self.retry_interval = self.browser.retry_interval
self._auto_handle_alert = self.browser._auto_handle_alert
@property
def set(self):
if self._set is None:
self._set = ChromiumPageSetter(self)
return self._set
@property
def wait(self):
if self._wait is None:
self._wait = ChromiumPageWaiter(self)
return self._wait
@property
def browser(self):
return self._browser
@property
def tabs_count(self):
return self.browser.tabs_count
@property
def tab_ids(self):
return self.browser.tab_ids
@property
def latest_tab(self):
return self.browser._get_tab(id_or_num=self.tab_ids[0], as_id=not Settings.singleton_tab_obj)
@property
def process_id(self):
return self.browser.process_id
@property
def browser_version(self):
return self._browser.version
@property
def address(self):
return self.browser.address
@property
def download_path(self):
return self.browser.download_path
def save(self, path=None, name=None, as_pdf=False, **kwargs):
return save_page(self, path, name, as_pdf, kwargs)
def get_tab(self, id_or_num=None, title=None, url=None, tab_type='page', as_id=False):
return self.browser._get_tab(id_or_num=id_or_num, title=title, url=url,
tab_type=tab_type, mix=False, as_id=as_id)
def get_tabs(self, title=None, url=None, tab_type='page', as_id=False):
return self.browser._get_tabs(title=title, url=url, tab_type=tab_type, mix=False, as_id=as_id)
def new_tab(self, url=None, new_window=False, background=False, new_context=False):
return self.browser._new_tab(False, url=url, new_window=new_window,
background=background, new_context=new_context)
def activate_tab(self, id_ind_tab):
self.browser.activate_tab(id_ind_tab)
def close(self):
self.browser._close_tab(self)
def close_tabs(self, tabs_or_ids, others=False):
self.browser.close_tabs(tabs_or_ids=tabs_or_ids, others=others)
def quit(self, timeout=5, force=True, del_data=False):
self.browser.quit(timeout, force, del_data=del_data)
def _on_disconnect(self):
ChromiumPage._PAGES.pop(self._browser.id, None)
@@ -0,0 +1,224 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from pathlib import Path
from typing import Union, Tuple, List, Optional
from .._base.chromium import Chromium
from .._configs.chromium_options import ChromiumOptions
from .._pages.chromium_base import ChromiumBase
from .._pages.chromium_tab import ChromiumTab
from .._units.rect import TabRect
from .._units.setter import ChromiumPageSetter
from .._units.waiter import ChromiumPageWaiter
class ChromiumPage(ChromiumBase):
"""用于管理浏览器和一个标签页的类"""
_PAGES: dict = ...
tab: ChromiumPage = ...
_browser: Chromium = ...
_rect: Optional[TabRect] = ...
_is_exist: bool = ...
def __new__(cls,
addr_or_opts: Union[str, int, ChromiumOptions] = None,
tab_id: str = None):
"""
:param addr_or_opts: 浏览器地址:端口、ws地址、ChromiumOptions对象或端口数字(int)
:param tab_id: 要控制的标签页id,不指定默认为激活的
"""
...
def __init__(self,
addr_or_opts: Union[str, int, ChromiumOptions] = None,
tab_id: str = None):
"""
:param addr_or_opts: 浏览器地址:端口、ws地址、ChromiumOptions对象或端口数字(int)
:param tab_id: 要控制的标签页id,不指定默认为激活的
"""
...
@property
def set(self) -> ChromiumPageSetter:
"""返回用于设置的对象"""
...
@property
def wait(self) -> ChromiumPageWaiter:
"""返回用于等待的对象"""
...
@property
def browser(self) -> Chromium:
"""返回浏览器对象"""
...
@property
def tabs_count(self) -> int:
"""返回标签页数量"""
...
@property
def tab_ids(self) -> List[str]:
"""返回所有标签页id组成的列表"""
...
@property
def latest_tab(self) -> Union[ChromiumTab, ChromiumPage, str]:
"""返回最新的标签页,最新标签页指最后创建或最后被激活的
当Settings.singleton_tab_obj==True时返回Tab对象,否则返回tab id"""
...
@property
def process_id(self) -> Optional[int]:
"""返回浏览器进程id"""
...
@property
def browser_version(self) -> str:
"""返回所控制的浏览器版本号"""
...
@property
def address(self) -> str:
"""返回浏览器地址ip:port"""
...
def save(self,
path: Union[str, Path] = None,
name: str = None,
as_pdf: bool = False,
landscape: bool = ...,
displayHeaderFooter: bool = ...,
printBackground: bool = ...,
scale: float = ...,
paperWidth: float = ...,
paperHeight: float = ...,
marginTop: float = ...,
marginBottom: float = ...,
marginLeft: float = ...,
marginRight: float = ...,
pageRanges: str = ...,
headerTemplate: str = ...,
footerTemplate: str = ...,
preferCSSPageSize: bool = ...,
generateTaggedPDF: bool = ...,
generateDocumentOutline: bool = ...) -> Union[bytes, str]:
"""把当前页面保存为文件,如果path和name参数都为None,只返回文本
:param path: 保存路径,为None且name不为None时保存在当前路径
:param name: 文件名,为None且path不为None时用title属性值
:param as_pdf: 为Ture保存为pdf,否则为mhtml且忽略kwargs参数
:param landscape: 纸张方向,as_pdf为True时才生效
:param displayHeaderFooter: 是否显示页头页脚,as_pdf为True时才生效
:param printBackground: 是否打印背景图片,as_pdf为True时才生效
:param scale: 缩放比例,as_pdf为True时才生效
:param paperWidth: 页面宽度(英寸),as_pdf为True时才生效
:param paperHeight: 页面高度(英寸),as_pdf为True时才生效
:param marginTop: 上边距(英寸),as_pdf为True时才生效
:param marginBottom: 下边距(英寸),as_pdf为True时才生效
:param marginLeft: 左边距(英寸),as_pdf为True时才生效
:param marginRight: 右边距(英寸),as_pdf为True时才生效
:param pageRanges: 页面范围,格式'1-5, 8, 11-13',as_pdf为True时才生效
:param headerTemplate: 页头HTML模板,as_pdf为True时才生效
模板可包含以下class:
- date:日期
- title:文档标题
- url:文档url
- pageNumber:当前页码
- totalPages:总页数
示例:<span class=title></span>
:param footerTemplate: 页脚HTML模板,格式与页头的一样,as_pdf为True时才生效
:param preferCSSPageSize: 是否使用css定义的页面大小,as_pdf为True时才生效
:param generateTaggedPDF: 是否生成带标签的(可访问的)PDF。默认为嵌入器选择,as_pdf为True时才生效
:param generateDocumentOutline: 是否将文档大纲嵌入到PDF中。,as_pdf为True时才生效
:return: as_pdf为True时返回bytes,否则返回文件文本
"""
...
def get_tab(self,
id_or_num: Union[str, ChromiumTab, int] = None,
title: str = None,
url: str = None,
tab_type: Union[str, list, tuple] = 'page',
as_id: bool = False) -> Union[ChromiumTab, str, None]:
"""获取一个标签页对象,id_or_num不为None时,后面几个参数无效
:param id_or_num: 要获取的标签页id或序号,序号从1开始,可传入负数获取倒数第几个,不是视觉排列顺序,而是激活顺序
:param title: 要匹配title的文本,模糊匹配,为None则匹配所有
:param url: 要匹配url的文本,模糊匹配,为None则匹配所有
:param tab_type: tab类型,可用列表输入多个,如 'page', 'iframe' 等,为None则匹配所有
:param as_id: 是否返回标签页id而不是标签页对象
:return: ChromiumTab对象
"""
...
def get_tabs(self,
title: str = None,
url: str = None,
tab_type: Union[str, list, tuple] = 'page',
as_id: bool = False) -> Union[List[ChromiumTab], List[str]]:
"""查找符合条件的tab,返回它们组成的列表
:param title: 要匹配title的文本,模糊匹配,为None则匹配所有
:param url: 要匹配url的文本,模糊匹配,为None则匹配所有
:param tab_type: tab类型,可用列表输入多个,如 'page', 'iframe' 等,为None则匹配所有
:param as_id: 是否返回标签页id而不是标签页对象
:return: ChromiumTab对象组成的列表
"""
...
