Settings

The Scrapy settings allows you to customize the behaviour of all Scrapy components, including the core, extensions, pipelines and spiders themselves.

The infrastructure of the settings provides a global namespace of key-value mappings that the code can use to pull configuration values from. The settings can be populated through different mechanisms, which are described below.

The settings are also the mechanism for selecting the currently active Scrapy project (in case you have many).

For a list of available built-in settings see: Built-in settings reference.

Designating the settings

When you use Scrapy, you have to tell it which settings you’re using. You can do this by using an environment variable, SCRAPY_SETTINGS_MODULE.

The value of SCRAPY_SETTINGS_MODULE should be in Python path syntax, e.g. myproject.settings. Note that the settings module should be on the Python import search path.

Populating the settings

Settings can be populated using different mechanisms, each of which has a different precedence:

1. Command-line settings

Settings set in the command line have the highest precedence, overriding any other settings.

You can explicitly override one or more settings using the -s (or --set) command-line option.

Example:

scrapy crawl myspider -s LOG_LEVEL=INFO -s LOG_FILE=scrapy.log

2. Spider settings

Spiders can define their own settings that will take precedence and override the project ones.

Note

Pre-crawler settings cannot be defined per spider, and reactor settings should not have a different value per spider when running multiple spiders in the same process.

One way to do so is by setting their custom_settings attribute:

import scrapy


class MySpider(scrapy.Spider):
    name = "myspider"

    custom_settings = {
        "SOME_SETTING": "some value",
    }

It’s often better to implement update_settings() instead, and settings set there should use the "spider" priority explicitly:

import scrapy


class MySpider(scrapy.Spider):
    name = "myspider"

    @classmethod
    def update_settings(cls, settings):
        super().update_settings(settings)
        settings.set("SOME_SETTING", "some value", priority="spider")

Added in version 2.11.

It’s also possible to modify the settings in the from_crawler() method, e.g. based on spider arguments or other logic:

import scrapy


class MySpider(scrapy.Spider):
    name = "myspider"

    @classmethod
    def from_crawler(cls, crawler, *args, **kwargs):
        spider = super().from_crawler(crawler, *args, **kwargs)
        if "some_argument" in kwargs:
            spider.settings.set(
                "SOME_SETTING", kwargs["some_argument"], priority="spider"
            )
        return spider

3. Project settings

Scrapy projects include a settings module, usually a file called settings.py, where you should populate most settings that apply to all your spiders.

See also

Designating the settings

4. Add-on settings

Add-ons can modify settings. They should do this with "addon" priority where possible.

5. Command-specific default settings

Each Scrapy command can have its own default settings, which override the global default settings.

Those command-specific default settings are specified in the default_settings attribute of each command class.

6. Default global settings

The scrapy.settings.default_settings module defines global default values for some built-in settings.

Note

startproject generates a settings.py file that sets some settings to different values.

The reference documentation of settings indicates the default value if one exists. If startproject sets a value, that value is documented as default, and the value from scrapy.settings.default_settings is documented as “fallback”.

Compatibility with pickle

Setting values must be picklable.

Import paths and classes

Added in version 2.4.0.

When a setting references a callable object to be imported by Scrapy, such as a class or a function, there are two different ways you can specify that object:

  • As a string containing the import path of that object

  • As the object itself

For example:

from mybot.pipelines.validate import ValidateMyItem

ITEM_PIPELINES = {
    # passing the classname...
    ValidateMyItem: 300,
    # ...equals passing the class path
    "mybot.pipelines.validate.ValidateMyItem": 300,
}

Note

Passing non-callable objects is not supported.

How to access settings

In a spider, settings are available through self.settings:

class MySpider(scrapy.Spider):
    name = "myspider"
    start_urls = ["http://example.com"]

    def parse(self, response):
        print(f"Existing settings: {self.settings.attributes.keys()}")

Note

The settings attribute is set in the base Spider class after the spider is initialized. If you want to use settings before the initialization (e.g., in your spider’s __init__() method), you’ll need to override the from_crawler() method.

Components can also access settings.

The settings object can be used like a dict (e.g. settings["LOG_ENABLED"]). However, to support non-string setting values, which may be passed from the command line as strings, it is recommended to use one of the methods provided by the Settings API.

Component priority dictionaries

A component priority dictionary is a dict where keys are components and values are component priorities. For example:

{
    "path.to.ComponentA": None,
    ComponentB: 100,
}

A component can be specified either as a class object or through an import path.

Warning

Component priority dictionaries are regular dict objects. Be careful not to define the same component more than once, e.g. with different import path strings or defining both an import path and a type object.

A priority can be an int or None.

A component with priority 1 goes before a component with priority 2. What going before entails, however, depends on the corresponding setting. For example, in the DOWNLOADER_MIDDLEWARES setting, components have their process_request() method executed before that of later components, but have their process_response() method executed after that of later components.

A component with priority None is disabled.

Some component priority dictionaries get merged with some built-in value. For example, DOWNLOADER_MIDDLEWARES is merged with DOWNLOADER_MIDDLEWARES_BASE. This is where None comes in handy, allowing you to disable a component from the base setting in the regular setting:

DOWNLOADER_MIDDLEWARES = {
    "scrapy.downloadermiddlewares.offsite.OffsiteMiddleware": None,
}

Special settings

The following settings work slightly differently than all other settings.

Pre-crawler settings

Pre-crawler settings are settings used before the Crawler object is created.

These settings cannot be set from a spider.

These settings are SPIDER_LOADER_CLASS and settings used by the corresponding component, e.g. SPIDER_MODULES and SPIDER_LOADER_WARN_ONLY for the default component.

Reactor settings

Reactor settings are settings tied to the Twisted reactor.

