diff --git a/docs/topics/dynamic-content.rst b/docs/topics/dynamic-content.rst index aa32c8943..ea5d06210 100644 --- a/docs/topics/dynamic-content.rst +++ b/docs/topics/dynamic-content.rst @@ -246,24 +246,46 @@ Using a headless browser ======================== A `headless browser`_ is a special web browser that provides an API for -automation. +automation. By installing the :ref:`asyncio reactor `, +it is possible to integrate ``asyncio``-based libraries which handle headless browsers. -The easiest way to use a headless browser with Scrapy is to use Selenium_, -along with `scrapy-selenium`_ for seamless integration. +One such library is `playwright-python`_ (an official Python port of `playwright`_). +The following is a simple snippet to illustrate its usage within a Scrapy spider:: + import scrapy + from playwright.async_api import async_playwright + + class PlaywrightSpider(scrapy.Spider): + name = "playwright" + start_urls = ["data:,"] # avoid using the default Scrapy downloader + + async def parse(self, response): + async with async_playwright() as pw: + browser = await pw.chromium.launch() + page = await browser.new_page() + await page.goto("https:/example.org") + title = await page.title() + return {"title": title} + + +However, using `playwright-python`_ directly as in the above example +circumvents most of the Scrapy components (middlewares, dupefilter, etc). +We recommend using `scrapy-playwright`_ for a better integration. .. _AJAX: https://en.wikipedia.org/wiki/Ajax_%28programming%29 -.. _chompjs: https://github.com/Nykakin/chompjs .. _CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets +.. _JavaScript: https://en.wikipedia.org/wiki/JavaScript +.. _Splash: https://github.com/scrapinghub/splash +.. _chompjs: https://github.com/Nykakin/chompjs .. _curl: https://curl.haxx.se/ .. _headless browser: https://en.wikipedia.org/wiki/Headless_browser -.. _JavaScript: https://en.wikipedia.org/wiki/JavaScript .. _js2xml: https://github.com/scrapinghub/js2xml +.. _playwright-python: https://github.com/microsoft/playwright-python +.. _playwright: https://github.com/microsoft/playwright +.. _pyppeteer: https://pyppeteer.github.io/pyppeteer/ .. _pytesseract: https://github.com/madmaze/pytesseract -.. _scrapy-selenium: https://github.com/clemfromspace/scrapy-selenium +.. _scrapy-playwright: https://github.com/scrapy-plugins/scrapy-playwright .. _scrapy-splash: https://github.com/scrapy-plugins/scrapy-splash -.. _Selenium: https://www.selenium.dev/ -.. _Splash: https://github.com/scrapinghub/splash .. _tabula-py: https://github.com/chezou/tabula-py .. _wget: https://www.gnu.org/software/wget/ -.. _wgrep: https://github.com/stav/wgrep \ No newline at end of file +.. _wgrep: https://github.com/stav/wgrep diff --git a/docs/topics/feed-exports.rst b/docs/topics/feed-exports.rst index 216a8bc52..2b3217d62 100644 --- a/docs/topics/feed-exports.rst +++ b/docs/topics/feed-exports.rst @@ -135,7 +135,7 @@ Here are some examples to illustrate: - ``s3://mybucket/scraping/feeds/%(name)s/%(time)s.json`` -.. note:: :ref:`Spider arguments ` become spider attributes, hence +.. note:: :ref:`Spider arguments ` become spider attributes, hence they can also be used as storage URI parameters. @@ -200,6 +200,9 @@ passed through the following settings: - :setting:`AWS_ACCESS_KEY_ID` - :setting:`AWS_SECRET_ACCESS_KEY` +- :setting:`AWS_SESSION_TOKEN` (only needed for `temporary security credentials`_) + +.. _temporary security credentials: https://docs.aws.amazon.com/general/latest/gr/aws-sec-cred-types.html#temporary-access-keys You can also define a custom ACL