def new_tab(self,
url: str = None,
new_window: bool = False,
background: bool = False,
new_context: bool = False) -> ChromiumTab:
"""新建一个标签页
:param url: 新标签页跳转到的网址
:param new_window: 是否在新窗口打开标签页
:param background: 是否不激活新标签页,如new_window为True则无效
:param new_context: 是否创建新的上下文
:return: 新标签页对象
"""
...
def activate_tab(self,
id_ind_tab: Union[int, str, ChromiumTab]) -> None:
"""使标签页变为活动状态
:param id_ind_tab: 标签页id(str)、Tab对象或标签页序号(int),序号从1开始
:return: None
"""
...
def close(self) -> None:
"""关闭Page管理的标签页"""
...
def close_tabs(self,
tabs_or_ids: Union[str, ChromiumTab, List[Union[str, ChromiumTab]],
Tuple[Union[str, ChromiumTab]]],
others: bool = False) -> None:
"""关闭传入的标签页,可传入多个
:param tabs_or_ids: 要关闭的标签页对象或id,可传入列表或元组
:param others: 是否关闭指定标签页之外的
:return: None
"""
...
def quit(self,
timeout: float = 5,
force: bool = True,
del_data: bool = False) -> None:
"""关闭浏览器
:param timeout: 等待浏览器关闭超时时间(秒)
:param force: 关闭超时是否强制终止进程
:param del_data: 是否删除用户文件夹
:return: None
"""
...
def _on_disconnect(self) -> None:
"""浏览器退出时执行"""
...
@@ -0,0 +1,76 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from copy import copy
from time import sleep
from .._functions.settings import Settings
from .._functions.web import save_page
from .._pages.chromium_base import ChromiumBase
from .._units.setter import TabSetter
from .._units.waiter import TabWaiter
class ChromiumTab(ChromiumBase):
_TABS = {}
def __new__(cls, browser, tab_id):
if Settings.singleton_tab_obj and tab_id in cls._TABS:
r = cls._TABS[tab_id]
while not hasattr(r, '_frame_id'):
sleep(.05)
return r
r = object.__new__(cls)
cls._TABS[tab_id] = r
return r
def __init__(self, browser, tab_id):
if Settings.singleton_tab_obj and hasattr(self, '_created'):
return
self._created = True
super().__init__(browser, tab_id)
self._tab = self
self._type = 'ChromiumTab'
def __repr__(self):
return f'<ChromiumTab browser_id={self.browser.id} tab_id={self.tab_id}>'
def _d_set_runtime_settings(self):
self._timeouts = copy(self.browser.timeouts)
self.retry_times = self.browser.retry_times
self.retry_interval = self.browser.retry_interval
self._load_mode = self.browser._load_mode
self._download_path = self.browser.download_path
self._auto_handle_alert = self.browser._auto_handle_alert
self._none_ele_return_value = self.browser._none_ele_return_value
self._none_ele_value = self.browser._none_ele_value
def close(self, others=False):
if others:
self.browser.close_tabs(self.tab_id, others=True)
else:
self.browser._close_tab(self)
@property
def set(self):
if self._set is None:
self._set = TabSetter(self)
return self._set
@property
def wait(self):
if self._wait is None:
self._wait = TabWaiter(self)
return self._wait
def save(self, path=None, name=None, as_pdf=False, **kwargs):
return save_page(self, path, name, as_pdf, kwargs)
def _on_disconnect(self):
if not self._disconnect_flag:
ChromiumTab._TABS.pop(self.tab_id, None)
@@ -0,0 +1,110 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from pathlib import Path
from typing import Union, Optional
from .chromium_base import ChromiumBase
from .._base.chromium import Chromium
from .._units.rect import TabRect
from .._units.setter import TabSetter
from .._units.waiter import TabWaiter
class ChromiumTab(ChromiumBase):
"""实现浏览器标签页的类"""
_TABS: dict = ...
_tab: ChromiumTab = ...
_rect: Optional[TabRect] = ...
def __new__(cls, browser: Chromium, tab_id: str):
"""
:param browser: Browser对象
:param tab_id: 标签页id
"""
...
def __init__(self, browser: Chromium, tab_id: str):
"""
:param browser: Browser对象
:param tab_id: 标签页id
"""
...
def _d_set_runtime_settings(self) -> None:
"""重写设置浏览器运行参数方法"""
...
def close(self, others: bool = False) -> None:
"""关闭标签页
:param others: 是否关闭其它,保留自己
:return: None
"""
...
@property
def set(self) -> TabSetter:
"""返回用于设置的对象"""
...
@property
def wait(self) -> TabWaiter:
"""返回用于等待的对象"""
...
def save(self,
path: Union[str, Path] = None,
name: str = None,
as_pdf: bool = False,
landscape: bool = False,
displayHeaderFooter: bool = False,
printBackground: bool = False,
scale: float = 1,
paperWidth: float = 8.5,
paperHeight: float = 11,
marginTop: float = 11,
marginBottom: float = 1,
marginLeft: float = 1,
marginRight: float = 1,
pageRanges: str = '',
headerTemplate: str = '',
footerTemplate: str = '',
preferCSSPageSize: bool = False,
generateTaggedPDF: bool = ...,
generateDocumentOutline: bool = ...) -> Union[bytes, str]:
"""把当前页面保存为文件,如果path和name参数都为None,只返回文本
:param path: 保存路径,为None且name不为None时保存在当前路径
:param name: 文件名,为None且path不为None时用title属性值
:param as_pdf: 为Ture保存为pdf,否则为mhtml且忽略kwargs参数
:param landscape: 纸张方向,as_pdf为True时才生效
:param displayHeaderFooter: 是否显示页头页脚,as_pdf为True时才生效
:param printBackground: 是否打印背景图片,as_pdf为True时才生效
:param scale: 缩放比例,as_pdf为True时才生效
:param paperWidth: 页面宽度(英寸),as_pdf为True时才生效
:param paperHeight: 页面高度(英寸),as_pdf为True时才生效
:param marginTop: 上边距(英寸),as_pdf为True时才生效
:param marginBottom: 下边距(英寸),as_pdf为True时才生效
:param marginLeft: 左边距(英寸),as_pdf为True时才生效
:param marginRight: 右边距(英寸),as_pdf为True时才生效
:param pageRanges: 页面范围,格式'1-5, 8, 11-13',as_pdf为True时才生效
:param headerTemplate: 页头HTML模板,as_pdf为True时才生效
模板可包含以下class:
- date:日期
- title:文档标题
- url:文档url
- pageNumber:当前页码
- totalPages:总页数
示例:<span class=title></span>
:param footerTemplate: 页脚HTML模板,格式与页头的一样,as_pdf为True时才生效
:param preferCSSPageSize: 是否使用css定义的页面大小,as_pdf为True时才生效
:param generateTaggedPDF: 是否生成带标签的(可访问的)PDF。默认为嵌入器选择,as_pdf为True时才生效
:param generateDocumentOutline: 是否将文档大纲嵌入到PDF中。,as_pdf为True时才生效
:return: as_pdf为True时返回bytes,否则返回文件文本
"""
...
def _on_disconnect(self): ...