These settings can be defined from a spider. However, because only 1 reactor can be used per process, these settings cannot use a different value per spider when running multiple spiders in the same process.

In general, if different spiders define different values, the first defined value is used. However, if two spiders request a different reactor, an exception is raised.

These settings are:

ASYNCIO_EVENT_LOOP and TWISTED_REACTOR are used upon installing the reactor. The rest of the settings are applied when starting the reactor.

Built-in settings reference

Here’s a list of all available Scrapy settings, in alphabetical order, along with their default values and the scope where they apply.

The scope, where available, shows where the setting is being used, if it’s tied to any particular component. In that case the module of that component will be shown, typically an extension, middleware or pipeline. It also means that the component must be enabled in order for the setting to have any effect.

ADDONS

Default: {}

A dict containing paths to the add-ons enabled in your project and their priorities. For more information, see Add-ons.

AWS_ACCESS_KEY_ID

Default: None

The AWS access key used by code that requires access to Amazon Web services, such as the S3 feed storage backend.

AWS_SECRET_ACCESS_KEY

Default: None

The AWS secret key used by code that requires access to Amazon Web services, such as the S3 feed storage backend.

AWS_SESSION_TOKEN

Default: None

The AWS security token used by code that requires access to Amazon Web services, such as the S3 feed storage backend, when using temporary security credentials.

AWS_ENDPOINT_URL

Default: None

Endpoint URL used for S3-like storage, for example Minio or s3.scality.

AWS_USE_SSL

Default: None

Use this option if you want to disable SSL connection for communication with S3 or S3-like storage. By default SSL will be used.

AWS_VERIFY

Default: None

Verify SSL connection between Scrapy and S3 or S3-like storage. By default SSL verification will occur.

AWS_REGION_NAME

Default: None

The name of the region associated with the AWS client.

ASYNCIO_EVENT_LOOP

Default: None

Import path of a given asyncio event loop class.

If the asyncio reactor is enabled (see TWISTED_REACTOR) this setting can be used to specify the asyncio event loop to be used with it. Set the setting to the import path of the desired asyncio event loop class. If the setting is set to None the default asyncio event loop will be used.

If you are installing the asyncio reactor manually using the install_reactor() function, you can use the event_loop_path parameter to indicate the import path of the event loop class to be used.

Note that the event loop class must inherit from asyncio.AbstractEventLoop.

Caution

Please be aware that, when using a non-default event loop (either defined via ASYNCIO_EVENT_LOOP or installed with install_reactor()), Scrapy will call asyncio.set_event_loop(), which will set the specified event loop as the current loop for the current OS thread.

BOT_NAME

Default: <project name> (fallback: 'scrapybot')

The name of the bot implemented by this Scrapy project (also known as the project name). This name will be used for the logging too.

It’s automatically populated with your project name when you create your project with the startproject command.

CONCURRENT_ITEMS

Default: 100

Maximum number of concurrent items (per response) to process in parallel in item pipelines.

CONCURRENT_REQUESTS

Default: 16

The maximum number of concurrent (i.e. simultaneous) requests that will be performed by the Scrapy downloader.

CONCURRENT_REQUESTS_PER_DOMAIN

Default: 8

The maximum number of concurrent (i.e. simultaneous) requests that will be performed to any single domain.

See also: AutoThrottle extension and its AUTOTHROTTLE_TARGET_CONCURRENCY option.

CONCURRENT_REQUESTS_PER_IP

Default: 0

The maximum number of concurrent (i.e. simultaneous) requests that will be performed to any single IP. If non-zero, the CONCURRENT_REQUESTS_PER_DOMAIN setting is ignored, and this one is used instead. In other words, concurrency limits will be applied per IP, not per domain.

This setting also affects DOWNLOAD_DELAY and AutoThrottle extension: if CONCURRENT_REQUESTS_PER_IP is non-zero, download delay is enforced per IP, not per domain.

DEFAULT_DROPITEM_LOG_LEVEL

Default: "WARNING"

Default log level of messages about dropped items.

When an item is dropped by raising scrapy.exceptions.DropItem from the process_item() method of an item pipeline, a message is logged, and by default its log level is the one configured in this setting.

You may specify this log level as an integer (e.g. 20), as a log level constant (e.g. logging.INFO) or as a string with the name of a log level constant (e.g. "INFO").

When writing an item pipeline, you can force a different log level by setting scrapy.exceptions.DropItem.log_level in your scrapy.exceptions.DropItem exception. For example:

from scrapy.exceptions import DropItem


class MyPipeline:
    def process_item(self, item, spider):
        if not item.get("price"):
            raise DropItem("Missing price data", log_level="INFO")
        return item

DEFAULT_ITEM_CLASS

Default: 'scrapy.Item'

The default class that will be used for instantiating items in the the Scrapy shell.

DEFAULT_REQUEST_HEADERS

Default:

{
    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
    "Accept-Language": "en",
}

The default headers used for Scrapy HTTP Requests. They’re populated in the DefaultHeadersMiddleware.

Caution

Cookies set via the Cookie header are not considered by the CookiesMiddleware. If you need to set cookies for a request, use the Request.cookies parameter. This is a known current limitation that is being worked on.

DEPTH_LIMIT

Default: 0

Scope: scrapy.spidermiddlewares.depth.DepthMiddleware

The maximum depth that will be allowed to crawl for any site. If zero, no limit will be imposed.

DEPTH_PRIORITY

Default: 0

Scope: scrapy.spidermiddlewares.depth.DepthMiddleware

An integer that is used to adjust the priority of a Request based on its depth.

