`site` — MediaWiki sites#

Library module representing MediaWiki sites (wikis).

`BaseSite` — Base Class for Sites#

Objects with site methods independent of the communication interface.

class pywikibot.site._basesite.BaseSite(code, fam=None, user=None)[source]#

Bases: ComparableMixin

Site methods that are independent of the communication interface.

Parameters:

code (str) – the site’s language code
fam (str or pywikibot.family.Family) – wiki family name (optional)
user (str) – bot user name (optional)

category_redirects(fallback: str = '_default')#

Return list of category redirect templates.

See also

family.Family.category_redirects()

Return type:: list[str]

get_edit_restricted_templates()#

Return tuple of edit restricted templates.

Added in version 3.0.

Return type:: tuple[str, …]

get_archived_page_templates()#

Return tuple of edit restricted templates.

Added in version 3.0.

Return type:: tuple[str, …]

disambig(fallback='_default')#

Return list of disambiguation templates.

See also

family.Family.disambig()

Parameters:: fallback (str | None)
Return type:: list[str]

protocol()#

The protocol to use to connect to the site.

May be overridden to return ‘http’. Other protocols are not supported.

Changed in version 8.2: https is returned instead of http.

See also

family.Family.protocol()

Returns:: protocol that this family uses

category_on_one_line()[source]#: Return True if this site wants all category links on one line.

property code#

The identifying code for this Site equal to the wiki prefix.

By convention, this is usually an ISO language code, but it does not have to be.

property codes: set[str]#: Return set of all valid site codes for this site’s Family.

Added in version 9.6.

See also

family.Family.codes

disambcategory()[source]#: Return Category in which disambig pages are listed.

property doc_subpage: tuple#: Return the documentation subpage for this Site.

property family#: The Family object for this Site’s wiki family.

getSite(code)[source]#: Return Site object for language ‘code’ in this Family.

interwiki_putfirst()[source]#: Return list of language codes for ordering of interwiki links.

isInterwikiLink(text)[source]#

Return True if text is in the form of an interwiki link.

If a link object constructed using “text” as the link text parses as belonging to a different site, this method returns True.

property lang#

The ISO language code for this Site.

Presumed to be equal to the site code, but this can be overridden.

languages()[source]#

Return list of all valid site codes for this site’s Family.

Deprecated since version 9.6: Use codes() instead.

Return type:: list[str]

lock_page(page, block=True)[source]#

Lock page for writing. Must be called before writing any page.

We don’t want different threads trying to write to the same page at the same time, even to different sections.

Parameters:

page (Page) – the page to be locked
block (bool) – if true, wait until the page is available to be locked; otherwise, raise an exception if page can’t be locked

property namespaces#: Return dict of valid namespaces on this wiki.

ns_normalize(value)[source]#

Return canonical local form of namespace name.

Parameters:: value (str) – A namespace name

pagename2codes()[source]#

Return list of localized PAGENAMEE tags for the site.

Return type:: list[str]

pagenamecodes()[source]#

Return list of localized PAGENAME tags for the site.

Return type:: list[str]

redirect()[source]#

Return a default redirect tag for the site.

Changed in version 8.4: return a single generic redirect tag instead of a list of tags. For the list use redirects() instead.

Return type:: str

property redirect_regex: Pattern[str]#

Return a compiled regular expression matching on redirect pages.

Group 1 in the regex match object will be the target title.

A redirect starts with hash (#), followed by a keyword, then arbitrary stuff, then a wikilink. The wikilink may contain a label, although this is not useful.

Added in version 8.4: moved from class:APISite

redirects()[source]#

Return list of generic redirect tags for the site.

See also

redirect() for the default redirect tag.

Added in version 8.4.

Return type:: list[str]

sametitle(title1, title2)[source]#

Return True if title1 and title2 identify the same wiki page.

title1 and title2 may be unequal but still identify the same page, if they use different aliases for the same namespace.

Parameters:

title1 (str)
title2 (str)

Return type:

bool

property sitename#: String representing this Site’s name and code.

property throttle#: Return this Site’s throttle. Initialize a new one if needed.

unlock_page(page)[source]#

Unlock page. Call as soon as a write operation has completed.

Parameters:: page (Page) – the page to be locked
Return type:: None

property use_hard_category_redirects#

Hard redirects are used for this site.

Originally create as property for future use for a proposal to replace category redirect templates with hard redirects. This was never implemented and is not used inside the framework.

Deprecated since version 8.5.

user()[source]#

Return the currently-logged in bot username, or None.

Return type:: str | None

username()[source]#

Return the username used for the site.

