async def connect(options: dict = None, **kwargs: Any) -> Browser:
"""Connect to the existing chrome.
``browserWSEndpoint`` option is necessary to connect to the chrome. The
format is ``ws://${host}:${port}/devtools/browser/<id>``. This value can
get by :attr:`~pyppeteer.browser.Browser.wsEndpoint`.
Available options are:
* ``browserWSEndpoint`` (str): A browser websocket endpoint to connect to.
(**required**)
* ``ignoreHTTPSErrors`` (bool): Whether to ignore HTTPS errors. Defaults to
``False``.
* ``slowMo`` (int|float): Slow down pyppeteer's by the specified amount of
milliseconds.
* ``logLevel`` (int|str): Log level to print logs. Defaults to same as the
root logger.
* ``loop`` (asyncio.AbstractEventLoop): Event loop (**experimental**).
"""
options = merge_dict(options, kwargs)
logLevel = options.get('logLevel')
if logLevel:
logging.getLogger('pyppeteer').setLevel(logLevel)
browserWSEndpoint = options.get('browserWSEndpoint')
if not browserWSEndpoint:
raise BrowserError('Need `browserWSEndpoint` option.')
connectionDelay = options.get('slowMo', 0)
connection = Connection(browserWSEndpoint,
options.get('loop', asyncio.get_event_loop()),
connectionDelay)
return await Browser.create(connection, options, None,
lambda: connection.send('Browser.close'))
async def connect(options: dict = None, **kwargs: Any) -> Browser:
"""Connect to the existing chrome.
``browserWSEndpoint`` or ``browserURL`` option is necessary to connect to
the chrome. The format of ``browserWSEndpoint`` is
``ws://${host}:${port}/devtools/browser/<id>`` and format of ``browserURL``
is ``http://127.0.0.1:9222```.
The value of ``browserWSEndpoint`` can get by :attr:`~pyppeteer.browser.Browser.wsEndpoint`.
Available options are:
* ``browserWSEndpoint`` (str): A browser websocket endpoint to connect to.
* ``browserURL`` (str): A browser URL to connect to.
* ``ignoreHTTPSErrors`` (bool): Whether to ignore HTTPS errors. Defaults to
``False``.
* ``defaultViewport`` (dict): Set a consistent viewport for each page.
Defaults to an 800x600 viewport. ``None`` disables default viewport.
* ``width`` (int): page width in pixels.
* ``height`` (int): page height in pixels.
* ``deviceScaleFactor`` (int|float): Specify device scale factor (can be
thought as dpr). Defaults to ``1``.
* ``isMobile`` (bool): Whether the ``meta viewport`` tag is taken into
account. Defaults to ``False``.
* ``hasTouch`` (bool): Specify if viewport supports touch events.
Defaults to ``False``.
* ``isLandscape`` (bool): Specify if viewport is in landscape mode.
Defaults to ``False``.
* ``slowMo`` (int|float): Slow down pyppeteer's by the specified amount of
milliseconds.
* ``logLevel`` (int|str): Log level to print logs. Defaults to same as the
root logger.
* ``loop`` (asyncio.AbstractEventLoop): Event loop (**experimental**).
"""
options = merge_dict(options, kwargs)
logLevel = options.get('logLevel')
if logLevel:
logging.getLogger('pyppeteer').setLevel(logLevel)
browserWSEndpoint = options.get('browserWSEndpoint')
if not browserWSEndpoint:
browserURL = options.get('browserURL')
if not browserURL:
raise BrowserError(
'Need `browserWSEndpoint` or `browserURL` option.')