and custom endpoint for exported feeds using this setting: @@ -357,7 +360,7 @@ For instance:: 'item_export_kwargs': { 'export_empty_fields': True, }, - }, + }, '/home/user/documents/items.xml': { 'format': 'xml', 'fields': ['name', 'price'], diff --git a/docs/topics/media-pipeline.rst b/docs/topics/media-pipeline.rst index 3438cb637..46bd2859b 100644 --- a/docs/topics/media-pipeline.rst +++ b/docs/topics/media-pipeline.rst @@ -111,25 +111,82 @@ For the Images Pipeline, set the :setting:`IMAGES_STORE` setting:: IMAGES_STORE = '/path/to/valid/dir' +.. _topics-file-naming: + +File Naming +=========== + +Default File Naming +------------------- + +By default, files are stored using an `SHA-1 hash`_ of their URLs for the file names. + +For example, the following image URL:: + + http://www.example.com/image.jpg + +Whose ``SHA-1 hash`` is:: + + 3afec3b4765f8f0a07b78f98c07b83f013567a0a + +Will be downloaded and stored using your chosen :ref:`storage method ` and the following file name:: + + 3afec3b4765f8f0a07b78f98c07b83f013567a0a.jpg + +Custom File Naming +------------------- + +You may wish to use a different calculated file name for saved files. +For example, classifying an image by including meta in the file name. + +Customize file names by overriding the ``file_path`` method of your +media pipeline. + +For example, an image pipeline with image URL:: + + http://www.example.com/product/images/large/front/0000000004166 + +Can be processed into a file name with a condensed hash and the perspective +``front``:: + + 00b08510e4_front.jpg + +By overriding ``file_path`` like this: + +.. code-block:: python + + import hashlib + from os.path import splitext + + def file_path(self, request, response=None, info=None, *, item=None): + image_url_hash = hashlib.shake_256(request.url.encode()).hexdigest(5) + image_perspective = request.url.split('/')[-2] + image_filename = f'{image_url_hash}_{image_perspective}.jpg' + + return image_filename + +.. warning:: + If your custom file name scheme relies on meta data that can vary between + scrapes it may lead to unexpected re-downloading of existing media using + new file names. + + For example, if your custom file name scheme uses a product title and the + site changes an item's product title between scrapes, Scrapy will re-download + the same media using updated file names. + +For more information about the ``file_path`` method, see :ref:`topics-media-pipeline-override`. + +.. _topics-supported-storage: + Supported Storage ================= File system storage ------------------- -The files are stored using a `SHA1 hash`_ of their URLs for the file names. +File system storage will save files to the following path:: -For example, the following image URL:: - - http://www.example.com/image.jpg - -Whose ``SHA1 hash`` is:: - - 3afec3b4765f8f0a07b78f98c07b83f013567a0a - -Will be downloaded and stored in the following file:: - - /full/3afec3b4765f8f0a07b78f98c07b83f013567a0a.jpg + /full/ Where: @@ -139,6 +196,9 @@ Where: * ``full`` is a sub-directory to separate full images from thumbnails (if used). For more info see :ref:`topics-images-thumbnails`. +* ```` is the file name assigned to the file. For more info see :ref:`topics-file-naming`. + + .. _media-pipeline-ftp: FTP server storage @@ -353,9 +413,9 @@ Where: * ```` is the one specified in the :setting:`IMAGES_THUMBS` dictionary keys (``small``, ``big``, etc) -* ```` is the `SHA1 hash`_ of the image url +* ```` is the `SHA-1 hash`_ of the image url -.. _SHA1 hash: https://en.wikipedia.org/wiki/SHA_hash_functions +.. _SHA-1 hash: https://en.wikipedia.org/wiki/SHA_hash_functions