Return type:: str | None

validLanguageLinks()[source]#: Return list of language codes to be used in interwiki links.

class pywikibot.site._basesite.BaseSite(code, fam=None, user=None)[source]#

Bases: ComparableMixin

Site methods that are independent of the communication interface.

Parameters:

code (str) – the site’s language code
fam (str or pywikibot.family.Family) – wiki family name (optional)
user (str) – bot user name (optional)

category_on_one_line()[source]#: Return True if this site wants all category links on one line.

property code#

The identifying code for this Site equal to the wiki prefix.

By convention, this is usually an ISO language code, but it does not have to be.

property codes: set[str]#: Return set of all valid site codes for this site’s Family.

Added in version 9.6.

See also

family.Family.codes

disambcategory()[source]#: Return Category in which disambig pages are listed.

property doc_subpage: tuple#: Return the documentation subpage for this Site.

property family#: The Family object for this Site’s wiki family.

getSite(code)[source]#: Return Site object for language ‘code’ in this Family.

interwiki_putfirst()[source]#: Return list of language codes for ordering of interwiki links.

isInterwikiLink(text)[source]#

Return True if text is in the form of an interwiki link.

If a link object constructed using “text” as the link text parses as belonging to a different site, this method returns True.

property lang#

The ISO language code for this Site.

Presumed to be equal to the site code, but this can be overridden.

languages()[source]#

Return list of all valid site codes for this site’s Family.

Deprecated since version 9.6: Use codes() instead.

Return type:: list[str]

lock_page(page, block=True)[source]#

Lock page for writing. Must be called before writing any page.

We don’t want different threads trying to write to the same page at the same time, even to different sections.

Parameters:

page (Page) – the page to be locked
block (bool) – if true, wait until the page is available to be locked; otherwise, raise an exception if page can’t be locked

property namespaces#: Return dict of valid namespaces on this wiki.

ns_normalize(value)[source]#

Return canonical local form of namespace name.

Parameters:: value (str) – A namespace name

pagename2codes()[source]#

Return list of localized PAGENAMEE tags for the site.

Return type:: list[str]

pagenamecodes()[source]#

Return list of localized PAGENAME tags for the site.

Return type:: list[str]

redirect()[source]#

Return a default redirect tag for the site.

Changed in version 8.4: return a single generic redirect tag instead of a list of tags. For the list use redirects() instead.

Return type:: str

property redirect_regex: Pattern[str]#

Return a compiled regular expression matching on redirect pages.

Group 1 in the regex match object will be the target title.

A redirect starts with hash (#), followed by a keyword, then arbitrary stuff, then a wikilink. The wikilink may contain a label, although this is not useful.

Added in version 8.4: moved from class:APISite

redirects()[source]#

Return list of generic redirect tags for the site.

See also

redirect() for the default redirect tag.

Added in version 8.4.

Return type:: list[str]

sametitle(title1, title2)[source]#

Return True if title1 and title2 identify the same wiki page.

title1 and title2 may be unequal but still identify the same page, if they use different aliases for the same namespace.

Parameters:

title1 (str)
title2 (str)

Return type:

bool

property sitename#: String representing this Site’s name and code.

property throttle#: Return this Site’s throttle. Initialize a new one if needed.

unlock_page(page)[source]#

Unlock page. Call as soon as a write operation has completed.

Parameters:: page (Page) – the page to be locked
Return type:: None

property use_hard_category_redirects#

Hard redirects are used for this site.

Originally create as property for future use for a proposal to replace category redirect templates with hard redirects. This was never implemented and is not used inside the framework.

Deprecated since version 8.5.

user()[source]#

Return the currently-logged in bot username, or None.

Return type:: str | None

username()[source]#

Return the username used for the site.

Return type:: str | None

validLanguageLinks()[source]#: Return list of language codes to be used in interwiki links.

`APISite` — API Interface for Sites#

Objects representing API interface to MediaWiki site.

class pywikibot.site._apisite.APISite(code, fam=None, user=None)[source]#

Bases: BaseSite, EchoMixin, GeneratorsMixin, GeoDataMixin, GlobalUsageMixin, LinterMixin, PageImagesMixin, ProofreadPageMixin, TextExtractsMixin, ThanksMixin, UrlShortenerMixin, WikibaseClientMixin

API interface to MediaWiki site.

Do not instantiate directly; use pywikibot.Site function.

Parameters:

code (str)
fam (str | pywikibot.family.Family | None)
user (str | None)

property articlepath: str#: Get the nice article path with ``{}``placeholder.

Added in version 7.0.

static assert_valid_iter_params(msg_prefix, start, end, reverse, is_ts=True)[source]#

Validate iterating API parameters.