@@ -0,0 +1,199 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from .chromium_tab import ChromiumTab
from .._base.base import BasePage
from .._configs.session_options import SessionOptions
from .._functions.cookies import set_session_cookies, set_tab_cookies
from .._functions.settings import Settings as _S
from .._pages.session_page import SessionPage
from .._units.setter import MixTabSetter
class MixTab(SessionPage, ChromiumTab, BasePage):
def __init__(self, browser, tab_id):
if _S.singleton_tab_obj and hasattr(self, '_created'):
return
self._d_mode = True
self._session_options = None
self._headers = None
self._response = None
self._session = None
self._encoding = None
self._timeout = 10
super(SessionPage, self).__init__(browser=browser, tab_id=tab_id)
self._type = 'MixTab'
def __call__(self, locator, index=1, timeout=None):
return super(SessionPage, self).__call__(locator, index=index, timeout=timeout) if self._d_mode \
else super().__call__(locator, index=index)
def __repr__(self):
return f'<MixTab browser_id={self.browser.id} tab_id={self.tab_id}>'
@property
def set(self):
if self._set is None:
self._set = MixTabSetter(self)
return self._set
@property
def url(self):
return self._browser_url if self._d_mode else self._session_url
@property
def _browser_url(self):
return super(SessionPage, self).url if self._driver else None
@property
def title(self):
return super(SessionPage, self).title if self._d_mode else super().title
@property
def raw_data(self):
return super(SessionPage, self).html if self._d_mode else super().raw_data
@property
def html(self):
return super(SessionPage, self).html if self._d_mode else super().html
@property
def json(self):
return super(SessionPage, self).json if self._d_mode else super().json
@property
def response(self):
return self._response
@property
def mode(self):
return 'd' if self._d_mode else 's'
@property
def user_agent(self):
return super(SessionPage, self).user_agent if self._d_mode else super().user_agent
@property
def _session_url(self):
return self._response.url if self._response else None
@property
def timeout(self):
return self.timeouts.base if self._d_mode else self._timeout
def get(self, url, show_errmsg=False, retry=None, interval=None, timeout=None, **kwargs):
if self._d_mode:
if kwargs:
raise ValueError(_S._lang.join(_S._lang.S_MODE_ONLY, ARGS=", ".join(kwargs.keys())))
return super(SessionPage, self).get(url, show_errmsg, retry, interval, timeout)
if timeout is None:
timeout = self.timeouts.page_load
return super().get(url, show_errmsg, retry, interval, timeout, **kwargs)
def post(self, url, show_errmsg=False, retry=None, interval=None, timeout=None, **kwargs):
if self.mode == 'd':
self.cookies_to_session()
if timeout is None:
kwargs['timeout'] = self.timeouts.page_load
if self._session is None:
self._create_session()
super().post(url, show_errmsg, retry, interval, **kwargs)
return self.response
def ele(self, locator, index=1, timeout=None):
return (super(SessionPage, self).ele(locator, index=index, timeout=timeout)
if self._d_mode else super().ele(locator, index=index))
def eles(self, locator, timeout=None):
return super(SessionPage, self).eles(locator, timeout=timeout) if self._d_mode else super().eles(locator)
def s_ele(self, locator=None, index=1, timeout=None):
return (super(SessionPage, self).s_ele(locator, index=index, timeout=timeout)
if self._d_mode else super().s_ele(locator, index=index, timeout=timeout))
def s_eles(self, locator, timeout=None):
return (super(SessionPage, self).s_eles(locator, timeout=timeout)
if self._d_mode else super().s_eles(locator, timeout=timeout))
def change_mode(self, mode=None, go=True, copy_cookies=True):
if mode:
mode = mode.lower()
if mode is not None and ((mode == 'd' and self._d_mode) or (mode == 's' and not self._d_mode)):
return
self._d_mode = not self._d_mode
# s模式转d模式
if self._d_mode:
if self._driver is None or not self._driver.is_running:
self._driver_init(self.tab_id)
self._get_document()
self._url = super(SessionPage, self).url
if self._session_url:
if copy_cookies:
self.cookies_to_browser()
if go:
self.get(self._session_url)
return
# d模式转s模式
if self._session is None:
self._set_session_options(
self.browser._session_options or SessionOptions(read_file=self.browser._session_options is None))
self._create_session()
self._url = self._session_url
if self._driver:
if copy_cookies:
self.cookies_to_session()
if go:
url = super(SessionPage, self).url
if url.startswith('http'):
self.get(url)
def cookies_to_session(self, copy_user_agent=True):
if not self._session:
return
if copy_user_agent:
user_agent = self._run_cdp('Runtime.evaluate', expression='navigator.userAgent;')['result']['value']
self._headers.update({"User-Agent": user_agent})
set_session_cookies(self.session, super(SessionPage, self).cookies())
def cookies_to_browser(self):
if self._driver is None or not self._driver.is_running:
return
set_tab_cookies(self, super().cookies())
def cookies(self, all_domains=False, all_info=False):
return super(SessionPage, self).cookies(all_domains, all_info) if self._d_mode \
else super().cookies(all_domains, all_info)
def close(self, others=False, session=False):
if others:
self.browser.close_tabs(self.tab_id, others=True)
else:
self.browser._close_tab(self)
if session and self._session:
self._session.close()
if self._response is not None:
self._response.close()
def _find_elements(self, locator, timeout, index=1, relative=False, raise_err=None):
return super(SessionPage, self)._find_elements(locator, timeout=timeout, index=index, relative=relative) \
if self._d_mode else super()._find_elements(locator, index=index, timeout=timeout)
def _set_session_options(self, session_or_options=None):
if session_or_options is None:
session_or_options = self.browser._session_options or SessionOptions(
read_file=self.browser._session_options is None)
super()._set_session_options(session_or_options)
@@ -0,0 +1,298 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from http.cookiejar import CookieJar
from typing import Union, Tuple, Any, Optional, Literal
from requests import Session, Response
from .chromium_frame import ChromiumFrame
from .chromium_tab import ChromiumTab
from .session_page import SessionPage
from .._base.chromium import Chromium
from .._elements.chromium_element import ChromiumElement
from .._elements.session_element import SessionElement
from .._functions.cookies import CookiesList
from .._functions.elements import SessionElementsList, ChromiumElementsList
from .._units.setter import MixTabSetter
from .._units.waiter import MixTabWaiter
class MixTab(SessionPage, ChromiumTab):
_tab: MixTab = ...
_d_mode: bool = ...
_set: MixTabSetter = ...
def __init__(self, browser: Chromium, tab_id: str):
"""
:param browser: Chromium对象
:param tab_id: 标签页id
"""
...
def __call__(self,
locator: Union[Tuple[str, str], str, ChromiumElement, SessionElement],
index: int = 1,
timeout: float = None) -> Union[ChromiumElement, SessionElement]:
"""在内部查找元素
例:ele = page('@id=ele_id')
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 超时时间(秒)
:return: 子元素对象
"""
...
@property
def set(self) -> MixTabSetter:
"""返回用于设置的对象"""
...
@property
def wait(self) -> MixTabWaiter:
"""返回用于等待的对象"""
...
@property
def url(self) -> Union[str, None]:
"""返回浏览器当前url"""
...
@property
def _browser_url(self) -> Union[str, None]:
"""返回浏览器当前url"""
...
@property
def title(self) -> str:
"""返回当前页面title"""
...
@property
def raw_data(self) -> Union[str, bytes]:
"""返回页码原始数据数据"""
...
@property
def html(self) -> str:
"""返回页面html文本"""
...
@property
def json(self) -> dict:
"""当返回内容是json格式时,返回对应的字典"""
...
@property
def response(self) -> Response:
"""返回 s 模式获取到的 Response 对象,切换到 s 模式"""
...
@property
def mode(self) -> Literal['s', 'd']:
"""返回当前模式,'s'或'd' """
...
@property
def user_agent(self) -> str:
"""返回user agent"""
...
@property
def session(self) -> Session:
"""返回Session对象,如未初始化则按配置信息创建"""
...
@property
def _session_url(self) -> str:
"""返回 session 保存的url"""
...