The priority of a request is adjusted as follows:

request.priority = request.priority - (depth * DEPTH_PRIORITY)

As depth increases, positive values of DEPTH_PRIORITY decrease request priority (BFO), while negative values increase request priority (DFO). See also Does Scrapy crawl in breadth-first or depth-first order?.

Note

This setting adjusts priority in the opposite way compared to other priority settings REDIRECT_PRIORITY_ADJUST and RETRY_PRIORITY_ADJUST.

DEPTH_STATS_VERBOSE

Default: False

Scope: scrapy.spidermiddlewares.depth.DepthMiddleware

Whether to collect verbose depth stats. If this is enabled, the number of requests for each depth is collected in the stats.

DNSCACHE_ENABLED

Default: True

Whether to enable DNS in-memory cache.

DNSCACHE_SIZE

Default: 10000

DNS in-memory cache size.

DNS_RESOLVER

Added in version 2.0.

Default: 'scrapy.resolver.CachingThreadedResolver'

The class to be used to resolve DNS names. The default scrapy.resolver.CachingThreadedResolver supports specifying a timeout for DNS requests via the DNS_TIMEOUT setting, but works only with IPv4 addresses. Scrapy provides an alternative resolver, scrapy.resolver.CachingHostnameResolver, which supports IPv4/IPv6 addresses but does not take the DNS_TIMEOUT setting into account.

DNS_TIMEOUT

Default: 60

Timeout for processing of DNS queries in seconds. Float is supported.

DOWNLOADER

Default: 'scrapy.core.downloader.Downloader'

The downloader to use for crawling.

DOWNLOADER_HTTPCLIENTFACTORY

Default: 'scrapy.core.downloader.webclient.ScrapyHTTPClientFactory'

Defines a Twisted protocol.ClientFactory class to use for HTTP/1.0 connections (for HTTP10DownloadHandler).

Note

HTTP/1.0 is rarely used nowadays and its Scrapy support is deprecated, so you can safely ignore this setting, unless you really want to use HTTP/1.0 and override DOWNLOAD_HANDLERS for http(s) scheme accordingly, i.e. to 'scrapy.core.downloader.handlers.http.HTTP10DownloadHandler'.

DOWNLOADER_CLIENTCONTEXTFACTORY

Default: 'scrapy.core.downloader.contextfactory.ScrapyClientContextFactory'

Represents the classpath to the ContextFactory to use.

Here, “ContextFactory” is a Twisted term for SSL/TLS contexts, defining the TLS/SSL protocol version to use, whether to do certificate verification, or even enable client-side authentication (and various other things).

Note

Scrapy default context factory does NOT perform remote server certificate verification. This is usually fine for web scraping.

If you do need remote server certificate verification enabled, Scrapy also has another context factory class that you can set, 'scrapy.core.downloader.contextfactory.BrowserLikeContextFactory', which uses the platform’s certificates to validate remote endpoints.

If you do use a custom ContextFactory, make sure its __init__ method accepts a method parameter (this is the OpenSSL.SSL method mapping DOWNLOADER_CLIENT_TLS_METHOD), a tls_verbose_logging parameter (bool) and a tls_ciphers parameter (see DOWNLOADER_CLIENT_TLS_CIPHERS).

DOWNLOADER_CLIENT_TLS_CIPHERS

Default: 'DEFAULT'

Use this setting to customize the TLS/SSL ciphers used by the default HTTP/1.1 downloader.

The setting should contain a string in the OpenSSL cipher list format, these ciphers will be used as client ciphers. Changing this setting may be necessary to access certain HTTPS websites: for example, you may need to use 'DEFAULT:!DH' for a website with weak DH parameters or enable a specific cipher that is not included in DEFAULT if a website requires it.

DOWNLOADER_CLIENT_TLS_METHOD

Default: 'TLS'

Use this setting to customize the TLS/SSL method used by the default HTTP/1.1 downloader.

This setting must be one of these string values:

  • 'TLS': maps to OpenSSL’s TLS_method() (a.k.a SSLv23_method()), which allows protocol negotiation, starting from the highest supported by the platform; default, recommended

  • 'TLSv1.0': this value forces HTTPS connections to use TLS version 1.0 ; set this if you want the behavior of Scrapy<1.1

  • 'TLSv1.1': forces TLS version 1.1

  • 'TLSv1.2': forces TLS version 1.2

DOWNLOADER_CLIENT_TLS_VERBOSE_LOGGING

Default: False

Setting this to True will enable DEBUG level messages about TLS connection parameters after establishing HTTPS connections. The kind of information logged depends on the versions of OpenSSL and pyOpenSSL.

This setting is only used for the default DOWNLOADER_CLIENTCONTEXTFACTORY.

DOWNLOADER_MIDDLEWARES

Default:: {}

A dict containing the downloader middlewares enabled in your project, and their orders. For more info see Activating a downloader middleware.

DOWNLOADER_MIDDLEWARES_BASE

Default:

{
    "scrapy.downloadermiddlewares.offsite.OffsiteMiddleware": 50,
    "scrapy.downloadermiddlewares.robotstxt.RobotsTxtMiddleware": 100,
    "scrapy.downloadermiddlewares.httpauth.HttpAuthMiddleware": 300,
    "scrapy.downloadermiddlewares.downloadtimeout.DownloadTimeoutMiddleware": 350,
    "scrapy.downloadermiddlewares.defaultheaders.DefaultHeadersMiddleware": 400,
    "scrapy.downloadermiddlewares.useragent.UserAgentMiddleware": 500,
    "scrapy.downloadermiddlewares.retry.RetryMiddleware": 550,
    "scrapy.downloadermiddlewares.ajaxcrawl.AjaxCrawlMiddleware": 560,
    "scrapy.downloadermiddlewares.redirect.MetaRefreshMiddleware": 580,
    "scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware": 590,
    "scrapy.downloadermiddlewares.redirect.RedirectMiddleware": 600,
    "scrapy.downloadermiddlewares.cookies.CookiesMiddleware": 700,
    "scrapy.downloadermiddlewares.httpproxy.HttpProxyMiddleware": 750,
    "scrapy.downloadermiddlewares.stats.DownloaderStats": 850,
    "scrapy.downloadermiddlewares.httpcache.HttpCacheMiddleware": 900,
}

A dict containing the downloader middlewares enabled by default in Scrapy. Low orders are closer to the engine, high orders are closer to the downloader. You should never modify this setting in your project, modify DOWNLOADER_MIDDLEWARES instead. For more info see Activating a downloader middleware.