Parameters:

msg_prefix (str) – The calling method name
start (datetime | int | str) – The start value to compare
end (datetime | int | str) – The end value to compare
reverse (bool) – The reverse option
is_ts (bool) – When comparing timestamps (with is_ts=True) the start is usually greater than end. Comparing titles this is vice versa.

Raises:

AssertionError – start/end values are not comparabel types or are in the wrong order

Return type:

None

blockuser(user, expiry, reason, anononly=True, nocreate=True, autoblock=True, noemail=False, reblock=False, allowusertalk=False)[source]#

Block a user for certain amount of time and for a certain reason.

See also

API:Block

Parameters:

user (pywikibot.page.User) – The username/IP to be blocked without a namespace.
expiry (datetime.datetime | str | bool) –
The length or date/time when the block expires. If ‘never’, ‘infinite’, ‘indefinite’ it never does. If the value is given as a str it’s parsed by php’s strtotime function:

https://www.php.net/manual/en/function.strtotime.php

The relative format is described there:

https://www.php.net/manual/en/datetime.formats.relative.php

It is recommended to not use a str if possible to be independent of the API.
reason (str) – The reason for the block.
anononly (bool) – Disable anonymous edits for this IP.
nocreate (bool) – Prevent account creation.
autoblock (bool) – Automatically block the last used IP address and all subsequent IP addresses from which this account logs in.
noemail (bool) – Prevent user from sending email through the wiki.
reblock (bool) – If the user is already blocked, overwrite the existing block.
allowusertalk (bool) – Whether the user can edit their talk page while blocked.

Returns:

The data retrieved from the API request.

Return type:

dict[str, Any]

categoryinfo(category)[source]#

Retrieve data on contents of category.

Parameters:: category (Category)
Return type:: dict[str, int]

compare(old, diff, difftype='table')[source]#

Corresponding method to the ‘action=compare’ API action.

Hint

Use diff.html_comparator() function to parse result.

See also

API:Compare

Parameters:

old (_CompType) – starting revision ID, title, Page, or Revision
diff (_CompType) – ending revision ID, title, Page, or Revision
difftype (str) – type of diff. One of ‘table’ or ‘inline’.

Returns:

Returns an HTML string of a diff between two revisions.

Return type:

str

data_repository()[source]#

Return the data repository connected to this site.

Returns:: The data repository if one is connected or None otherwise.
Return type:: DataSite | None

dbName()[source]#

Return this site’s internal id.

Return type:: str

delete(page, reason, *, deletetalk=False, oldimage=None)[source]#

Delete a page or a specific old version of a file from the wiki.

Requires appropriate privileges.

See also

API:Delete

Page to be deleted can be given either as Page object or as pageid. To delete a specific version of an image the oldimage identifier must be provided.

Added in version 6.1: renamed from deletepage

Changed in version 6.1: keyword only parameter oldimage was added.

Changed in version 7.1: keyword only parameter deletetalk was added.

Changed in version 8.1: raises exceptions.NoPageError if page does not exist.

Parameters:

page (BasePage | int | str) – Page to be deleted or its pageid.
reason (str) – Deletion reason.
deletetalk (bool) – Also delete the talk page, if it exists.
oldimage (str | None) – oldimage id of the file version to be deleted. If a BasePage object is given with page parameter, it has to be a FilePage.

Raises:

TypeError, ValueError – page has wrong type/value.

Return type:

None

deleterevs(targettype, ids, *, hide=None, show=None, reason='', target=None)[source]#

Delete or undelete specified page revisions, file versions or logs.

See also

API:Revisiondelete

If more than one target id is provided, the same action is taken for all of them.

Added in version 6.0.

Parameters:

targettype (str) – Type of target. One of “archive”, “filearchive”, “logging”, “oldimage”, “revision”.
ids (int | str | list[int | str]) – Identifiers for the revision, log, file version or archive.
hide (str | list[str] | None) – What to delete. Can be “comment”, “content”, “user” or a combination of them in pipe-separate form such as “comment|user”.
show (str | list[str] | None) – What to undelete. Can be “comment”, “content”, “user” or a combination of them in pipe-separate form such as “comment|user”.
reason (str) – Deletion reason.
target (pywikibot.page.Page | str | None) – Page object or page title, if required for the type.

Return type:

None

editpage(page, summary=None, minor=True, notminor=False, bot=True, recreate=True, createonly=False, nocreate=False, watch=None, **kwargs)[source]#

Submit an edit to be saved to the wiki.