@property
def timeout(self) -> float:
"""返回通用timeout设置"""
...
def get(self,
url: str,
show_errmsg: bool = False,
retry: Optional[int] = None,
interval: Optional[float] = None,
timeout: Optional[float] = None,
params: Optional[dict] = None,
data: Any = None,
json: Any = None,
headers: Optional[dict] = None,
cookies: Union[CookieJar, dict] = None,
files: Optional[Any] = None,
auth: Optional[Any] = None,
allow_redirects: bool = True,
proxies: Optional[dict] = None,
hooks: Optional[Any] = None,
stream: bool = None,
verify: Union[bool, str] = None,
cert: [str, Tuple[str, str]] = None) -> Union[bool, None]:
"""跳转到一个url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间
:param params: url中的参数
:param data: 携带的数据
:param json: 要发送的 JSON 数据,会自动设置 Content-Type 为 application/json
:param headers: 请求头
:param cookies: cookies信息
:param files: 要上传的文件,可以是一个字典,其中键是文件名,值是文件对象或文件路径
:param auth: 身份认证信息
:param allow_redirects: 是否允许重定向
:param proxies: 代理信息
:param hooks: 回调方法
:param stream: 是否使用流式传输
:param verify: 是否验证 SSL 证书
:param cert: SSL客户端证书文件的路径(.pem格式),或('cert', 'key')元组
:return: s模式时返回url是否可用,d模式时返回获取到的Response对象
"""
...
def post(self,
url: str,
show_errmsg: bool = False,
retry: Optional[int] = None,
interval: Optional[float] = None,
timeout: Optional[float] = None,
params: Optional[dict] = None,
data: Any = None,
json: Any = None,
headers: Optional[dict] = None,
cookies: Union[CookieJar, dict] = None,
files: Optional[Any] = None,
auth: Optional[Any] = None,
allow_redirects: bool = True,
proxies: Optional[dict] = None,
hooks: Optional[Any] = None,
stream: bool = None,
verify: Union[bool, str] = None,
cert: [str, Tuple[str, str]] = None) -> Response:
"""用post方式跳转到url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间
:param params: url中的参数
:param data: 携带的数据
:param json: 要发送的 JSON 数据,会自动设置 Content-Type 为 application/json
:param headers: 请求头
:param cookies: cookies信息
:param files: 要上传的文件,可以是一个字典,其中键是文件名,值是文件对象或文件路径
:param auth: 身份认证信息
:param allow_redirects: 是否允许重定向
:param proxies: 代理信息
:param hooks: 回调方法
:param stream: 是否使用流式传输
:param verify: 是否验证 SSL 证书
:param cert: SSL客户端证书文件的路径(.pem格式),或('cert', 'key')元组
:return: 获取到的Response对象
"""
...
def ele(self,
locator: Union[Tuple[str, str], str, ChromiumElement, SessionElement],
index: int = 1,
timeout: float = None) -> Union[ChromiumElement, SessionElement]:
"""返回第一个符合条件的元素、属性或节点文本
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: 元素对象或属性、文本节点文本
"""
...
def eles(self,
locator: Union[Tuple[str, str], str],
timeout: float = None) -> Union[ChromiumElementsList, SessionElementsList]:
"""返回页面中所有符合条件的元素、属性或节点文本
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: 元素对象或属性、文本组成的列表
"""
...
def s_ele(self,
locator: Union[Tuple[str, str], str] = None,
index: int = 1,
timeout: float = None) -> SessionElement:
"""查找第一个符合条件的元素以SessionElement形式返回,d模式处理复杂页面时效率很高
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: SessionElement对象或属性、文本
"""
...
def s_eles(self, locator: Union[Tuple[str, str], str], timeout: float = None) -> SessionElementsList:
"""查找所有符合条件的元素以SessionElement形式返回,d模式处理复杂页面时效率很高
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: SessionElement对象或属性、文本组成的列表
"""
...
def change_mode(self, mode: str = None, go: bool = True, copy_cookies: bool = True) -> None:
"""切换模式,接收's'或'd',除此以外的字符串会切换为 d 模式
如copy_cookies为True,切换时会把当前模式的cookies复制到目标模式
切换后,如果go是True,调用相应的get函数使访问的页面同步
:param mode: 模式字符串
:param go: 是否跳转到原模式的url
:param copy_cookies: 是否复制cookies到目标模式
:return: None
"""
...
def cookies_to_session(self, copy_user_agent: bool = True) -> None:
"""把浏览器的cookies复制到session对象
:param copy_user_agent: 是否复制ua信息
:return: None
"""
...
def cookies_to_browser(self) -> None:
"""把session对象的cookies复制到浏览器"""
...
def cookies(self, all_domains: bool = False, all_info: bool = False) -> CookiesList:
"""返回cookies
:param all_domains: 是否返回所有域的cookies
:param all_info: 是否返回所有信息,False则只返回name、value、domain
:return: cookies信息
"""
...
def close(self, others: bool = False) -> None:
"""关闭标签页
:param others: 是否关闭其它,保留自己
:return: None
"""
...
def _find_elements(self,
locator: Union[Tuple[str, str], str, ChromiumElement, SessionElement, ChromiumFrame],
timeout: float,
index: Optional[int] = 1,
relative: bool = False,
raise_err: bool = None) \
-> Union[ChromiumElement, SessionElement, ChromiumFrame, SessionElementsList, ChromiumElementsList]:
"""返回页面中符合条件的元素、属性或节点文本,默认返回第一个
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:param timeout: 查找元素超时时间(秒),d模式专用
:param index: 第几个结果,从1开始,可传入负数获取倒数第几个,为None返回所有
:param relative: MixTab用的表示是否相对定位的参数
:param raise_err: 找不到元素是是否抛出异常,为None时根据全局设置
:return: 元素对象或属性、文本节点文本
"""
...
@@ -0,0 +1,296 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from pathlib import Path
from re import search, DOTALL
from time import sleep
from urllib.parse import urlparse
from requests import Response
from requests.structures import CaseInsensitiveDict
from tldextract import TLDExtract
from .._base.base import BasePage
from .._elements.session_element import SessionElement, make_session_ele
from .._functions.cookies import cookie_to_dict, CookiesList
from .._functions.settings import Settings as _S
from .._functions.web import format_headers
from .._units.setter import SessionPageSetter
class SessionPage(BasePage):
def __init__(self, session_or_options=None, timeout=None):
super().__init__()
self._response = None
self._set = None
self._encoding = None
self._type = 'SessionPage'
self._page = self
self._set_session_options(session_or_options)
self._s_set_runtime_settings()
if timeout is not None: # 即将废弃
self._timeout = timeout
if not self._session:
self._create_session()
def __repr__(self):
return f'<SessionPage url={self.url}>'
def _s_set_runtime_settings(self):
self._timeout = self._session_options.timeout
self._download_path = str(Path(self._session_options.download_path or '.').absolute())
self.retry_times = self._session_options.retry_times
self.retry_interval = self._session_options.retry_interval
def __call__(self, locator, index=1, timeout=None):
return self.ele(locator, index=index)
# -----------------共有属性和方法-------------------
@property
def title(self):
ele = self._ele('xpath://title', raise_err=False)
return ele.text if ele else None
@property
def url(self):
return self._url
@property
def _session_url(self):
return self._url
@property
def raw_data(self):
return self.response.content if self.response else b''
@property
def html(self):
return self.response.text if self.response else ''
@property
def json(self):
try:
return self.response.json()
except Exception:
return None
@property
def user_agent(self):
return self._headers.get('user-agent', '')
@property
def session(self):
return self._session
@property
def response(self):
return self._response
@property
def encoding(self):
return self._encoding
@property
def set(self):
if self._set is None:
self._set = SessionPageSetter(self)
return self._set
@property
def timeout(self):
return self._timeout
def get(self, url, show_errmsg=False, retry=None, interval=None, timeout=None, **kwargs):
"""用get方式跳转到url,可输入文件路径
:param url: 目标url,可指定本地文件路径
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间(秒),为None时使用页面对象timeout属性值
:param kwargs: 连接参数
:return: url是否可用
"""
retry, interval, is_file = self._before_connect(url.replace('file:///', '', 1), retry, interval)
if is_file:
with open(self._url, 'rb') as f:
r = Response()
r._content = f.read()
r.status_code = 200
r.url = self._url
self._response = r
return True
return self._s_connect(self._url, 'get', show_errmsg, retry, interval, **kwargs)
def post(self, url, show_errmsg=False, retry=None, interval=None, **kwargs):
"""用post方式跳转到url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象timeout属性值
:param kwargs: 连接参数
:return: url是否可用
"""