DOWNLOADER_STATS

Default: True

Whether to enable downloader stats collection.

DOWNLOAD_DELAY

Default: 0

Minimum seconds to wait between 2 consecutive requests to the same domain.

Use DOWNLOAD_DELAY to throttle your crawling speed, to avoid hitting servers too hard.

Decimal numbers are supported. For example, to send a maximum of 4 requests every 10 seconds:

DOWNLOAD_DELAY = 2.5

This setting is also affected by the RANDOMIZE_DOWNLOAD_DELAY setting, which is enabled by default.

When CONCURRENT_REQUESTS_PER_IP is non-zero, delays are enforced per IP address instead of per domain.

Note that DOWNLOAD_DELAY can lower the effective per-domain concurrency below CONCURRENT_REQUESTS_PER_DOMAIN. If the response time of a domain is lower than DOWNLOAD_DELAY, the effective concurrency for that domain is 1. When testing throttling configurations, it usually makes sense to lower CONCURRENT_REQUESTS_PER_DOMAIN first, and only increase DOWNLOAD_DELAY once CONCURRENT_REQUESTS_PER_DOMAIN is 1 but a higher throttling is desired.

Note

This delay can be set per spider using download_delay spider attribute.

It is also possible to change this setting per domain, although it requires non-trivial code. See the implementation of the AutoThrottle extension for an example.

DOWNLOAD_HANDLERS

Default: {}

A dict containing the request downloader handlers enabled in your project. See DOWNLOAD_HANDLERS_BASE for example format.

DOWNLOAD_HANDLERS_BASE

Default:

{
    "data": "scrapy.core.downloader.handlers.datauri.DataURIDownloadHandler",
    "file": "scrapy.core.downloader.handlers.file.FileDownloadHandler",
    "http": "scrapy.core.downloader.handlers.http.HTTPDownloadHandler",
    "https": "scrapy.core.downloader.handlers.http.HTTPDownloadHandler",
    "s3": "scrapy.core.downloader.handlers.s3.S3DownloadHandler",
    "ftp": "scrapy.core.downloader.handlers.ftp.FTPDownloadHandler",
}

A dict containing the request download handlers enabled by default in Scrapy. You should never modify this setting in your project, modify DOWNLOAD_HANDLERS instead.

You can disable any of these download handlers by assigning None to their URI scheme in DOWNLOAD_HANDLERS. E.g., to disable the built-in FTP handler (without replacement), place this in your settings.py:

DOWNLOAD_HANDLERS = {
    "ftp": None,
}

The default HTTPS handler uses HTTP/1.1. To use HTTP/2:

  1. Install Twisted[http2]>=17.9.0 to install the packages required to enable HTTP/2 support in Twisted.

  2. Update DOWNLOAD_HANDLERS as follows:

    DOWNLOAD_HANDLERS = {
    "https": "scrapy.core.downloader.handlers.http2.H2DownloadHandler",
}

Warning

HTTP/2 support in Scrapy is experimental, and not yet recommended for production environments. Future Scrapy versions may introduce related changes without a deprecation period or warning.

Note

Known limitations of the current HTTP/2 implementation of Scrapy include:

  • No support for HTTP/2 Cleartext (h2c), since no major browser supports HTTP/2 unencrypted (refer http2 faq).

  • No setting to specify a maximum frame size larger than the default value, 16384. Connections to servers that send a larger frame will fail.

  • No support for server pushes, which are ignored.

  • No support for the bytes_received and headers_received signals.

DOWNLOAD_SLOTS

Default: {}

Allows to define concurrency/delay parameters on per slot (domain) basis:

DOWNLOAD_SLOTS = {
    "quotes.toscrape.com": {"concurrency": 1, "delay": 2, "randomize_delay": False},
    "books.toscrape.com": {"delay": 3, "randomize_delay": False},
}

Note

For other downloader slots default settings values will be used:

DOWNLOAD_TIMEOUT

Default: 180

The amount of time (in secs) that the downloader will wait before timing out.

Note

This timeout can be set per spider using download_timeout spider attribute and per-request using download_timeout Request.meta key.

DOWNLOAD_MAXSIZE

Default: 1073741824 (1 GiB)

The maximum response body size (in bytes) allowed. Bigger responses are aborted and ignored.

This applies both before and after compression. If decompressing a response body would exceed this limit, decompression is aborted and the response is ignored.

Use 0 to disable this limit.

This limit can be set per spider using the download_maxsize spider attribute and per request using the download_maxsize Request.meta key.

DOWNLOAD_WARNSIZE

Default: 33554432 (32 MiB)

If the size of a response exceeds this value, before or after compression, a warning will be logged about it.

Use 0 to disable this limit.

This limit can be set per spider using the download_warnsize spider attribute and per request using the download_warnsize Request.meta key.

DOWNLOAD_FAIL_ON_DATALOSS

Default: True

Whether or not to fail on broken responses, that is, declared Content-Length does not match content sent by the server or chunked response was not properly finish. If True, these responses raise a ResponseFailed([_DataLoss]) error. If False, these responses are passed through and the flag dataloss is added to the response, i.e.: 'dataloss' in response.flags is True.