Example of image files stored using ``small`` and ``big`` thumbnail names:: diff --git a/docs/topics/request-response.rst b/docs/topics/request-response.rst index a6a3daf31..d3e08efd4 100644 --- a/docs/topics/request-response.rst +++ b/docs/topics/request-response.rst @@ -670,9 +670,6 @@ Response objects .. autoclass:: Response - A :class:`Response` object represents an HTTP response, which is usually - downloaded (by the Downloader) and fed to the Spiders for processing. - :param url: the URL of this response :type url: str @@ -829,6 +826,8 @@ Response objects handlers, i.e. for ``http(s)`` responses. For other handlers, :attr:`protocol` is always ``None``. + .. autoattribute:: Response.attributes + .. method:: Response.copy() Returns a new Response which is a copy of this Response. @@ -925,6 +924,8 @@ TextResponse objects A :class:`~scrapy.Selector` instance using the response as target. The selector is lazily instantiated on first access. + .. autoattribute:: TextResponse.attributes + :class:`TextResponse` objects support the following methods in addition to the standard :class:`Response` ones: diff --git a/docs/topics/settings.rst b/docs/topics/settings.rst index 1290b4a5e..1a1a833df 100644 --- a/docs/topics/settings.rst +++ b/docs/topics/settings.rst @@ -204,6 +204,19 @@ Default: ``None`` The AWS secret key used by code that requires access to `Amazon Web services`_, such as the :ref:`S3 feed storage backend `. +.. setting:: AWS_SESSION_TOKEN + +AWS_SESSION_TOKEN +----------------- + +Default: ``None`` + +The AWS security token used by code that requires access to `Amazon Web services`_, +such as the :ref:`S3 feed storage backend `, when using +`temporary security credentials`_. + +.. _temporary security credentials: https://docs.aws.amazon.com/general/latest/gr/aws-sec-cred-types.html#temporary-access-keys + .. setting:: AWS_ENDPOINT_URL AWS_ENDPOINT_URL diff --git a/scrapy/core/downloader/handlers/s3.py b/scrapy/core/downloader/handlers/s3.py index 1966570d4..51ca1ed5e 100644 --- a/scrapy/core/downloader/handlers/s3.py +++ b/scrapy/core/downloader/handlers/s3.py @@ -1,5 +1,3 @@ -from urllib.parse import unquote - from scrapy.core.downloader.handlers.http import HTTPDownloadHandler from scrapy.exceptions import NotConfigured from scrapy.utils.boto import is_botocore_available @@ -12,6 +10,7 @@ class S3DownloadHandler: def __init__(self, settings, *, crawler=None, aws_access_key_id=None, aws_secret_access_key=None, + aws_session_token=None, httpdownloadhandler=HTTPDownloadHandler, **kw): if not is_botocore_available(): raise NotConfigured('missing botocore library') @@ -20,6 +19,8 @@ class S3DownloadHandler: aws_access_key_id = settings['AWS_ACCESS_KEY_ID'] if not aws_secret_access_key: aws_secret_access_key = settings['AWS_SECRET_ACCESS_KEY'] + if not aws_session_token: + aws_session_token = settings['AWS_SESSION_TOKEN'] # If no credentials could be found anywhere, # consider this an anonymous connection request by default; @@ -38,7 +39,7 @@ class S3DownloadHandler: if not self.anon: SignerCls = botocore.auth.AUTH_TYPE_MAPS['s3'] self._signer = SignerCls(botocore.credentials.Credentials( - aws_access_key_id, aws_secret_access_key)) + aws_access_key_id, aws_secret_access_key, aws_session_token)) _http_handler = create_instance( objcls=httpdownloadhandler, @@ -59,7 +60,7 @@ class S3DownloadHandler: url = f'{scheme}://{bucket}.s3.amazonaws.com{path}' if self.anon: request = request.replace(url=url) - elif self._signer is not None: + else: import botocore.awsrequest awsrequest = botocore.awsrequest.AWSRequest( method=request.method, @@ -69,14 +70,4 @@ class S3DownloadHandler: self._signer.add_auth(awsrequest) request = request.replace( url=url, headers=awsrequest.headers.items()) - else: - signed_headers = self.conn.make_request( - method=request.method, - bucket=bucket, - key=unquote(p.path), - query_args=unquote(p.query), - headers=request.headers, - data=request.body, - ) - request = request.replace(url=url, headers=signed_headers) return self._download_http(request, spider) diff --git a/scrapy/extensions/feedexport.py b/scrapy/extensions/feedexport.py index bd4808e2b..564c736f2 100644 --- a/scrapy/extensions/feedexport.py +++ b/scrapy/extensions/feedexport.py @@ -154,13 +154,14 @@ class FileFeedStorage: class S3FeedStorage(BlockingFeedStorage): def __init__(self, uri, access_key=None, secret_key=None, acl=None, endpoint_url=None, *, - feed_options=None): + feed_options=None, session_token=None): if not is_botocore_available(): raise NotConfigured('missing botocore library') u = urlparse(uri) self.bucketname = u.hostname self.access_key = u.username or access_key self.secret_key = u.password or secret_key + self.session_token = session_token self.keyname = u.path[1:] # remove first "/" self.acl = acl self.endpoint_url = endpoint_url @@ -169,6 +170,7 @@ class S3FeedStorage(BlockingFeedStorage): self.s3_client = session.create_client( 's3', aws_access_key_id=self.access_key, aws_secret_access_key=self.secret_key, + aws_session_token=self.session_token, endpoint_url=self.endpoint_url) if feed_options and feed_options.get('overwrite', True) is False: logger.warning('S3 does not support appending to files. To ' @@ -182,6 +184,7 @@ class S3FeedStorage(BlockingFeedStorage): uri, access_key=crawler.settings['AWS_ACCESS_KEY_ID'], secret_key=crawler.settings['AWS_SECRET_ACCESS_KEY'], + session_token=crawler.settings['AWS_SESSION_TOKEN'], acl=crawler.settings['FEED_STORAGE_S3_ACL'] or None, endpoint_url=crawler.settings['AWS_ENDPOINT_URL'] or None, feed_options=feed_options, diff --git a/scrapy/http/response/__init__.py b/scrapy/http/response/__init__.py index 185a9bb67..4de6c9b5b 100644 --- a/scrapy/http/response/__init__.py +++ b/scrapy/http/response/__init__.py @@ -4,7 +4,7 @@ responses in Scrapy. See documentation in docs/topics/request-response.rst """ -from typing import Generator +from typing import Generator, Tuple from urllib.parse import urljoin from scrapy.exceptions import NotSupported @@ -16,6 +16,19 @@ from scrapy.utils.trackref import object_ref class Response(object_ref): + """An object that represents an HTTP response, which is usually + downloaded (by the Downloader) and fed to the Spiders for processing. + """ + + attributes: Tuple[str, ...] = ( + "url", "status", "headers", "body", "flags", "request", "certificate", "ip_address", "protocol", + ) + """A tuple of :class:`str` objects containing the name of all public + attributes of the class that are also keyword parameters of the + ``__init__`` method. + + Currently used by :meth:`Response.replace`. + """ def __init__( self, @@ -97,12 +110,8 @@ class Response(object_ref): return self.replace() def replace(self, *args, **kwargs): - """Create a new Response with the same attributes except for those - given new values. - """ - for x in [ - "url", "status", "headers", "body", "request", "flags", "certificate", "ip_address", "protocol", - ]: + """Create a new Response with the same attributes except for those given new values""" + for x in self.attributes: kwargs.setdefault(x, getattr(self, x)) cls = kwargs.pop('cls', self.