return self._s_connect(url, 'post', show_errmsg, retry, interval, **kwargs)
def ele(self, locator, index=1, timeout=None):
return self._ele(locator, index=index, method='ele()')
def eles(self, locator, timeout=None):
return self._ele(locator, index=None)
def s_ele(self, locator=None, index=1):
return make_session_ele(self) if locator is None else self._ele(locator, index=index, method='s_ele()')
def s_eles(self, locator):
return self._ele(locator, index=None)
def _find_elements(self, locator, timeout, index=1, relative=True, raise_err=None):
return locator if isinstance(locator, SessionElement) else make_session_ele(self, locator, index=index)
def cookies(self, all_domains=False, all_info=False):
if all_domains:
cookies = self.session.cookies
else:
if self.url:
ex_url = TLDExtract(
suffix_list_urls=["https://publicsuffix.org/list/public_suffix_list.dat",
f"file:///{_S.suffixes_list}"]).extract_str(self._session_url)
domain = f'{ex_url.domain}.{ex_url.suffix}' if ex_url.suffix else ex_url.domain
cookies = tuple(c for c in self.session.cookies if domain in c.domain or c.domain == '')
else:
cookies = tuple(c for c in self.session.cookies)
if all_info:
r = CookiesList()
for c in cookies:
r.append(cookie_to_dict(c))
else:
r = CookiesList()
for c in cookies:
c = cookie_to_dict(c)
r.append({'name': c['name'], 'value': c['value'], 'domain': c['domain']})
return r
def close(self):
self._session.close()
if self._response is not None:
self._response.close()
def _s_connect(self, url, mode, show_errmsg=False, retry=None, interval=None, **kwargs):
retry, interval, is_file = self._before_connect(url, retry, interval)
self._response = self._make_response(self._url, mode, retry, interval, show_errmsg, **kwargs)
if self._response is None:
self._url_available = False
else:
if self._response.ok:
self._url_available = True
else:
if show_errmsg:
raise ConnectionError(_S._lang.join(_S._lang.STATUS_CODE_, self._response.status_code))
self._url_available = False
return self._url_available
def _make_response(self, url, mode='get', retry=None, interval=None, show_errmsg=False, **kwargs):
kwargs = CaseInsensitiveDict(kwargs)
if 'headers' in kwargs:
kwargs['headers'] = CaseInsensitiveDict(format_headers(kwargs['headers']))
else:
kwargs['headers'] = CaseInsensitiveDict()
# 设置referer和host值
parsed_url = urlparse(url)
hostname = parsed_url.netloc
scheme = parsed_url.scheme
if not check_headers(kwargs['headers'], self._headers, 'Referer'):
kwargs['headers']['Referer'] = self.url if self.url else f'{scheme}://{hostname}'
elif not kwargs['headers']['Referer']:
kwargs['headers'].pop('Referer')
if not check_headers(kwargs['headers'], self._headers, 'Host'):
kwargs['headers']['Host'] = hostname
elif not kwargs['headers']['Host']:
kwargs['headers'].pop('Host')
if not check_headers(kwargs, self._headers, 'timeout'):
kwargs['timeout'] = self.timeout
h = CaseInsensitiveDict(self._headers)
for k, v in kwargs['headers'].items():
h[k] = v
kwargs['headers'] = h
r = err = None
retry = retry if retry is not None else self.retry_times
interval = interval if interval is not None else self.retry_interval
for i in range(retry + 1):
try:
if mode == 'get':
r = self.session.get(url, **kwargs)
elif mode == 'post':
r = self.session.post(url, **kwargs)
if r and r.content:
if self._encoding:
r.encoding = self._encoding
return r
return set_charset(r)
except Exception as e:
err = e
# if r and r.status_code in (403, 404):
# break
if i < retry:
sleep(interval)
if show_errmsg:
print(f'{_S._lang.RETRY} {url}')
if show_errmsg:
if err:
raise err
elif r is not None:
raise (ConnectionError(_S._lang.join(_S._lang.STATUS_CODE_, r.status_code)) if r.content
else ConnectionError(_S._lang.join(_S._lang.CONTENT_IS_EMPTY)))
else:
raise ConnectionError(_S._lang.join(_S._lang.CONNECT_ERR))
else:
if r is not None:
return r if r.content else None
else:
return None
def check_headers(kwargs, headers, arg):
return arg in kwargs or arg in headers
def set_charset(response):
# 在headers中获取编码
content_type = response.headers.get('content-type', '').lower()
if not content_type.endswith(';'):
content_type += ';'
charset = search(r'charset[=: ]*(.*)?;?', content_type)
if charset:
response.encoding = charset.group(1)
# 在headers中获取不到编码,且如果是网页
elif content_type.replace(' ', '').startswith('text/html'):
re_result = search(b'<meta.*?charset=[ \\\'"]*([^"\\\' />]+).*?>', response.content, DOTALL)
if re_result:
charset = re_result.group(1).decode()
else:
charset = response.apparent_encoding
response.encoding = charset
return response
@@ -0,0 +1,322 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from pathlib import Path
from typing import Any, Union, Tuple, Optional
from requests import Session, Response
from requests.structures import CaseInsensitiveDict
from .._base.base import BasePage
from .._configs.session_options import SessionOptions
from .._elements.session_element import SessionElement
from .._functions.cookies import CookiesList
from .._functions.elements import SessionElementsList
from .._units.setter import SessionPageSetter
class SessionPage(BasePage):
"""SessionPage封装了页面操作的常用功能,使用requests来获取、解析网页"""
_session_options: Optional[SessionOptions] = ...
_url: str = ...
_response: Optional[Response] = ...
_url_available: bool = ...
_timeout: float = ...
retry_times: int = ...
retry_interval: float = ...
_set: Optional[SessionPageSetter] = ...
_encoding: Optional[str] = ...
_page: SessionPage = ...
def __init__(self, session_or_options: Union[Session, SessionOptions] = None):
"""
:param session_or_options: Session对象或SessionOptions对象
"""
...
def _s_set_runtime_settings(self) -> None:
"""设置运行时用到的属性"""
...
def __call__(self,
locator: Union[Tuple[str, str], str, SessionElement],
index: int = 1,
timeout: float = None) -> SessionElement:
"""在内部查找元素
例:ele2 = ele1('@id=ele_id')
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 不起实际作用,用于和ChromiumElement对应,便于无差别调用
:return: SessionElement对象或属性文本
"""
...
@property
def title(self) -> str:
"""返回网页title"""
...
@property
def url(self) -> str:
"""返回当前访问url"""
...
@property
def _session_url(self) -> str:
"""返回当前访问url"""
...
@property
def raw_data(self) -> Union[str, bytes]:
"""返回页面原始数据"""
...
@property
def html(self) -> str:
"""返回页面的html文本"""
...
@property
def json(self) -> Union[dict, None]:
"""当返回内容是json格式时,返回对应的字典,非json格式时返回None"""
...
@property
def user_agent(self) -> str:
"""返回user agent"""
...
@property
def session(self) -> Session:
"""返回Session对象"""
...
@property
def response(self) -> Response:
"""返回访问url得到的Response对象"""
...
@property
def encoding(self) -> str:
"""返回设置的编码,s模式专用"""
...
@property
def set(self) -> SessionPageSetter:
"""返回用于设置的对象"""
...
@property
def timeout(self) -> float:
"""返回超时设置"""
...
def get(self,
url: Union[Path, str],
show_errmsg: bool | None = False,
retry: int | None = None,
interval: float | None = None,
timeout: float | None = None,
params: dict | None = ...,
data: Any = ...,
json: Any = ...,
headers: Union[dict, str, None] = ...,
cookies: Any | None = ...,
files: Any | None = ...,
auth: Any | None = ...,
allow_redirects: bool = ...,
proxies: dict | None = ...,
hooks: Any | None = ...,
stream: Any | None = ...,
verify: Any | None = ...,
cert: Any | None = ...) -> bool:
"""用get方式跳转到url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间
:param params: url中的参数
:param data: 携带的数据
:param json: 要发送的 JSON 数据,会自动设置 Content-Type 为 application/json
:param headers: 请求头
:param cookies: cookies信息
:param files: 要上传的文件,可以是一个字典,其中键是文件名,值是文件对象或文件路径
:param auth: 身份认证信息
:param allow_redirects: 是否允许重定向
:param proxies: 代理信息
:param hooks: 回调方法
:param stream: 是否使用流式传输
:param verify: 是否验证 SSL 证书
:param cert: SSL客户端证书文件的路径(.pem格式),或('cert', 'key')元组
:return: s模式时返回url是否可用,d模式时返回获取到的Response对象
"""
...