Optionally, this can be set per-request basis by using the download_fail_on_dataloss Request.meta key to False.

Note

A broken response, or data loss error, may happen under several circumstances, from server misconfiguration to network errors to data corruption. It is up to the user to decide if it makes sense to process broken responses considering they may contain partial or incomplete content. If RETRY_ENABLED is True and this setting is set to True, the ResponseFailed([_DataLoss]) failure will be retried as usual.

Warning

This setting is ignored by the H2DownloadHandler download handler (see DOWNLOAD_HANDLERS). In case of a data loss error, the corresponding HTTP/2 connection may be corrupted, affecting other requests that use the same connection; hence, a ResponseFailed([InvalidBodyLengthError]) failure is always raised for every request that was using that connection.

DUPEFILTER_CLASS

Default: 'scrapy.dupefilters.RFPDupeFilter'

The class used to detect and filter duplicate requests.

The default, RFPDupeFilter, filters based on the REQUEST_FINGERPRINTER_CLASS setting.

To change how duplicates are checked, you can point DUPEFILTER_CLASS to a custom subclass of RFPDupeFilter that overrides its __init__ method to use a different request fingerprinting class. For example:

from scrapy.dupefilters import RFPDupeFilter
from scrapy.utils.request import fingerprint


class CustomRequestFingerprinter:
    def fingerprint(self, request):
        return fingerprint(request, include_headers=["X-ID"])


class CustomDupeFilter(RFPDupeFilter):

    def __init__(self, path=None, debug=False, *, fingerprinter=None):
        super().__init__(
            path=path, debug=debug, fingerprinter=CustomRequestFingerprinter()
        )

To disable duplicate request filtering set DUPEFILTER_CLASS to 'scrapy.dupefilters.BaseDupeFilter'. Note that not filtering out duplicate requests may cause crawling loops. It is usually better to set the dont_filter parameter to True on the __init__ method of a specific Request object that should not be filtered out.

A class assigned to DUPEFILTER_CLASS must implement the following interface:

class MyDupeFilter:

    @classmethod
    def from_settings(cls, settings):
        """Returns an instance of this duplicate request filtering class
        based on the current crawl settings."""
        return cls()

    def request_seen(self, request):
        """Returns ``True`` if *request* is a duplicate of another request
        seen in a previous call to :meth:`request_seen`, or ``False``
        otherwise."""
        return False

    def open(self):
        """Called before the spider opens. It may return a deferred."""
        pass

    def close(self, reason):
        """Called before the spider closes. It may return a deferred."""
        pass

    def log(self, request, spider):
        """Logs that a request has been filtered out.

        It is called right after a call to :meth:`request_seen` that
        returns ``True``.

        If :meth:`request_seen` always returns ``False``, such as in the
        case of :class:`~scrapy.dupefilters.BaseDupeFilter`, this method
        may be omitted.
        """
        pass
class scrapy.dupefilters.BaseDupeFilter[source]

Dummy duplicate request filtering class (DUPEFILTER_CLASS) that does not filter out any request.

class scrapy.dupefilters.RFPDupeFilter(path: str | None = None, debug: bool = False, *, fingerprinter: RequestFingerprinterProtocol | None = None)[source]

Duplicate request filtering class (DUPEFILTER_CLASS) that filters out requests with the canonical (w3lib.url.canonicalize_url()) url, method and body.

DUPEFILTER_DEBUG

Default: False

By default, RFPDupeFilter only logs the first duplicate request. Setting DUPEFILTER_DEBUG to True will make it log all duplicate requests.

EDITOR

Default: vi (on Unix systems) or the IDLE editor (on Windows)

The editor to use for editing spiders with the edit command. Additionally, if the EDITOR environment variable is set, the edit command will prefer it over the default setting.

EXTENSIONS

Default:: {}

Component priority dictionary of enabled extensions. See Extensions.

EXTENSIONS_BASE

Default:

{
    "scrapy.extensions.corestats.CoreStats": 0,
    "scrapy.extensions.telnet.TelnetConsole": 0,
    "scrapy.extensions.memusage.MemoryUsage": 0,
    "scrapy.extensions.memdebug.MemoryDebugger": 0,
    "scrapy.extensions.closespider.CloseSpider": 0,
    "scrapy.extensions.feedexport.FeedExporter": 0,
    "scrapy.extensions.logstats.LogStats": 0,
    "scrapy.extensions.spiderstate.SpiderState": 0,
    "scrapy.extensions.throttle.AutoThrottle": 0,
}

A dict containing the extensions available by default in Scrapy, and their orders. This setting contains all stable built-in extensions. Keep in mind that some of them need to be enabled through a setting.

For more information See the extensions user guide and the list of available extensions.

FEED_TEMPDIR

The Feed Temp dir allows you to set a custom folder to save crawler temporary files before uploading with FTP feed storage and Amazon S3.

FEED_STORAGE_GCS_ACL

The Access Control List (ACL) used when storing items to Google Cloud Storage. For more information on how to set this value, please refer to the column JSON API in Google Cloud documentation.

FTP_PASSIVE_MODE

Default: True

Whether or not to use passive mode when initiating FTP transfers.

FTP_PASSWORD

Default: "guest"

The password to use for FTP connections when there is no "ftp_password" in Request meta.

Note

Paraphrasing RFC 1635, although it is common to use either the password “guest” or one’s e-mail address for anonymous FTP, some FTP servers explicitly ask for the user’s e-mail address and will not allow login with the “guest” password.

FTP_USER

Default: "anonymous"

The username to use for FTP connections when there is no "ftp_user" in Request meta.