browserWSEndpoint = get_ws_endpoint(browserURL)
connectionDelay = options.get('slowMo', 0)
connection = Connection(browserWSEndpoint,
options.get('loop', asyncio.get_event_loop()),
connectionDelay)
browserContextIds = (await
connection.send('Target.getBrowserContexts')).get(
'browserContextIds', [])
ignoreHTTPSErrors = bool(options.get('ignoreHTTPSErrors', False))
defaultViewport = options.get('defaultViewport', {
'width': 800,
'height': 600
})
return await Browser.create(connection, browserContextIds,
ignoreHTTPSErrors, defaultViewport, None,
lambda: connection.send('Browser.close'))
async def connect(options: dict = None, **kwargs: Any) -> Browser:
"Connect to the existing chrome.\n\n ``browserWSEndpoint`` option is necessary to connect to the chrome. The\n format is ``ws://${host}:${port}/devtools/browser/<id>``. This value can\n get by :attr:`~pyppeteer.browser.Browser.wsEndpoint`.\n\n Available options are:\n\n * ``browserWSEndpoint`` (str): A browser websocket endpoint to connect to.\n (**required**)\n * ``ignoreHTTPSErrors`` (bool): Whether to ignore HTTPS errors. Defaults to\n ``False``.\n * ``slowMo`` (int|float): Slow down pyppeteer's by the specified amount of\n milliseconds.\n "
options = merge_dict(options, kwargs)
browserWSEndpoint = options.get('browserWSEndpoint')
if (not browserWSEndpoint):
raise BrowserError('Need `browserWSEndpoint` option.')
connection = Connection(browserWSEndpoint)
return (await Browser.create(connection, options, None,
(lambda: connection.send('Browser.close'))))
async def launch(self) -> Browser:
"""Start chrome process and return `Browser` object."""
env = self.options.get('env')
self.chromeClosed = False
self.connection: Optional[Connection] = None
self.proc = subprocess.Popen(
self.cmd,
stdout=subprocess.DEVNULL,
stderr=subprocess.DEVNULL,
env=env,
)
def _close_process(*args: Any, **kwargs: Any) -> None:
if not self.chromeClosed:
asyncio.get_event_loop().run_until_complete(self.killChrome())
# dont forget to close browser process
atexit.register(_close_process)
if self.options.get('handleSIGINT', True):
signal.signal(signal.SIGINT, _close_process)
if self.options.get('handleSIGTERM', True):
signal.signal(signal.SIGTERM, _close_process)
if self.options.get('handleSIGHUP', True):
signal.signal(signal.SIGHUP, _close_process)
connectionDelay = self.options.get('slowMo', 0)
self.browserWSEndpoint = self._get_ws_endpoint()
logger.info(f'Browser listening on: {self.browserWSEndpoint}')
self.connection = Connection(self.browserWSEndpoint, connectionDelay)
return await Browser.create(self.connection, self.options, self.proc,
self.killChrome)
async def launch(self):
env = self.options.get("env")
self.proc = await asyncio.subprocess.create_subprocess_exec(
*self.cmd,
stdout=asyncio.subprocess.DEVNULL,
stderr=asyncio.subprocess.DEVNULL,
env=env,
)
self.chromeClosed = False
self.connection = None
def _close_process(*args, **kwargs):
if not self.chromeClosed:
asyncio.ensure_future(self.killChrome())
# dont forget to close browser process
atexit.register(_close_process)
if self.options.get("handleSIGINT", True):
signal.signal(signal.SIGINT, _close_process)
if self.options.get("handleSIGTERM", True):
signal.signal(signal.SIGTERM, _close_process)
if not sys.platform.startswith("win"):
# SIGHUP is not defined on windows
if self.options.get("handleSIGHUP", True):
signal.signal(signal.SIGHUP, _close_process)
connectionDelay = self.options.get("slowMo", 0.1)
self.browserWSEndpoint = self._get_ws_endpoint()
self.connection = Connection(self.browserWSEndpoint, connectionDelay)
return await Browser.create(self.connection, self.options, self.proc,
self.killChrome)
async def launch(self) -> Browser:
'Start chrome process and return `Browser` object.'