See also

API:Edit
BasePage.save() (should be preferred)

Parameters:

page (BasePage) – The Page to be saved. By default its .text property will be used as the new text to be saved to the wiki
summary (str | None) – The edit summary for the modification (optional, but most wikis strongly encourage its use)
minor (bool) – if True (default), mark edit as minor
notminor (bool) – if True, override account preferences to mark edit as non-minor
recreate (bool) – if True (default), create new page even if this title has previously been deleted
createonly (bool) – if True, raise an error if this title already exists on the wiki
nocreate (bool) – if True, raise a exceptions.NoCreateError exception if the page does not exist
watch (str | None) –
Specify how the watchlist is affected by this edit, set to one of watch, unwatch, preferences, nochange:
- watch — add the page to the watchlist
- unwatch — remove the page from the watchlist
- preferences — use the preference settings (default)
- nochange — don’t change the watchlist
If None (default), follow bot account’s default settings
bot (bool) – if True and bot right is given, mark edit with bot flag
kwargs (Any)

Keyword Arguments:

text (str) – Overrides Page.text
section (int | str) – Edit an existing numbered section or a new section (‘new’)
prependtext (str) – Prepend text. Overrides Page.text
appendtext (str) – Append text. Overrides Page.text.
undo (int) – Revision id to undo. Overrides Page.text

Returns:

True if edit succeeded, False if it failed

Raises:

AbuseFilterDisallowedError – This action has been automatically identified as harmful, and therefore disallowed
CaptchaError – config.solve_captcha is False and saving the page requires solving a captcha
CascadeLockedPageError – The page is protected with protection cascade
EditConflictError – an edit conflict occurred
Error – No text to be saved or API editing not enabled on site or user is not authorized to edit, create pages or create image redirects on site or bot is not logged in and anon users are not authorized to edit, create pages or to create image redirects or the edit was filtered or the content is too big
KeyError – No ‘result’ found in API response
LockedNoPageError – The page title is protected
LockedPageError – The page has been protected to prevent editing or other actions
NoCreateError – The page you specified doesn’t exist and nocreate is set
NoPageError – recreate is disabled and page does not exist
PageCreatedConflictError – The page you tried to create has been created already
PageDeletedConflictError – The page has been deleted in meantime
SpamblacklistError – The title is blacklisted as spam
TitleblacklistError – The title is blacklisted
ValueError – text keyword is used with one of the override keywords appendtext, prependtext or undo or more than one of the override keywords are used or no text keyword is used together with section keyword.

Return type:

bool

expand_text(text, title=None, includecomments=None)[source]#

Parse the given text for preprocessing and rendering.

e.g expand templates and strip comments if includecomments parameter is not True. Keeps text inside <nowiki></nowiki> tags unchanges etc. Can be used to parse magic parser words like {{CURRENTTIMESTAMP}}.

Parameters:

text (str) – text to be expanded
title (str | None) – page title without section
includecomments (bool | None) – if True do not strip comments

Return type:

str

property file_extensions: list[str]#: File extensions enabled on the wiki.

Added in version 8.4.

Changed in version 9.2: also include extensions from the image repository

static fromDBName(dbname, site=None)[source]#

Create a site from a database name using the sitematrix.

Changed in version 8.3.3: changed from classmethod to staticmethod.

Parameters:

dbname (str) – database name
site (BaseSite | None) – Site to load sitematrix from. (Default meta.wikimedia.org)

Returns:

site object for the database name

Return type:

BaseSite

get_globaluserinfo(user=None, force=False)[source]#

Retrieve globaluserinfo from site and cache it.

Added in version 7.0.

Parameters:

user (str | int | None) – The user name or user ID whose global info is retrieved. Defaults to the current user.
force (bool) – Whether the cache should be discarded.

Returns:

A dict with the following keys and values:

id: user id (numeric str)
home: dbname of home wiki
registration: registration date as Timestamp
groups: list of groups (could be empty)
rights: list of rights (could be empty)
editcount: global editcount

Raises:

TypeError – Inappropriate argument type of ‘user’

Return type:

dict[str, Any]

get_parsed_page(page)[source]#

Retrieve parsed text of the page using action=parse.

Changed in version 7.1: raises KeyError instead of AssertionError

See also

Parameters:: page (BasePage)
Return type:: str

get_property_names(force=False)[source]#

Get property names for pages_with_property().

See also

API:Pagepropnames

Parameters:: force (bool) – force to retrieve userinfo ignoring cache
Return type:: list[str]

get_searched_namespaces(force=False)[source]#

Retrieve the default searched namespaces for the user.

If no user is logged in, it returns the namespaces used by default. Otherwise it returns the user preferences. It caches the last result and returns it, if the username or login status hasn’t changed.