GCS_PROJECT_ID

Default: None

The Project ID that will be used when storing data on Google Cloud Storage.

ITEM_PIPELINES

Default: {}

A dict containing the item pipelines to use, and their orders. Order values are arbitrary, but it is customary to define them in the 0-1000 range. Lower orders process before higher orders.

Example:

ITEM_PIPELINES = {
    "mybot.pipelines.validate.ValidateMyItem": 300,
    "mybot.pipelines.validate.StoreMyItem": 800,
}

ITEM_PIPELINES_BASE

Default: {}

A dict containing the pipelines enabled by default in Scrapy. You should never modify this setting in your project, modify ITEM_PIPELINES instead.

JOBDIR

Default: None

A string indicating the directory for storing the state of a crawl when pausing and resuming crawls.

LOG_ENABLED

Default: True

Whether to enable logging.

LOG_ENCODING

Default: 'utf-8'

The encoding to use for logging.

LOG_FILE

Default: None

File name to use for logging output. If None, standard error will be used.

LOG_FILE_APPEND

Default: True

If False, the log file specified with LOG_FILE will be overwritten (discarding the output from previous runs, if any).

LOG_FORMAT

Default: '%(asctime)s [%(name)s] %(levelname)s: %(message)s'

String for formatting log messages. Refer to the Python logging documentation for the whole list of available placeholders.

LOG_DATEFORMAT

Default: '%Y-%m-%d %H:%M:%S'

String for formatting date/time, expansion of the %(asctime)s placeholder in LOG_FORMAT. Refer to the Python datetime documentation for the whole list of available directives.

LOG_FORMATTER

Default: scrapy.logformatter.LogFormatter

The class to use for formatting log messages for different actions.

LOG_LEVEL

Default: 'DEBUG'

Minimum level to log. Available levels are: CRITICAL, ERROR, WARNING, INFO, DEBUG. For more info see Logging.

LOG_STDOUT

Default: False

If True, all standard output (and error) of your process will be redirected to the log. For example if you print('hello') it will appear in the Scrapy log.

LOG_SHORT_NAMES

Default: False

If True, the logs will just contain the root path. If it is set to False then it displays the component responsible for the log output

LOG_VERSIONS

Default: ["lxml", "libxml2", "cssselect", "parsel", "w3lib", "Twisted", "Python", "pyOpenSSL", "cryptography", "Platform"]

Logs the installed versions of the specified items.

An item can be any installed Python package.

The following special items are also supported:

LOGSTATS_INTERVAL

Default: 60.0

The interval (in seconds) between each logging printout of the stats by LogStats.

MEMDEBUG_ENABLED

Default: False

Whether to enable memory debugging.

MEMDEBUG_NOTIFY

Default: []

When memory debugging is enabled a memory report will be sent to the specified addresses if this setting is not empty, otherwise the report will be written to the log.

Example:

MEMDEBUG_NOTIFY = ['user@example.com']

MEMUSAGE_ENABLED

Default: True

Scope: scrapy.extensions.memusage

Whether to enable the memory usage extension. This extension keeps track of a peak memory used by the process (it writes it to stats). It can also optionally shutdown the Scrapy process when it exceeds a memory limit (see MEMUSAGE_LIMIT_MB), and notify by email when that happened (see MEMUSAGE_NOTIFY_MAIL).

See Memory usage extension.

MEMUSAGE_LIMIT_MB

Default: 0

Scope: scrapy.extensions.memusage

The maximum amount of memory to allow (in megabytes) before shutting down Scrapy (if MEMUSAGE_ENABLED is True). If zero, no check will be performed.

See Memory usage extension.

MEMUSAGE_CHECK_INTERVAL_SECONDS

Default: 60.0

Scope: scrapy.extensions.memusage

The Memory usage extension checks the current memory usage, versus the limits set by MEMUSAGE_LIMIT_MB and MEMUSAGE_WARNING_MB, at fixed time intervals.

This sets the length of these intervals, in seconds.

See Memory usage extension.

MEMUSAGE_NOTIFY_MAIL

Default: False

Scope: scrapy.extensions.memusage

A list of emails to notify if the memory limit has been reached.

Example:

MEMUSAGE_NOTIFY_MAIL = ['user@example.com']

See Memory usage extension.

MEMUSAGE_WARNING_MB

Default: 0

Scope: scrapy.extensions.memusage

The maximum amount of memory to allow (in megabytes) before sending a warning email notifying about it. If zero, no warning will be produced.

NEWSPIDER_MODULE

Default: "<project name>.spiders" (fallback: "")

Module where to create new spiders using the genspider command.

Example:

NEWSPIDER_MODULE = 'mybot.spiders_dev'

RANDOMIZE_DOWNLOAD_DELAY

Default: True

If enabled, Scrapy will wait a random amount of time (between 0.5 * DOWNLOAD_DELAY and 1.5 * DOWNLOAD_DELAY) while fetching requests from the same website.

This randomization decreases the chance of the crawler being detected (and subsequently blocked) by sites which analyze requests looking for statistically significant similarities in the time between their requests.

The randomization policy is the same used by wget --random-wait option.

If DOWNLOAD_DELAY is zero (default) this option has no effect.

REACTOR_THREADPOOL_MAXSIZE

Default: 10

The maximum limit for Twisted Reactor thread pool size. This is common multi-purpose thread pool used by various Scrapy components. Threaded DNS Resolver, BlockingFeedStorage, S3FilesStore just to name a few. Increase this value if you’re experiencing problems with insufficient blocking IO.

REDIRECT_PRIORITY_ADJUST

Default: +2

Scope: scrapy.downloadermiddlewares.redirect.RedirectMiddleware

Adjust redirect request priority relative to original request:

  • a positive priority adjust (default) means higher priority.