env = self.options.get('env')
self.chromeClosed = False
self.connection = None
self.proc = subprocess.Popen(self.cmd,
stdout=subprocess.DEVNULL,
stderr=subprocess.DEVNULL,
env=env)
def _close_process(*args: Any, **kwargs: Any) -> None:
if (not self.chromeClosed):
asyncio.get_event_loop().run_until_complete(self.killChrome())
atexit.register(_close_process)
if self.options.get('handleSIGINT', True):
signal.signal(signal.SIGINT, _close_process)
if self.options.get('handleSIGTERM', True):
signal.signal(signal.SIGTERM, _close_process)
if (not sys.platform.startswith('win')):
if self.options.get('handleSIGHUP', True):
signal.signal(signal.SIGHUP, _close_process)
connectionDelay = self.options.get('slowMo', 0)
self.browserWSEndpoint = self._get_ws_endpoint()
logger.info(''.join(
['Browser listening on: ', '{}'.format(self.browserWSEndpoint)]))
self.connection = Connection(self.browserWSEndpoint, connectionDelay)
return (await Browser.create(self.connection, self.options, self.proc,
self.killChrome))
def launch(self) -> Browser:
"""Start chromium process."""
self.proc = subprocess.Popen(
self.cmd,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
)
atexit.register(self.killChrome)
import time
for _ in range(100):
# wait for DevTools port to open for at least 10sec
# setting timeout timer is bettter
time.sleep(0.1)
if self.proc.poll() is not None:
raise BrowserError('Unexpectedly chrome process closed with '
f'return code: {self.proc.returncode}')
msg = self.proc.stdout.readline().decode()
if not msg:
continue
m = re.match(r'DevTools listening on (ws://.*)$', msg)
if m is not None:
break
else:
# This block called only when `for`-loop does not `break`
raise BrowserError('Failed to connect DevTools port.')
logger.debug(m.group(0))
connectionDelay = self.options.get('slowMo', 0)
connection = Connection(m.group(1).strip(), connectionDelay)
return Browser(connection, self.options.get('ignoreHTTPSErrors',
False), self.killChrome)
async def connect(options: dict = None, **kwargs: Any) -> Browser:
"""Connect to the existing chrome.
``browserWSEndpoint`` option is necessary to connect to the chrome. The
format is ``ws://${host}:${port}/devtools/browser/<id>``. This value can
get by :attr:`~pyppeteer.browser.Browser.wsEndpoint`.
Available options are:
* ``browserWSEndpoint`` (str): A browser websocket endpoint to connect to.
(**required**)
* ``ignoreHTTPSErrors`` (bool): Whether to ignore HTTPS errors. Defaults to
``False``.
* ``slowMo`` (int|float): Slow down pyppeteer's by the specified amount of
milliseconds.
"""
options = merge_dict(options, kwargs)
browserWSEndpoint = options.get('browserWSEndpoint')
if not browserWSEndpoint:
raise BrowserError('Need `browserWSEndpoint` option.')
connection = Connection(browserWSEndpoint)
return await Browser.create(
connection, options, None, lambda: connection.send('Browser.close'))
async def connect(options: dict = None, **kwargs: Any) -> Browser:
"""Connect to the existing chrome.
``browserWSEndpoint`` option is necessary to connect to the chrome. The
format is ``ws://${host}:${port}/devtools/browser/<id>``. This value can
get by :attr:`~pyppeteer.browser.Browser.wsEndpoint`.
Available options are:
* ``browserWSEndpoint`` (str): A browser websocket endpoint to connect to.
(**required**)
* ``ignoreHTTPSErrors`` (bool): Whether to ignore HTTPS errors. Defaults to
``False``.
* ``slowMo`` (int|float): Slow down pyppeteer's by the specified amount of
milliseconds.
* ``logLevel`` (int|str): Log level to print logs. Defaults to same as the
root logger.
* ``loop`` (asyncio.AbstractEventLoop): Event loop (**experimental**).