def post(self,
url: str,
show_errmsg: bool = False,
retry: int | None = None,
interval: float | None = None,
timeout: float | None = ...,
params: dict | None = ...,
data: Any = ...,
json: Any = ...,
headers: Union[dict, str, None] = ...,
cookies: Any | None = ...,
files: Any | None = ...,
auth: Any | None = ...,
allow_redirects: bool = ...,
proxies: dict | None = ...,
hooks: Any | None = ...,
stream: Any | None = ...,
verify: Any | None = ...,
cert: Any | None = ...) -> bool:
"""用post方式跳转到url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间
:param params: url中的参数
:param data: 携带的数据
:param json: 要发送的 JSON 数据,会自动设置 Content-Type 为 application/json
:param headers: 请求头
:param cookies: cookies信息
:param files: 要上传的文件,可以是一个字典,其中键是文件名,值是文件对象或文件路径
:param auth: 身份认证信息
:param allow_redirects: 是否允许重定向
:param proxies: 代理信息
:param hooks: 回调方法
:param stream: 是否使用流式传输
:param verify: 是否验证 SSL 证书
:param cert: SSL客户端证书文件的路径(.pem格式),或('cert', 'key')元组
:return: s模式时返回url是否可用,d模式时返回获取到的Response对象
"""
...
def ele(self,
locator: Union[Tuple[str, str], str, SessionElement],
index: int = 1,
timeout: float = None) -> SessionElement:
"""返回页面中符合条件的一个元素、属性或节点文本
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 不起实际作用,用于和ChromiumElement对应,便于无差别调用
:return: SessionElement对象或属性、文本
"""
...
def eles(self,
locator: Union[Tuple[str, str], str],
timeout: float = None) -> SessionElementsList:
"""返回页面中所有符合条件的元素、属性或节点文本
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param timeout: 不起实际作用,用于和ChromiumElement对应,便于无差别调用
:return: SessionElement对象或属性、文本组成的列表
"""
...
def s_ele(self,
locator: Union[Tuple[str, str], str, SessionElement] = None,
index: int = 1) -> SessionElement:
"""返回页面中符合条件的一个元素、属性或节点文本
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:return: SessionElement对象或属性、文本
"""
...
def s_eles(self, locator: Union[Tuple[str, str], str]) -> SessionElementsList:
"""返回页面中符合条件的所有元素、属性或节点文本
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:return: SessionElement对象或属性、文本
"""
...
def _find_elements(self,
locator: Union[Tuple[str, str], str, SessionElement],
timeout: float,
index: Optional[int] = 1,
relative: bool = True,
raise_err: bool = None) -> Union[SessionElement, SessionElementsList]:
"""返回页面中符合条件的元素、属性或节点文本,默认返回第一个
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:param timeout: 不起实际作用,用于和父类对应
:param index: 第几个结果,从1开始,可传入负数获取倒数第几个,为None返回所有
:param raise_err: 找不到元素是是否抛出异常,为None时根据全局设置
:return: SessionElement对象
"""
...
def cookies(self,
all_domains: bool = False,
all_info: bool = False) -> CookiesList:
"""返回cookies
:param all_domains: 是否返回所有域的cookies
:param all_info: 是否返回所有信息,False则只返回name、value、domain
:return: cookies组成的列表
"""
...
def close(self) -> None:
"""关闭Session对象"""
...
def _s_connect(self,
url: str,
mode: str,
show_errmsg: bool = False,
retry: int = None,
interval: float = None,
**kwargs) -> bool:
"""执行get或post连接
:param url: 目标url
:param mode: 'get' 或 'post'
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数
:param interval: 重试间隔(秒)
:param kwargs: 连接参数
:return: url是否可用
"""
...
def _make_response(self,
url: str,
mode: str = 'get',
retry: int = None,
interval: float = None,
show_errmsg: bool = False,
**kwargs) -> Response:
"""生成Response对象
:param url: 目标url
:param mode: 'get' 或 'post'
:param show_errmsg: 是否显示和抛出异常
:param kwargs: 其它参数
:return: Response对象
"""
...
def check_headers(kwargs: Union[dict, CaseInsensitiveDict],
headers: Union[dict, CaseInsensitiveDict],
arg: str) -> bool:
"""检查kwargs或headers中是否有arg所示属性
:param kwargs: 要检查的参数dict
:param headers: 要检查的headers
:param arg: 属性名称
:return: 检查结果
"""
...
def set_charset(response: Response) -> Response:
"""设置Response对象的编码
:param response: Response对象
:return: Response对象
"""
...
@@ -0,0 +1,260 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from .chromium_page import ChromiumPage
from .session_page import SessionPage
from .._base.base import BasePage
from .._configs.chromium_options import ChromiumOptions
from .._configs.session_options import SessionOptions
from .._functions.cookies import set_session_cookies, set_tab_cookies
from .._functions.settings import Settings as _S
from .._units.setter import WebPageSetter
class WebPage(SessionPage, ChromiumPage, BasePage):
def __new__(cls, mode='d', timeout=None, chromium_options=None, session_or_options=None):
# 即将废弃timeout
return super().__new__(cls, chromium_options)
def __init__(self, mode='d', timeout=None, chromium_options=None, session_or_options=None):
# 即将废弃timeout
if hasattr(self, '_created'):
return
mode = mode.lower()
if mode not in ('s', 'd'):
raise ValueError(_S._lang.join(_S._lang.INCORRECT_VAL_, 'mode', ALLOW_VAL='"s", "d"', CURR_VAL=mode))
self._d_mode = mode == 'd'
self._has_driver = True
self._has_session = True
super().__init__(session_or_options=session_or_options)
if not chromium_options:
chromium_options = ChromiumOptions(read_file=chromium_options)
chromium_options.set_timeouts(base=self._timeout).set_paths(download_path=self.download_path)
super(SessionPage, self).__init__(addr_or_opts=chromium_options, timeout=timeout) # 即将废弃timeout
self._type = 'WebPage'
self.change_mode(mode, go=False, copy_cookies=False)
def __call__(self, locator, index=1, timeout=None):
if self._d_mode:
return super(SessionPage, self).__call__(locator, index=index, timeout=timeout)
return super().__call__(locator, index=index)
def __repr__(self):
return f'<WebPage browser_id={self.browser.id} tab_id={self.tab_id}>'
@property
def latest_tab(self):
return self.browser._get_tab(id_or_num=self.tab_ids[0], mix=True, as_id=not _S.singleton_tab_obj)
@property
def set(self):
if self._set is None:
self._set = WebPageSetter(self)
return self._set
@property
def url(self):
return self._browser_url if self._d_mode else self._session_url
@property
def _browser_url(self):
return super(SessionPage, self).url if self._driver else None
@property
def title(self):
return super(SessionPage, self).title if self._d_mode else super().title
@property
def raw_data(self):
if self._d_mode:
return super(SessionPage, self).html if self._has_driver else ''
return super().raw_data
@property
def html(self):
if self._d_mode:
return super(SessionPage, self).html if self._has_driver else ''
return super().html
@property
def json(self):
return super(SessionPage, self).json if self._d_mode else super().json
@property
def response(self):
return self._response
@property
def mode(self):
return 'd' if self._d_mode else 's'
@property
def user_agent(self):
return super(SessionPage, self).user_agent if self._d_mode else super().user_agent
@property
def session(self):
if self._session is None:
self._create_session()
return self._session
@property
def _session_url(self):
return self._response.url if self._response else None
@property
def timeout(self):
return self.timeouts.base if self._d_mode else self._timeout
@property
def download_path(self):
return self.browser.download_path
def get(self, url, show_errmsg=False, retry=None, interval=None, timeout=None, **kwargs):
if self._d_mode:
return super(SessionPage, self).get(url, show_errmsg, retry, interval, timeout)
if timeout is None:
timeout = self.timeouts.page_load if self._has_driver else self.timeout
return super().get(url, show_errmsg, retry, interval, timeout, **kwargs)