Parameters:: force (bool) – Whether the cache should be discarded.
Returns:: The namespaces which are searched by default.
Return type:: set[Namespace]

get_tokens(types, *args, **kwargs)[source]#

Preload one or multiple tokens.

Usage

>>> site = pywikibot.Site()
>>> tokens = site.get_tokens([])  # get all tokens
>>> list(tokens.keys())  # result depends on user
['createaccount', 'login']
>>> tokens = site.get_tokens(['csrf', 'patrol'])
>>> list(tokens.keys())
['csrf', 'patrol']
>>> token = site.get_tokens(['csrf']).get('csrf')  # get a single token
>>> token
'a9f...0a0+\\'
>>> token = site.get_tokens(['unknown'])  # try an invalid token
...
... # invalid token names shows a warning and the key is not in result
...
WARNING: API warning (tokens) of unknown format:
... {'warnings': 'Unrecognized value for parameter "type": foo'}
{}

You should not call this method directly, especially if you only need a specific token. Use tokens property instead.

Changed in version 8.0: all parameter is deprecated. Use an empty list for types instead.

Note

args and kwargs are not used for deprecation warning only.

See also

API:Tokens

Parameters:: types (list[str]) – the types of token (e.g., “csrf”, “login”, “patrol”). If the list is empty all available tokens are loaded. See API documentation for full list of types.
Returns:: a dict with retrieved valid tokens.
Return type:: dict[str, str]

getcategoryinfo(category)[source]#

Retrieve data on contents of category.

See also

API:Categoryinfo

Parameters:: category (Category)
Return type:: None

getcurrenttimestamp()[source]#

Return the server time as a MediaWiki timestamp string.

It calls server_time first so it queries the server to get the current server time.

Returns:: the server time (as ‘yyyymmddhhmmss’)
Return type:: str

getmagicwords(word)[source]#

Return list of localized “word” magic words for the site.

Parameters:: word (str)
Return type:: list[str]

getredirtarget(page, *, ignore_section=True)[source]#

Return page object for the redirect target of page.

Added in version 9.3: ignore_section parameter

See also

page.BasePage.getRedirectTarget()

Parameters:

page (BasePage) – page to search redirects for
ignore_section (bool) – do not include section to the target even the link has one

Returns:

redirect target of page

Raises:

CircularRedirectError – page is a circular redirect
InterwikiRedirectPageError – the redirect target is on another site
IsNotRedirectPageError – page is not a redirect
RuntimeError – no redirects found
SectionError – the section is not found on target page and ignore_section is not set

Return type:

pywikibot.page.Page

property globaluserinfo: dict[str, Any]#

Retrieve globaluserinfo of the current user from site.

To get globaluserinfo for a given user or user ID use get_globaluserinfo() method instead

Added in version 3.0.

has_all_mediawiki_messages(keys, lang=None)[source]#

Confirm that the site defines a set of MediaWiki messages.

Parameters:

keys (Iterable[str]) – names of MediaWiki messages
lang (str | None) – a language code, default is self.lang

Return type:

bool

property has_data_repository: bool#: Return True if site has a shared data repository like Wikidata.

has_extension(name)[source]#

Determine whether extension name is loaded.

Parameters:: name (str) – The extension to check for, case sensitive
Returns:: If the extension is loaded
Return type:: bool

has_group(group)[source]#

Return true if and only if the user is a member of specified group.

Possible values of ‘group’ may vary depending on wiki settings, but will usually include bot.

See also

API:Userinfo

Parameters:: group (str)
Return type:: bool

property has_image_repository: bool#: Return True if site has a shared image repository like Commons.

has_mediawiki_message(key, lang=None)[source]#

Determine if the site defines a MediaWiki message.

Parameters:

key (str) – name of MediaWiki message
lang (str | None) – a language code, default is self.lang

Return type:

bool

has_right(right)[source]#

Return true if and only if the user has a specific right.

Possible values of ‘right’ may vary depending on wiki settings.

See also

API:Userinfo

Parameters:: right (str) – a specific right to be validated
Return type:: bool

image_repository()[source]#

Return Site object for image repository e.g. commons.

Return type:: BaseSite | None

interwiki(prefix)[source]#

Return the site for a corresponding interwiki prefix.

Raises:

pywikibot.exceptions.SiteDefinitionError – if the url given in the interwiki table doesn’t match any of the existing families.
KeyError – if the prefix is not an interwiki prefix.

Parameters:

prefix (str)

Return type:

BaseSite

interwiki_prefix(site)[source]#

Return the interwiki prefixes going to that site.