  • a negative priority adjust means lower priority.

ROBOTSTXT_OBEY

Default: True (fallback: False)

If enabled, Scrapy will respect robots.txt policies. For more information see RobotsTxtMiddleware.

Note

While the default value is False for historical reasons, this option is enabled by default in settings.py file generated by scrapy startproject command.

ROBOTSTXT_PARSER

Default: 'scrapy.robotstxt.ProtegoRobotParser'

The parser backend to use for parsing robots.txt files. For more information see RobotsTxtMiddleware.

ROBOTSTXT_USER_AGENT

Default: None

The user agent string to use for matching in the robots.txt file. If None, the User-Agent header you are sending with the request or the USER_AGENT setting (in that order) will be used for determining the user agent to use in the robots.txt file.

SCHEDULER

Default: 'scrapy.core.scheduler.Scheduler'

The scheduler class to be used for crawling. See the Scheduler topic for details.

SCHEDULER_DEBUG

Default: False

Setting to True will log debug information about the requests scheduler. This currently logs (only once) if the requests cannot be serialized to disk. Stats counter (scheduler/unserializable) tracks the number of times this happens.

Example entry in logs:

1956-01-31 00:00:00+0800 [scrapy.core.scheduler] ERROR: Unable to serialize request:
<GET http://example.com> - reason: cannot serialize <Request at 0x9a7c7ec>
(type Request)> - no more unserializable requests will be logged
(see 'scheduler/unserializable' stats counter)

SCHEDULER_DISK_QUEUE

Default: 'scrapy.squeues.PickleLifoDiskQueue'

Type of disk queue that will be used by the scheduler. Other available types are scrapy.squeues.PickleFifoDiskQueue, scrapy.squeues.MarshalFifoDiskQueue, scrapy.squeues.MarshalLifoDiskQueue.

SCHEDULER_MEMORY_QUEUE

Default: 'scrapy.squeues.LifoMemoryQueue'

Type of in-memory queue used by the scheduler. Other available type is: scrapy.squeues.FifoMemoryQueue.

SCHEDULER_PRIORITY_QUEUE

Default: 'scrapy.pqueues.ScrapyPriorityQueue'

Type of priority queue used by the scheduler. Another available type is scrapy.pqueues.DownloaderAwarePriorityQueue. scrapy.pqueues.DownloaderAwarePriorityQueue works better than scrapy.pqueues.ScrapyPriorityQueue when you crawl many different domains in parallel. But currently scrapy.pqueues.DownloaderAwarePriorityQueue does not work together with CONCURRENT_REQUESTS_PER_IP.

SCHEDULER_START_DISK_QUEUE

Default: 'scrapy.squeues.PickleFifoDiskQueue'

Type of disk queue (see JOBDIR) that the scheduler uses for start requests.

For available choices, see SCHEDULER_DISK_QUEUE.

Use None or "" to disable these separate queues entirely, and instead have start requests share the same queues as other requests.

Note

Disabling separate start request queues makes start request order unintuitive: start requests will be sent in order only until CONCURRENT_REQUESTS is reached, then remaining start requests will be sent in reverse order.

SCHEDULER_START_MEMORY_QUEUE

Default: 'scrapy.squeues.FifoMemoryQueue'

Type of in-memory queue that the scheduler uses for start requests.

For available choices, see SCHEDULER_MEMORY_QUEUE.

Use None or "" to disable these separate queues entirely, and instead have start requests share the same queues as other requests.

Note

Disabling separate start request queues makes start request order unintuitive: start requests will be sent in order only until CONCURRENT_REQUESTS is reached, then remaining start requests will be sent in reverse order.

SCRAPER_SLOT_MAX_ACTIVE_SIZE

Added in version 2.0.

Default: 5_000_000

Soft limit (in bytes) for response data being processed.

While the sum of the sizes of all responses being processed is above this value, Scrapy does not process new requests.

SPIDER_CONTRACTS

Default:: {}

A dict containing the spider contracts enabled in your project, used for testing spiders. For more info see Spiders Contracts.

SPIDER_CONTRACTS_BASE

Default:

{
    "scrapy.contracts.default.UrlContract": 1,
    "scrapy.contracts.default.ReturnsContract": 2,
    "scrapy.contracts.default.ScrapesContract": 3,
}

A dict containing the Scrapy contracts enabled by default in Scrapy. You should never modify this setting in your project, modify SPIDER_CONTRACTS instead. For more info see Spiders Contracts.

You can disable any of these contracts by assigning None to their class path in SPIDER_CONTRACTS. E.g., to disable the built-in ScrapesContract, place this in your settings.py:

SPIDER_CONTRACTS = {
    "scrapy.contracts.default.ScrapesContract": None,
}

SPIDER_LOADER_CLASS

Default: 'scrapy.spiderloader.SpiderLoader'

The class that will be used for loading spiders, which must implement the SpiderLoader API.

SPIDER_LOADER_WARN_ONLY

Default: False

By default, when Scrapy tries to import spider classes from SPIDER_MODULES, it will fail loudly if there is any ImportError or SyntaxError exception. But you can choose to silence this exception and turn it into a simple warning by setting SPIDER_LOADER_WARN_ONLY = True.

Note

Some scrapy commands run with this setting to True already (i.e. they will only issue a warning and will not fail) since they do not actually need to load spider classes to work: scrapy runspider, scrapy settings, scrapy startproject, scrapy version.

SPIDER_MIDDLEWARES

Default:: {}

A dict containing the spider middlewares enabled in your project, and their orders. For more info see Activating a spider middleware.