def post(self, url, show_errmsg=False, retry=None, interval=None, **kwargs):
if self.mode == 'd':
self.cookies_to_session()
super().post(url, show_errmsg, retry, interval, **kwargs)
return self.response
def ele(self, locator, index=1, timeout=None):
return (super(SessionPage, self).ele(locator, index=index, timeout=timeout)
if self._d_mode else super().ele(locator, index=index))
def eles(self, locator, timeout=None):
return super(SessionPage, self).eles(locator, timeout=timeout) if self._d_mode else super().eles(locator)
def s_ele(self, locator=None, index=1, timeout=None):
return (super(SessionPage, self).s_ele(locator, index=index, timeout=timeout)
if self._d_mode else super().s_ele(locator, index=index, timeout=timeout))
def s_eles(self, locator, timeout=None):
return (super(SessionPage, self).s_eles(locator, timeout=timeout)
if self._d_mode else super().s_eles(locator, timeout=timeout))
def change_mode(self, mode=None, go=True, copy_cookies=True):
if mode:
mode = mode.lower()
if mode is not None and ((mode == 'd' and self._d_mode) or (mode == 's' and not self._d_mode)):
return
self._d_mode = not self._d_mode
# s模式转d模式
if self._d_mode:
if self._driver is None or not self._driver.is_running:
self._driver_init(self.tab_id)
self._get_document()
self._url = None if not self._has_driver else super(SessionPage, self).url
self._has_driver = True
if self._session_url:
if copy_cookies:
self.cookies_to_browser()
if go:
self.get(self._session_url)
return
# d模式转s模式
self._has_session = True
self._url = self._session_url
if self._has_driver and self._driver.is_running:
if copy_cookies:
self.cookies_to_session()
if go and not self.get(super(SessionPage, self).url):
raise ConnectionError(_S._lang.join(_S._lang.S_MODE_GET_FAILED))
def cookies_to_session(self, copy_user_agent=True):
if not self._has_session:
return
if copy_user_agent:
user_agent = self._run_cdp('Runtime.evaluate', expression='navigator.userAgent;')['result']['value']
self._headers.update({"User-Agent": user_agent})
set_session_cookies(self.session, super(SessionPage, self).cookies())
def cookies_to_browser(self):
if not self._has_driver:
return
set_tab_cookies(self, super().cookies())
def cookies(self, all_domains=False, all_info=False):
return super(SessionPage, self).cookies(all_domains, all_info) if self._d_mode \
else super().cookies(all_domains, all_info)
def get_tab(self, id_or_num=None, title=None, url=None, tab_type='page', as_id=False):
return self.browser._get_tab(id_or_num=id_or_num, title=title, url=url,
tab_type=tab_type, mix=True, as_id=as_id)
def get_tabs(self, title=None, url=None, tab_type='page', as_id=False):
return self.browser._get_tabs(title=title, url=url, tab_type=tab_type, mix=True, as_id=as_id)
def new_tab(self, url=None, new_window=False, background=False, new_context=False):
return self.browser._new_tab(url=url, new_window=new_window, background=background, new_context=new_context)
def close_driver(self):
if self._has_driver:
self.change_mode('s')
try:
self.driver.run('Browser.close')
except Exception:
pass
self._driver.stop()
self._driver = None
self._has_driver = None
def close_session(self):
if self._has_session:
self.change_mode('d')
self._session.close()
if self._response is not None:
self._response.close()
self._session = None
self._response = None
self._has_session = None
def close(self):
if self._has_driver:
self.browser._close_tab(self)
if self._session:
self._session.close()
if self._response is not None:
self._response.close()
def _find_elements(self, locator, timeout, index=1, relative=False, raise_err=None):
if self._d_mode:
return super(SessionPage, self)._find_elements(locator, timeout=timeout, index=index, relative=relative)
return super()._find_elements(locator, index=index, timeout=timeout)
def quit(self, timeout=5, force=True, del_data=False):
if self._has_session:
self._session.close()
self._session = None
self._response = None
self._has_session = None
if self._has_driver:
super(SessionPage, self).quit(timeout, force, del_data=del_data)
self._driver = None
self._has_driver = None
def _set_session_options(self, session_or_options=None):
if session_or_options is None:
session_or_options = self.browser._session_options or SessionOptions(
read_file=self.browser._session_options is None)
super()._set_session_options(session_or_options)
@@ -0,0 +1,380 @@
# -*- coding:utf-8 -*-
"""
@Author : g1879
@Contact : g1879@qq.com
@Website : https://DrissionPage.cn
@Copyright: (c) 2020 by g1879, Inc. All Rights Reserved.
"""
from http.cookiejar import CookieJar
from typing import Union, Tuple, List, Any, Optional, Literal
from requests import Session, Response
from .chromium_frame import ChromiumFrame
from .chromium_page import ChromiumPage
from .mix_tab import MixTab
from .session_page import SessionPage
from .._base.base import BasePage
from .._configs.chromium_options import ChromiumOptions
from .._configs.session_options import SessionOptions
from .._elements.chromium_element import ChromiumElement
from .._elements.session_element import SessionElement
from .._functions.cookies import CookiesList
from .._functions.elements import SessionElementsList, ChromiumElementsList
from .._units.setter import WebPageSetter
from .._units.waiter import WebPageWaiter
class WebPage(SessionPage, ChromiumPage, BasePage):
"""整合浏览器和request的页面类"""
_d_mode: bool = ...
_set: WebPageSetter = ...
_has_driver: Optional[bool] = ...
_has_session: Optional[bool] = ...
_session_options: Union[SessionOptions, None] = ...
_chromium_options: Union[ChromiumOptions, None] = ...
def __init__(self,
mode: str = 'd',
chromium_options: Union[ChromiumOptions, bool] = None,
session_or_options: Union[Session, SessionOptions, bool] = None) -> None:
"""初始化函数
:param mode: 'd' 或 's',即driver模式和session模式
:param chromium_options: ChromiumOptions对象,传入None时从默认ini文件读取,传入False时不读取ini文件,使用默认配置
:param session_or_options: Session对象或SessionOptions对象,传入None时从默认ini文件读取,传入False时不读取ini文件,使用默认配置
"""
...
def __call__(self,
locator: Union[Tuple[str, str], str, ChromiumElement, SessionElement],
index: int = 1,
timeout: float = None) -> Union[ChromiumElement, SessionElement]:
"""在内部查找元素
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 超时时间(秒)
:return: 子元素对象
"""
...
@property
def latest_tab(self) -> Union[MixTab, WebPage, str]:
"""返回最新的标签页,最新标签页指最后创建或最后被激活的
当Settings.singleton_tab_obj==True时返回Tab对象,否则返回tab id"""
...
@property
def set(self) -> WebPageSetter:
"""返回用于设置的对象"""
...
@property
def wait(self) -> WebPageWaiter:
"""返回用于等待的对象"""
...
@property
def url(self) -> Union[str, None]:
"""返回浏览器当前url"""
...
@property
def _browser_url(self) -> Union[str, None]:
"""返回浏览器当前url"""
...
@property
def title(self) -> str:
"""返回当前页面title"""
...
@property
def raw_data(self) -> Union[str, bytes]:
"""返回页码原始数据数据"""
...
@property
def html(self) -> str:
"""返回页面html文本"""
...
@property
def json(self) -> dict:
"""当返回内容是json格式时,返回对应的字典"""
...
@property
def response(self) -> Response:
"""返回 s 模式获取到的 Response 对象,切换到 s 模式"""
...
@property
def mode(self) -> Literal['s', 'd']:
"""返回当前模式,'s'或'd' """
...
@property
def user_agent(self) -> str:
"""返回user agent"""
...
@property
def session(self) -> Session:
"""返回Session对象,如未初始化则按配置信息创建"""
...
@property
def _session_url(self) -> str:
"""返回 session 保存的url"""
...