The interwiki prefixes are ordered first by length (shortest first) and then alphabetically. interwiki(prefix) is not guaranteed to equal site (i.e. the parameter passed to this function).

Parameters:: site (BaseSite) – The targeted site, which might be it’s own.
Raises:: KeyError – if there is no interwiki prefix for that site.
Return type:: list[str]

isBot(username)[source]#

Return True is username is a bot user.

Parameters:: username (str)
Return type:: bool

is_blocked(force=False)[source]#

Return True when logged in user is blocked.

To check whether a user can perform an action, the method has_right should be used.

See also

API:Userinfo

Added in version 7.0: The force parameter

Parameters:: force (bool) – Whether the cache should be discarded.
Return type:: bool

is_data_repository()[source]#

Return True if its data repository is itself.

Return type:: bool

is_image_repository()[source]#

Return True if Site object is the image repository.

Return type:: bool

is_locked(user=None, force=False)[source]#

Return True when given user is locked globally.

Added in version 7.0.

Parameters:

user (str | int | None) – The user name or user ID. Defaults to the current user.
force (bool) – Whether the cache should be discarded.

Return type:

bool

is_oauth_token_available()[source]#

Check whether OAuth token is set for this site.

Return type:: bool

is_uploaddisabled()[source]#

Return True if upload is disabled on site.

Example:

>>> site = pywikibot.Site('commons')
>>> site.is_uploaddisabled()
False
>>> site = pywikibot.Site('wikidata')
>>> site.is_uploaddisabled()
True

Return type:: bool

property lang: str#: Return the code for the language of this Site.

linktrail()[source]#

Build linktrail regex from siteinfo linktrail.

Letters that can follow a wikilink and are regarded as part of this link. This depends on the linktrail setting in LanguageXx.php

Added in version 7.3.

Returns:: The linktrail regex.
Return type:: str

list_to_text(args)[source]#

Convert a list of strings into human-readable text.

The MediaWiki messages ‘and’ and ‘word-separator’ are used as separator between the last two arguments. If more than two arguments are given, other arguments are joined using MediaWiki message ‘comma-separator’.

Parameters:: args (Iterable[str]) – text to be expanded
Return type:: str

loadimageinfo(page, history=False, url_width=None, url_height=None, url_param=None, timestamp=None)[source]#

Load image info from api and save in page attributes.

The following properties are loaded: timestamp, user, comment, url, size, sha1, mime, mediatype, archivename and bitdepth. metadata is loaded only if history is False. If url_width, url_height or url_param is given, additional properties thumbwidth, thumbheight, thumburl and responsiveUrls are given.

Note

Parameters validation and error handling left to the API call.

Changed in version 8.2: mediatype and bitdepth properties were added.

Changed in version 8.6.: Added timestamp parameter. Metadata are loaded only if history is False.

See also

API:Imageinfo

Parameters:

history (bool) – if true, return the image’s version history
url_width (int | None) – get info for a thumbnail with given width
url_height (int | None) – get info for a thumbnail with given height
url_param (str | None) – get info for a thumbnail with given param
timestamp (Timestamp | None) – timestamp of the image’s version to retrieve. It has effect only if history is False. If omitted, the latest version will be fetched.
page (FilePage)

Return type:

None

loadpageinfo(page, preload=False)[source]#

Load page info from api and store in page attributes.

See also

API:Info

Parameters:

page (BasePage)
preload (bool)

Return type:

None

loadpageprops(page)[source]#

Load page props for the given page.

Parameters:: page (BasePage)
Return type:: None

local_interwiki(prefix)[source]#

Return whether the interwiki prefix is local.

A local interwiki prefix is handled by the target site like a normal link. So if that link also contains an interwiki link it does follow it as long as it’s a local link.

Raises:

pywikibot.exceptions.SiteDefinitionError – if the url given in the interwiki table doesn’t match any of the existing families.
KeyError – if the prefix is not an interwiki prefix.

Parameters:

prefix (str)

Return type:

bool

logged_in()[source]#

Verify the bot is logged into the site as the expected user.

The expected usernames are those provided as the user parameter at instantiation.

Return type:: bool

login(autocreate=False, user=None, *, cookie_only=False)[source]#

Log the user in if not already logged in.

Changed in version 8.0: lazy load cookies when logging in. This was dropped in 8.0.4

Changed in version 8.0.4: the cookie_only parameter was added and cookies are loaded whenever the site is initialized.

See also

API:Login

Parameters:

autocreate (bool) – if true, allow auto-creation of the account using unified login
user (str | None) – bot user name. Overrides the username set by BaseSite initializer parameter or user config setting
cookie_only (bool) – Only try to login from cookie but do not force to login with username/password settings.