"""
options = merge_dict(options, kwargs)
logLevel = options.get('logLevel')
if logLevel:
logging.getLogger('pyppeteer').setLevel(logLevel)
browserWSEndpoint = options.get('browserWSEndpoint')
if not browserWSEndpoint:
raise BrowserError('Need `browserWSEndpoint` option.')
connectionDelay = options.get('slowMo', 0)
connection = Connection(browserWSEndpoint,
options.get('loop', asyncio.get_event_loop()),
connectionDelay)
browserContextIds = (await connection.send('Target.getBrowserContexts')
).get('browserContextIds', [])
ignoreHTTPSErrors = bool(options.get('ignoreHTTPSErrors', False))
return await Browser.create(
connection, browserContextIds, ignoreHTTPSErrors, True, None,
lambda: connection.send('Browser.close'))
async def launch(self):
env = self.options.get("env")
self.proc = await asyncio.subprocess.create_subprocess_exec(
*self.cmd,
stdout=asyncio.subprocess.DEVNULL,
stderr=asyncio.subprocess.DEVNULL,
env=env,
)
self.chromeClosed = False
self.connection = None
# Signal handlers for exits used to be here
connectionDelay = self.options.get("slowMo", 0.1)
self.browserWSEndpoint = self._get_ws_endpoint()
self.connection = Connection(self.browserWSEndpoint, connectionDelay)
return await Browser.create(self.connection, self.options, self.proc,
self.killChrome)
async def launch(self) -> Browser: # noqa: C901
"""Start chrome process and return `Browser` object."""
self.chromeClosed = False
self.connection: Optional[Connection] = None
options = dict()
options['env'] = self.env
if not self.dumpio:
# discard stdout, it's never read in any case.
options['stdout'] = subprocess.DEVNULL
options['stderr'] = subprocess.STDOUT
self.proc = subprocess.Popen( # type: ignore
self.cmd,
**options,
)
def _close_process(*args: Any, **kwargs: Any) -> None:
if not self.chromeClosed:
self._loop.run_until_complete(self.killChrome())
# don't forget to close browser process
if self.autoClose:
atexit.register(_close_process)
if self.handleSIGINT:
signal.signal(signal.SIGINT, _close_process)
if self.handleSIGTERM:
signal.signal(signal.SIGTERM, _close_process)
if not sys.platform.startswith('win'):
# SIGHUP is not defined on windows
if self.handleSIGHUP:
signal.signal(signal.SIGHUP, _close_process)
connectionDelay = self.slowMo
self.browserWSEndpoint = get_ws_endpoint(self.url)
logger.info(f'Browser listening on: {self.browserWSEndpoint}')
self.connection = Connection(
self.browserWSEndpoint,
self._loop,
connectionDelay,
)
browser = await Browser.create(self.connection, [],
self.ignoreHTTPSErrors,
self.defaultViewport, self.proc,
self.killChrome)
await self.ensureInitialPage(browser)
return browser
async def launch(self) -> Browser: # noqa: C901
"""Start chrome process and return `Browser` object."""
self.chromeClosed = False
self.connection: Optional[Connection] = None
options = dict()
options['env'] = self.options.get('env')
if not self.options.get('dumpio'):
options['stdout'] = subprocess.PIPE
options['stderr'] = subprocess.STDOUT
self.proc = subprocess.Popen( # type: ignore
self.cmd,
**options,
)
def _close_process(*args: Any, **kwargs: Any) -> None:
if not self.chromeClosed:
self._loop.run_until_complete(self.killChrome())
# don't forget to close browser process
if self.options.get('autoClose', True):
atexit.register(_close_process)
if self.options.get('handleSIGINT', True):
signal.signal(signal.SIGINT, _close_process)
if self.options.get('handleSIGTERM', True):
signal.signal(signal.SIGTERM, _close_process)
if not sys.platform.startswith('win'):
# SIGHUP is not defined on windows
if self.options.get('handleSIGHUP', True):
signal.signal(signal.SIGHUP, _close_process)
connectionDelay = self.options.get('slowMo', 0)
self.browserWSEndpoint = self._get_ws_endpoint()
logger.info(f'Browser listening on: {self.browserWSEndpoint}')
self.connection = Connection(self.browserWSEndpoint, self._loop,
connectionDelay)
ignoreHTTPSErrors = bool(self.options.get('ignoreHTTPSErrors', False))
setDefaultViewport = not self.options.get('appMode', False)
browser = await Browser.create(self.connection, [], ignoreHTTPSErrors,
setDefaultViewport, self.proc,
self.killChrome)
await self.ensureInitialPage(browser)
return browser
async def launch(self) -> Browser:
"""Start chrome process and return `Browser` object."""