SPIDER_MIDDLEWARES_BASE

Default:

{
    "scrapy.spidermiddlewares.httperror.HttpErrorMiddleware": 50,
    "scrapy.spidermiddlewares.referer.RefererMiddleware": 700,
    "scrapy.spidermiddlewares.urllength.UrlLengthMiddleware": 800,
    "scrapy.spidermiddlewares.depth.DepthMiddleware": 900,
}

A dict containing the spider middlewares enabled by default in Scrapy, and their orders. Low orders are closer to the engine, high orders are closer to the spider. For more info see Activating a spider middleware.

SPIDER_MODULES

Default: ["<project name>.spiders"] (fallback: [])

A list of modules where Scrapy will look for spiders.

Example:

SPIDER_MODULES = ["mybot.spiders_prod", "mybot.spiders_dev"]

STATS_CLASS

Default: 'scrapy.statscollectors.MemoryStatsCollector'

The class to use for collecting stats, who must implement the Stats Collector API.

STATS_DUMP

Default: True

Dump the Scrapy stats (to the Scrapy log) once the spider finishes.

For more info see: Stats Collection.

STATSMAILER_RCPTS

Default: [] (empty list)

Send Scrapy stats after spiders finish scraping. See StatsMailer for more info.

TELNETCONSOLE_ENABLED

Default: True

A boolean which specifies if the telnet console will be enabled (provided its extension is also enabled).

TEMPLATES_DIR

Default: templates dir inside scrapy module

The directory where to look for templates when creating new projects with startproject command and new spiders with genspider command.

The project name must not conflict with the name of custom files or directories in the project subdirectory.

TWISTED_REACTOR

Added in version 2.0.

Default: "twisted.internet.asyncioreactor.AsyncioSelectorReactor"

Import path of a given reactor.

Scrapy will install this reactor if no other reactor is installed yet, such as when the scrapy CLI program is invoked or when using the CrawlerProcess class.

If you are using the CrawlerRunner class, you also need to install the correct reactor manually. You can do that using install_reactor():

scrapy.utils.reactor.install_reactor(reactor_path: str, event_loop_path: str | None = None) → None[source]

Installs the reactor with the specified import path. Also installs the asyncio event loop with the specified import path if the asyncio reactor is enabled

If a reactor is already installed, install_reactor() has no effect.

CrawlerRunner.__init__ raises Exception if the installed reactor does not match the TWISTED_REACTOR setting; therefore, having top-level reactor imports in project files and imported third-party libraries will make Scrapy raise Exception when it checks which reactor is installed.

In order to use the reactor installed by Scrapy:

import scrapy
from twisted.internet import reactor


class QuotesSpider(scrapy.Spider):
    name = "quotes"

    def __init__(self, *args, **kwargs):
        self.timeout = int(kwargs.pop("timeout", "60"))
        super(QuotesSpider, self).__init__(*args, **kwargs)

    async def start(self):
        reactor.callLater(self.timeout, self.stop)

        urls = ["https://quotes.toscrape.com/page/1"]
        for url in urls:
            yield scrapy.Request(url=url, callback=self.parse)

    def parse(self, response):
        for quote in response.css("div.quote"):
            yield {"text": quote.css("span.text::text").get()}

    def stop(self):
        self.crawler.engine.close_spider(self, "timeout")

which raises Exception, becomes:

import scrapy


class QuotesSpider(scrapy.Spider):
    name = "quotes"

    def __init__(self, *args, **kwargs):
        self.timeout = int(kwargs.pop("timeout", "60"))
        super(QuotesSpider, self).__init__(*args, **kwargs)

    async def start(self):
        from twisted.internet import reactor

        reactor.callLater(self.timeout, self.stop)

        urls = ["https://quotes.toscrape.com/page/1"]
        for url in urls:
            yield scrapy.Request(url=url, callback=self.parse)

    def parse(self, response):
        for quote in response.css("div.quote"):
            yield {"text": quote.css("span.text::text").get()}

    def stop(self):
        self.crawler.engine.close_spider(self, "timeout")

If this setting is set None, Scrapy will use the existing reactor if one is already installed, or install the default reactor defined by Twisted for the current platform.

Changed in version 2.7: The startproject command now sets this setting to twisted.internet.asyncioreactor.AsyncioSelectorReactor in the generated settings.py file.

Changed in version 2.13: The default value was changed from None to "twisted.internet.asyncioreactor.AsyncioSelectorReactor".

For additional information, see Choosing a Reactor and GUI Toolkit Integration.

URLLENGTH_LIMIT

Default: 2083

Scope: spidermiddlewares.urllength

The maximum URL length to allow for crawled URLs.

This setting can act as a stopping condition in case of URLs of ever-increasing length, which may be caused for example by a programming error either in the target server or in your code. See also REDIRECT_MAX_TIMES and DEPTH_LIMIT.

Use 0 to allow URLs of any length.

The default value is copied from the Microsoft Internet Explorer maximum URL length, even though this setting exists for different reasons.

USER_AGENT

Default: "Scrapy/VERSION (+https://scrapy.org)"

The default User-Agent to use when crawling, unless overridden. This user agent is also used by RobotsTxtMiddleware if ROBOTSTXT_USER_AGENT setting is None and there is no overriding User-Agent header specified for the request.

WARN_ON_GENERATOR_RETURN_VALUE

Default: True

When enabled, Scrapy will warn if generator-based callback methods (like parse) contain return statements with non-None values. This helps detect potential mistakes in spider development.

Disable this setting to prevent syntax errors that may occur when dynamically modifying generator function source code during runtime, skip AST parsing of callback functions, or improve performance in auto-reloading development environments.

Settings documented elsewhere:

The following settings are documented elsewhere, please check each specific case to see how to enable and use them.