Raises:

pywikibot.exceptions.NoUsernameError – Username is not recognised by the site.

Return type:

None

logout()[source]#

Logout of the site and load details for the logged out user.

Also logs out of the global account if linked to the user.

See also

API:Logout

Raises:: APIError – Logout is not available when OAuth enabled.
Return type:: None

property logtypes: set[str]#: Return a set of log types available on current site.

property maxlimit: int#: Get the maximum limit of pages to be retrieved.

Added in version 7.0.

mediawiki_message(key, lang=None)[source]#

Fetch the text for a MediaWiki message.

Parameters:

key (str) – name of MediaWiki message
lang (str | None) – a language code, default is self.lang

Return type:

str

mediawiki_messages(keys, lang=None)[source]#

Fetch the text of a set of MediaWiki messages.

The returned dict uses each key to store the associated message.

See also

API:Allmessages

Parameters:

keys (Iterable[str]) – MediaWiki messages to fetch
lang (str | None) – a language code, default is self.lang

Return type:

OrderedDict[str, str]

merge_history(source, dest, timestamp=None, reason=None)[source]#

Merge revisions from one page into another.

See also

API:Mergehistory
page.BasePage.merge_history() (should be preferred)

Revisions dating up to the given timestamp in the source will be moved into the destination page history. History merge fails if the timestamps of source and dest revisions overlap (all source revisions must be dated before the earliest dest revision).

Parameters:

source (BasePage) – Source page from which revisions will be merged
dest (BasePage) – Destination page to which revisions will be merged
timestamp (Timestamp | None) – Revisions from this page dating up to this timestamp will be merged into the destination page (if not given or False, all revisions will be merged)
reason (str | None) – Optional reason for the history merge

Raises:

APIError – unexpected APIError
Error – expected APIError or unexpected response
NoPageError – source or dest does not exist
PageSaveRelatedError – source is equal to dest

Return type:

None

messages()[source]#

Return true if the user has new messages, and false otherwise.

Deprecated since version 8.0: Replaced by userinfo['messages'].

Return type:: bool

property months_names: list[tuple[str, str]]#

Obtain month names from the site messages.

The list is zero-indexed, ordered by month in calendar, and should be in the original site language.

Returns:: list of tuples (month name, abbreviation)

movepage(page, newtitle, summary, movetalk=True, noredirect=False, movesubpages=True)[source]#

Move a Page to a new title.

See also

API:Move

Changed in version 7.2: The movesubpages parameter was added

Parameters:

page (BasePage) – the Page to be moved (must exist)
newtitle (str) – the new title for the Page
summary (str) – edit summary (required!)
movetalk (bool) – if True (default), also move the talk page if possible
noredirect (bool) – if True, suppress creation of a redirect from the old title to the new one
movesubpages (bool) – Rename subpages, if applicable.

Returns:

Page object with the new title

Return type:

pywikibot.page.Page

property mw_version: MediaWikiVersion#

Return version() as a tools.MediaWikiVersion object.

Cache the result for 24 hours.

namespace(num, all_ns=False, all='[deprecated name of all_ns]')[source]#

Return string containing local name of namespace ‘num’.

If optional argument all_ns is true, return all recognized values for this namespace.

Changed in version 9.0: all parameter was renamed to all_ns.

Parameters:

num (int) – Namespace constant.
all_ns (bool) – If True return a Namespace object. Otherwise return the namespace name.

Returns:

local name or Namespace object

nice_get_address(title)[source]#

Return shorter URL path to retrieve page titled ‘title’.

Parameters:: title (str)
Return type:: str

page_can_be_edited(page, action='edit')[source]#

Determine if the page can be modified.

Return True if the bot has the permission of needed restriction level for the given action type.

See also

page.BasePage.has_permission() (should be preferred)

Parameters:

page (BasePage) – a pywikibot.page.BasePage object
action (str) – a valid restriction type like ‘edit’, ‘move’

Raises:

ValueError – invalid action parameter

Return type:

bool

page_from_repository(item)[source]#

Return a Page for this site object specified by Wikibase item.

Usage:

>>> site = pywikibot.Site('wikipedia:zh')
>>> page = site.page_from_repository('Q131303')
>>> page.title()
'Hello World'

This method is able to upcast categories:

>>> site = pywikibot.Site('commons')
>>> page = site.page_from_repository('Q131303')
>>> page.title()
'Category:Hello World'
>>> page
Category('Category:Hello World')

It also works for wikibase repositories:

>>> site = pywikibot.Site('wikidata')
>>> page = site.page_from_repository('Q5296')
>>> page.title()
'Wikidata:Main Page'

If no page exists for a given site, None is returned:

>>> site = pywikibot.Site('wikidata')
>>> page = site.page_from_repository('Q131303')
>>> page is None
True

Changed in version 7.7: No longer raise NotimplementedError if used with a Wikibase site.