@property
def timeout(self) -> float:
"""返回通用timeout设置"""
...
def get(self,
url: str,
show_errmsg: bool = False,
retry: Optional[int] = None,
interval: Optional[float] = None,
timeout: Optional[float] = None,
params: Optional[dict] = None,
data: Any = None,
json: Any = None,
headers: Optional[dict] = None,
cookies: Union[CookieJar, dict] = None,
files: Optional[Any] = None,
auth: Optional[Any] = None,
allow_redirects: bool = True,
proxies: Optional[dict] = None,
hooks: Optional[Any] = None,
stream: bool = None,
verify: Union[bool, str] = None,
cert: [str, Tuple[str, str]] = None) -> Union[bool, None]:
"""跳转到一个url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间
:param params: url中的参数
:param data: 携带的数据
:param json: 要发送的 JSON 数据,会自动设置 Content-Type 为 application/json
:param headers: 请求头
:param cookies: cookies信息
:param files: 要上传的文件,可以是一个字典,其中键是文件名,值是文件对象或文件路径
:param auth: 身份认证信息
:param allow_redirects: 是否允许重定向
:param proxies: 代理信息
:param hooks: 回调方法
:param stream: 是否使用流式传输
:param verify: 是否验证 SSL 证书
:param cert: SSL客户端证书文件的路径(.pem格式),或('cert', 'key')元组
:return: s模式时返回url是否可用,d模式时返回获取到的Response对象
"""
...
def post(self,
url: str,
show_errmsg: bool = False,
retry: Optional[int] = None,
interval: Optional[float] = None,
timeout: Optional[float] = None,
params: Optional[dict] = None,
data: Any = None,
json: Any = None,
headers: Optional[dict] = None,
cookies: Union[CookieJar, dict] = None,
files: Optional[Any] = None,
auth: Optional[Any] = None,
allow_redirects: bool = True,
proxies: Optional[dict] = None,
hooks: Optional[Any] = None,
stream: bool = None,
verify: Union[bool, str] = None,
cert: [str, Tuple[str, str]] = None) -> Response:
"""用post方式跳转到url
:param url: 目标url
:param show_errmsg: 是否显示和抛出异常
:param retry: 重试次数,为None时使用页面对象retry_times属性值
:param interval: 重试间隔(秒),为None时使用页面对象retry_interval属性值
:param timeout: 连接超时时间
:param params: url中的参数
:param data: 携带的数据
:param json: 要发送的 JSON 数据,会自动设置 Content-Type 为 application/json
:param headers: 请求头
:param cookies: cookies信息
:param files: 要上传的文件,可以是一个字典,其中键是文件名,值是文件对象或文件路径
:param auth: 身份认证信息
:param allow_redirects: 是否允许重定向
:param proxies: 代理信息
:param hooks: 回调方法
:param stream: 是否使用流式传输
:param verify: 是否验证 SSL 证书
:param cert: SSL客户端证书文件的路径(.pem格式),或('cert', 'key')元组
:return: 获取到的Response对象
"""
...
def ele(self,
locator: Union[Tuple[str, str], str, ChromiumElement, SessionElement],
index: int = 1,
timeout: float = None) -> Union[ChromiumElement, SessionElement]:
"""返回第一个符合条件的元素、属性或节点文本
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: 元素对象或属性、文本节点文本
"""
...
def eles(self,
locator: Union[Tuple[str, str], str],
timeout: float = None) -> Union[ChromiumElementsList, SessionElementsList]:
"""返回页面中所有符合条件的元素、属性或节点文本
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: 元素对象或属性、文本组成的列表
"""
...
def s_ele(self,
locator: Union[Tuple[str, str], str] = None,
index: int = 1,
timeout: float = None) -> SessionElement:
"""查找第一个符合条件的元素以SessionElement形式返回,d模式处理复杂页面时效率很高
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param index: 获取第几个,从1开始,可传入负数获取倒数第几个
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: SessionElement对象或属性、文本
"""
...
def s_eles(self, locator: Union[Tuple[str, str], str], timeout: float = None) -> SessionElementsList:
"""查找所有符合条件的元素以SessionElement形式返回,d模式处理复杂页面时效率很高
:param locator: 元素的定位信息,可以是loc元组,或查询字符串
:param timeout: 查找元素超时时间(秒),默认与页面等待时间一致
:return: SessionElement对象或属性、文本组成的列表
"""
...
def change_mode(self,
mode: str = None,
go: bool = True,
copy_cookies: bool = True) -> None:
"""切换模式,接收's'或'd',除此以外的字符串会切换为 d 模式
如copy_cookies为True,切换时会把当前模式的cookies复制到目标模式
切换后,如果go是True,调用相应的get函数使访问的页面同步
:param mode: 模式字符串
:param go: 是否跳转到原模式的url
:param copy_cookies: 是否复制cookies到目标模式
:return: None
"""
...
def cookies_to_session(self, copy_user_agent: bool = True) -> None:
"""把driver对象的cookies复制到session对象
:param copy_user_agent: 是否复制ua信息
:return: None
"""
...
def cookies_to_browser(self) -> None:
"""把session对象的cookies复制到浏览器"""
...
def cookies(self,
all_domains: bool = False,
all_info: bool = False) -> CookiesList:
"""返回cookies
:param all_domains: 是否返回所有域的cookies
:param all_info: 是否返回所有信息,False则只返回name、value、domain
:return: cookies信息
"""
...
def get_tab(self,
id_or_num: Union[str, MixTab, int] = None,
title: str = None,
url: str = None,
tab_type: Union[str, list, tuple] = 'page',
as_id: bool = False) -> Union[MixTab, str, None]:
"""获取一个标签页对象,id_or_num不为None时,后面几个参数无效
:param id_or_num: 要获取的标签页id或序号,序号从1开始,可传入负数获取倒数第几个,不是视觉排列顺序,而是激活顺序
:param title: 要匹配title的文本,模糊匹配,为None则匹配所有
:param url: 要匹配url的文本,模糊匹配,为None则匹配所有
:param tab_type: tab类型,可用列表输入多个,如 'page', 'iframe' 等,为None则匹配所有
:param as_id: 是否返回标签页id而不是标签页对象
:return: MixTab对象
"""
...
def get_tabs(self,
title: str = None,
url: str = None,
tab_type: Union[str, list, tuple] = 'page',
as_id: bool = False) -> Union[List[MixTab], List[str]]:
"""查找符合条件的tab,返回它们组成的列表
:param title: 要匹配title的文本,模糊匹配,为None则匹配所有
:param url: 要匹配url的文本,模糊匹配,为None则匹配所有
:param tab_type: tab类型,可用列表输入多个,如 'page', 'iframe' 等,为None则匹配所有
:param as_id: 是否返回标签页id而不是标签页对象
:return: ChromiumTab对象组成的列表
"""
...
def new_tab(self,
url: str = None,
new_window: bool = False,
background: bool = False,
new_context: bool = False) -> MixTab:
"""新建一个标签页
:param url: 新标签页跳转到的网址
:param new_window: 是否在新窗口打开标签页
:param background: 是否不激活新标签页,如new_window为True则无效
:param new_context: 是否创建新的上下文
:return: 新标签页对象
"""
...
def close_driver(self) -> None:
"""关闭driver及浏览器"""
...
def close_session(self) -> None:
"""关闭session"""
...
def close(self) -> None:
"""关闭标签页和Session"""
...
def _find_elements(self,
locator: Union[Tuple[str, str], str, ChromiumElement, SessionElement, ChromiumFrame],
timeout: float,
index: Optional[int] = 1,
relative: bool = False,
raise_err: bool = None) -> Union[
ChromiumElement, SessionElement, ChromiumFrame, SessionElementsList, ChromiumElementsList]:
"""返回页面中符合条件的元素、属性或节点文本,默认返回第一个
:param locator: 元素的定位信息,可以是元素对象,loc元组,或查询字符串
:param timeout: 查找元素超时时间(秒),d模式专用
:param index: 第几个结果,从1开始,可传入负数获取倒数第几个,为None返回所有
:param relative: MixTab用的表示是否相对定位的参数
:param raise_err: 找不到元素是是否抛出异常,为None时根据全局设置
:return: 元素对象或属性、文本节点文本
"""
...
def quit(self,
timeout: float = 5,
force: bool = True,
del_data: bool = False) -> None:
"""关闭浏览器和Session
:param timeout: 等待浏览器关闭超时时间(秒)
:param force: 关闭超时是否强制终止进程
:param del_data: 是否删除用户文件夹
:return: None
"""
...