__class__) return cls(*args, **kwargs) diff --git a/scrapy/http/response/text.py b/scrapy/http/response/text.py index e36e14880..27bd55c07 100644 --- a/scrapy/http/response/text.py +++ b/scrapy/http/response/text.py @@ -8,7 +8,7 @@ See documentation in docs/topics/request-response.rst import json import warnings from contextlib import suppress -from typing import Generator +from typing import Generator, Tuple from urllib.parse import urljoin import parsel @@ -30,6 +30,8 @@ class TextResponse(Response): _DEFAULT_ENCODING = 'ascii' _cached_decoded_json = _NONE + attributes: Tuple[str, ...] = Response.attributes + ("encoding",) + def __init__(self, *args, **kwargs): self._encoding = kwargs.pop('encoding', None) self._cached_benc = None @@ -53,10 +55,6 @@ class TextResponse(Response): else: super()._set_body(body) - def replace(self, *args, **kwargs): - kwargs.setdefault('encoding', self.encoding) - return Response.replace(self, *args, **kwargs) - @property def encoding(self): return self._declared_encoding() or self._body_inferred_encoding() diff --git a/scrapy/pipelines/files.py b/scrapy/pipelines/files.py index 13ecd4e6c..8766ef66f 100644 --- a/scrapy/pipelines/files.py +++ b/scrapy/pipelines/files.py @@ -79,6 +79,7 @@ class FSFilesStore: class S3FilesStore: AWS_ACCESS_KEY_ID = None AWS_SECRET_ACCESS_KEY = None + AWS_SESSION_TOKEN = None AWS_ENDPOINT_URL = None AWS_REGION_NAME = None AWS_USE_SSL = None @@ -98,6 +99,7 @@ class S3FilesStore: 's3', aws_access_key_id=self.AWS_ACCESS_KEY_ID, aws_secret_access_key=self.AWS_SECRET_ACCESS_KEY, + aws_session_token=self.AWS_SESSION_TOKEN, endpoint_url=self.AWS_ENDPOINT_URL, region_name=self.AWS_REGION_NAME, use_ssl=self.AWS_USE_SSL, @@ -349,6 +351,7 @@ class FilesPipeline(MediaPipeline): s3store = cls.STORE_SCHEMES['s3'] s3store.AWS_ACCESS_KEY_ID = settings['AWS_ACCESS_KEY_ID'] s3store.AWS_SECRET_ACCESS_KEY = settings['AWS_SECRET_ACCESS_KEY'] + s3store.AWS_SESSION_TOKEN = settings['AWS_SESSION_TOKEN'] s3store.AWS_ENDPOINT_URL = settings['AWS_ENDPOINT_URL'] s3store.AWS_REGION_NAME = settings['AWS_REGION_NAME'] s3store.AWS_USE_SSL = settings['AWS_USE_SSL'] diff --git a/scrapy/pipelines/images.py b/scrapy/pipelines/images.py index e3ab23ea5..9c99dc69e 100644 --- a/scrapy/pipelines/images.py +++ b/scrapy/pipelines/images.py @@ -92,6 +92,7 @@ class ImagesPipeline(FilesPipeline): s3store = cls.STORE_SCHEMES['s3'] s3store.AWS_ACCESS_KEY_ID = settings['AWS_ACCESS_KEY_ID'] s3store.AWS_SECRET_ACCESS_KEY = settings['AWS_SECRET_ACCESS_KEY'] + s3store.AWS_SESSION_TOKEN = settings['AWS_SESSION_TOKEN'] s3store.AWS_ENDPOINT_URL = settings['AWS_ENDPOINT_URL'] s3store.AWS_REGION_NAME = settings['AWS_REGION_NAME'] s3store.AWS_USE_SSL = settings['AWS_USE_SSL'] diff --git a/tests/test_feedexport.py b/tests/test_feedexport.py index da0b2c786..389808306 100644 --- a/tests/test_feedexport.py +++ b/tests/test_feedexport.py @@ -244,7 +244,8 @@ class S3FeedStorageTest(unittest.TestCase): def test_parse_credentials(self): skip_if_no_boto() aws_credentials = {'AWS_ACCESS_KEY_ID': 'settings_key', - 'AWS_SECRET_ACCESS_KEY': 'settings_secret'} + 'AWS_SECRET_ACCESS_KEY': 'settings_secret', + 'AWS_SESSION_TOKEN': 'settings_token'} crawler = get_crawler(settings_dict=aws_credentials) # Instantiate with crawler storage = S3FeedStorage.from_crawler( @@ -253,12 +254,15 @@ class S3FeedStorageTest(unittest.TestCase): ) self.assertEqual(storage.access_key, 'settings_key') self.assertEqual(storage.secret_key, 'settings_secret') + self.assertEqual(storage.session_token, 'settings_token') # Instantiate directly storage = S3FeedStorage('s3://mybucket/export.csv', aws_credentials['AWS_ACCESS_KEY_ID'], - aws_credentials['AWS_SECRET_ACCESS_KEY']) + aws_credentials['AWS_SECRET_ACCESS_KEY'], + session_token=aws_credentials['AWS_SESSION_TOKEN']) self.assertEqual(storage.access_key, 'settings_key') self.assertEqual(storage.secret_key, 'settings_secret') + self.assertEqual(storage.session_token, 'settings_token') # URI priority > settings priority storage = S3FeedStorage('s3://uri_key:uri_secret@mybucket/export.csv', aws_credentials['AWS_ACCESS_KEY_ID'], @@ -1957,8 +1961,8 @@ class FileFeedStoragePreFeedOptionsTest(unittest.TestCase): class S3FeedStorageWithoutFeedOptions(S3FeedStorage): - def __init__(self, uri, access_key, secret_key, acl, endpoint_url): - super().__init__(uri, access_key, secret_key, acl, endpoint_url) + def __init__(self, uri, access_key, secret_key, acl, endpoint_url, **kwargs): + super().__init__(uri, access_key, secret_key, acl, endpoint_url, **kwargs) class S3FeedStorageWithoutFeedOptionsWithFromCrawler(S3FeedStorage): diff --git a/tests/test_http_response.py b/tests/test_http_response.py index 04a594d03..cf34a9e5c 100644 --- a/tests/test_http_response.py +++ b/tests/test_http_response.py @@ -820,3 +820,62 @@ class XmlResponseTest(TextResponseTest): response.xpath("//s1:elem/text()", namespaces={'s1': 'http://scrapy.org'}).getall(), response.selector.xpath("//s2:elem/text()").getall(), ) + + +class CustomResponse(TextResponse): + attributes = TextResponse.attributes + ("foo", "bar") + + def __init__(self, *args, **kwargs) -> None: + self.foo = kwargs.pop("foo", None) + self.bar = kwargs.pop("bar", None) + self.lost = kwargs.pop("lost", None) + super().__init__(*args, **kwargs) + + +class CustomResponseTest(TextResponseTest): + response_class = CustomResponse + + def test_copy(self): + super().test_copy() + r1 = self.response_class(url="https://example.org", status=200, foo="foo", bar="bar", lost="lost") + r2 = r1.copy() + self.assertIsInstance(r2, self.response_class) + self.assertEqual(r1.foo, r2.foo) + self.assertEqual(r1.bar, r2.bar) + self.assertEqual(r1.lost, "lost") + self.assertIsNone(r2.lost) + + def test_replace(self): + super().test_replace() + r1 = self.response_class(url="https://example.org", status=200, foo="foo", bar="bar", lost="lost") + + r2 = r1.replace(foo="new-foo", bar="new-bar", lost="new-lost") + self.assertIsInstance(r2, self.response_class) + self.assertEqual(r1.foo, "foo") + self.assertEqual(r1.bar, "bar") + self.assertEqual(r1.lost, "lost") + self.assertEqual(r2.foo, "new-foo") + self.assertEqual(r2.bar, "new-bar") + self.assertEqual(r2.lost, "new-lost") + + r3 = r1.replace(foo="new-foo", bar="new-bar") + self.assertIsInstance(r3, self.response_class) + self.assertEqual(r1.foo, "foo") + self.assertEqual(r1.bar, "bar") + self.assertEqual(r1.lost, "lost") + self.assertEqual(r3.foo, "new-foo") + self.assertEqual(r3.bar, "new-bar") + self.assertIsNone(r3.lost) + + r4 = r1.replace(foo="new-foo") + self.assertIsInstance(r4, self.response_class) + self.assertEqual(r1.foo, "foo") + self.assertEqual(r1.bar, "bar") + self.assertEqual(r1.lost, "lost") + self.assertEqual(r4.foo, "new-foo") + self.assertEqual(r4.bar, "bar") + self.assertIsNone(r4.lost) + + with self.assertRaises(TypeError) as ctx: + r1.replace(unknown="unknown") + self.assertEqual(str(ctx.exception), "__init__() got an unexpected keyword argument 'unknown'") diff --git a/tox.ini b/tox.ini index 4c4bbff6e..96050223b 100644 --- a/tox.ini +++ b/tox.ini @@ -23,6 +23,7 @@ passenv = S3_TEST_FILE_URI AWS_ACCESS_KEY_ID AWS_SECRET_ACCESS_KEY + AWS_SESSION_TOKEN GCS_TEST_FILE_URI GCS_PROJECT_ID #allow tox virtualenv to upgrade pip/wheel/setuptools