env = self.options.get('env', {})
self.chromeClosed = False
self.connection: Optional[Connection] = None
self.proc = subprocess.Popen(
self.cmd,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
env=env,
)
def _close_process() -> None:
if not self.chromeClosed:
asyncio.get_event_loop().run_until_complete(self.killChrome())
# dont forget to close browser process
atexit.register(_close_process)
import time
for _ in range(100):
# wait for DevTools port to open for at least 10sec
# setting timeout timer is bettter
time.sleep(0.1)
if self.proc.poll() is not None:
self._cleanup_tmp_user_data_dir()
raise BrowserError('Unexpectedly chrome process closed with '
f'return code: {self.proc.returncode}')
msg = self.proc.stdout.readline().decode()
if not msg:
continue
m = re.match(r'DevTools listening on (ws://.*)$', msg)
if m is not None:
break
else:
# This block called only when `for`-loop does not `break`
raise BrowserError('Failed to connect DevTools port.')
logger.debug(m.group(0))
connectionDelay = self.options.get('slowMo', 0)
self.browserWSEndpoint = m.group(1).strip()
self.connection = Connection(self.browserWSEndpoint, connectionDelay)
return await Browser.create(self.connection, self.options,
self.killChrome)
async def launch_chrome(self):
self.chromeClosed = False
self.connection = None
self.proc = subprocess.Popen(
self._launch_cmd(), stdout=subprocess.PIPE, stderr=subprocess.STDOUT
)
# Signal handlers for exits used to be here
connectionDelay = self.slowMo
self.browserWSEndpoint = launcher.get_ws_endpoint(self.url)
self.connection = Connection(
self.browserWSEndpoint, self._loop, connectionDelay
)
return await Browser.create(
connection=self.connection,
contextIds=[],
ignoreHTTPSErrors=self.ignoreHTTPSErrors,
defaultViewport=self.defaultViewport,
process=self.proc,
closeCallBack=self.killChrome,
)
async def launch(self):
self.chromeClosed = False
self.connection = None
options = dict()
options['env'] = self.env
if not self.dumpio:
options['stdout'] = asyncio.subprocess.PIPE
options['stderr'] = asyncio.subprocess.STDOUT
self.proc = await asyncio.subprocess.create_subprocess_exec(
*self.cmd, **options)
# Signal handlers for exits used to be here
connectionDelay = self.slowMo
self.browserWSEndpoint = await self._get_ws_endpoint()
self.connection = Connection(self.browserWSEndpoint, self._loop,
connectionDelay)
browser = await Browser.create(
self.connection, [], self.ignoreHTTPSErrors, self.defaultViewport,
self.proc, self.killChrome)
await self.ensureInitialPage(browser)
return browser
def launch(self) -> Browser:
"""Start chromium process."""
self.proc = subprocess.Popen(
self.cmd,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
)
atexit.register(self.killChrome)
import time
while True:
time.sleep(0.1)
msg = self.proc.stdout.readline().decode()
if not msg:
continue
m = re.match(r'DevTools listening on (ws://.*)$', msg)
if m is not None:
break
logger.debug(m.group(0))
self.url = m.group(1).strip()
connectionDelay = self.options.get('slowMo', 0)
connection = Connection(self.url, connectionDelay)
return Browser(connection, self.options.get('ignoreHTTPSErrors',
False), self.killChrome)
async def launch(self) -> Browser: # noqa: C901
"""Start chrome process and return `Browser` object."""
self.chromeClosed = False
self.connection: Optional[Connection] = None
options = {}
options['env'] = self.options.get('env')
if not self.options.get('dumpio'):
options['stdout'] = subprocess.DEVNULL
options['stderr'] = subprocess.DEVNULL
self.proc = subprocess.Popen( # type: ignore
self.cmd,
**options,
)
def _close_process(*args: Any, **kwargs: Any) -> None:
if not self.chromeClosed:
asyncio.get_event_loop().run_until_complete(self.killChrome())
# dont forget to close browser process
atexit.register(_close_process)
if self.options.get('handleSIGINT', True):
signal.signal(signal.SIGINT, _close_process)
if self.options.get('handleSIGTERM', True):
signal.signal(signal.SIGTERM, _close_process)
if not sys.platform.startswith('win'):
# SIGHUP is not defined on windows
if self.options.get('handleSIGHUP', True):
signal.signal(signal.SIGHUP, _close_process)
connectionDelay = self.options.get('slowMo', 0)
self.browserWSEndpoint = self._get_ws_endpoint()
logger.info(f'Browser listening on: {self.browserWSEndpoint}')
self.connection = Connection(self.browserWSEndpoint, connectionDelay)
return await Browser.create(self.connection, self.options, self.proc,
self.killChrome)
async def launch(self) -> Browser:
'Start chrome process and return `Browser` object.'
self.chromeClosed = False
self.connection = None
options = dict()
options['env'] = self.options.get('env')
if (not self.options.get('dumpio')):
options['stdout'] = subprocess.PIPE
options['stderr'] = subprocess.STDOUT
self.proc = subprocess.Popen(self.cmd, **options)
def _close_process(*args: Any, **kwargs: Any) -> None:
if (not self.chromeClosed):
self._loop.run_until_complete(self.killChrome())
if self.options.get('autoClose', True):
atexit.register(_close_process)
if self.options.get('handleSIGINT', True):
signal.signal(signal.SIGINT, _close_process)
if self.options.get('handleSIGTERM', True):
signal.signal(signal.SIGTERM, _close_process)
if (not sys.platform.startswith('win')):
if self.options.get('handleSIGHUP', True):
signal.signal(signal.SIGHUP, _close_process)
connectionDelay = self.options.get('slowMo', 0)
self.browserWSEndpoint = self._get_ws_endpoint()
logger.info(''.join(
['Browser listening on: ', '{}'.format(self.browserWSEndpoint)]))
self.connection = Connection(self.browserWSEndpoint, self._loop,
connectionDelay)
ignoreHTTPSErrors = bool(self.options.get('ignoreHTTPSErrors', False))
setDefaultViewport = (not self.options.get('appMode', False))
browser = (await Browser.create(self.connection, [], ignoreHTTPSErrors,
setDefaultViewport, self.proc,
self.killChrome))
(await self.ensureInitialPage(browser))
return browser
async def to_pdf():
if app.config.get('CHROME_URL'):
version_url = urljoin(app.config['CHROME_URL'], 'json/version')
data = requests.get(version_url).json()
con = Connection(data['webSocketDebuggerUrl'])
browser = await Browser.create(con)
else:
chrome_dir = 'var/pyppeteer'
if not os.path.exists(chrome_dir):
os.mkdir(chrome_dir)
tmp_chrome_dir = tempfile.mkdtemp(dir=chrome_dir)
browser = await launch(executablePath='google-chrome',
userDataDir=tmp_chrome_dir)
page = await browser.newPage()
async def request_intercepted(request):
app.logger.debug('Intercepted URL: %s', request.url)
if request.url == url:
await request.respond({'body': html})
else:
await request.continue_()
page.on('request', request_intercepted)
await page.setRequestInterception(True)
await page.goto(url)
pdf = await page.pdf(format='A4')
await browser.close()
if not app.config.get('CHROME_URL'):
shutil.rmtree(tmp_chrome_dir)
return pdf