Parameters:: item (str) – id number of item, “Q###”,
Returns:: Page, or Category object given by Wikibase item number for this site object.
Raises:: pywikibot.exceptions.UnknownExtensionError – site has no Wikibase extension
Return type:: Page | None

page_isredirect(page)[source]#

Return True if and only if page is a redirect.

Parameters:: page (BasePage)
Return type:: bool

page_restrictions(page)[source]#

Return a dictionary reflecting page protections.

Example:

>>> site = pywikibot.Site('wikipedia:test')
>>> page = pywikibot.Page(site, 'Main Page')
>>> site.page_restrictions(page)
{'edit': ('sysop', 'infinity'), 'move': ('sysop', 'infinity')}

See also

page.BasePage.protection() (should be preferred)

Parameters:: page (BasePage)
Return type:: dict[str, tuple[str, str]]

pagename2codes()[source]#

Return list of localized PAGENAMEE tags for the site.

Return type:: list[str]

pagenamecodes()[source]#

Return list of localized PAGENAME tags for the site.

Return type:: list[str]

protect(page, protections, reason, expiry=None, **kwargs)[source]#

(Un)protect a wiki page. Requires protect right.

See also

Parameters:

protections (dict[str, str | None]) – A dict mapping type of protection to protection level of that type. Refer protection_types() for valid restriction types and protection_levels() for valid restriction levels. If None is given, however, that protection will be skipped.
reason (str) – Reason for the action
expiry (datetime.datetime | str | None) – When the block should expire. This expiry will be applied to all protections. If None, 'infinite', 'indefinite', 'never', or '' is given, there is no expiry.
page (BasePage)
kwargs (Any)

Return type:

None

protection_levels()[source]#

Return the protection levels available on this site.

Example:

>>> site = pywikibot.Site('wikipedia:test')
>>> sorted(site.protection_levels())
['', 'autoconfirmed', ... 'sysop', 'templateeditor']

See also

Siteinfo._get_default()

Returns:: protection types available
Return type:: set[str]

protection_types()[source]#

Return the protection types available on this site.

Example:

>>> site = pywikibot.Site('wikipedia:test')
>>> sorted(site.protection_types())
['create', 'edit', 'move', 'upload']

See also

Siteinfo._get_default()

Returns:: protection types available
Return type:: set[str]

purgepages(pages, forcelinkupdate=False, forcerecursivelinkupdate=False, converttitles=False, redirects=False)[source]#

Purge the server’s cache for one or multiple pages.

Parameters:

pages (list[BasePage]) – list of Page objects
redirects (bool) – Automatically resolve redirects.
converttitles (bool) – Convert titles to other variants if necessary. Only works if the wiki’s content language supports variant conversion.
forcelinkupdate (bool) – Update the links tables.
forcerecursivelinkupdate (bool) – Update the links table, and update the links tables for any page that uses this page as a template.

Returns:

True if API returned expected response; False otherwise

Return type:

bool

ratelimit(action)[source]#

Get the rate limit for a given action.

This method get the ratelimit for a given action and returns a tools.collections.RateLimit namedtuple which has the following fields and properties:

group — The current user group returned by the API. If the user is not logged in, the group will be ‘ip’.
hits — rate limit hits; API requests should not exceed this limit value for the given action.
seconds — time base in seconds for the maximum hits
delay — (property) calculated as seconds per hits which may be used for wait cycles.
ratio — (property) inverse of delay, calculated as hits per seconds. The result may be Infinite.

If the user has ‘noratelimit’ rights, maxlimit() is used for hits and seconds will be 0. ‘noratelimit’ is returned as group parameter in that case.

If no rate limit is found for the given action, maxlimit() is used for hits and seconds will be config.put_throttle. As group parameter ‘unknown’ is returned in that case.

Examples:

This is an example for a bot user which is not logged in. The rate limit user group is ‘ip’

>>> site = pywikibot.Site()
>>> limit = site.ratelimit('edit')  # get rate limit for 'edit' action
>>> limit
RateLimit(group='ip', hits=8, seconds=60)
>>> limit.delay  # delay and ratio must be get as attributes
7.5
>>> site.ratelimit('purge').hits  # get purge hits
30
>>> group, *limit = site.ratelimit('urlshortcode')
>>> group  # the user is not logged in, we get 'ip' as group
'ip'
>>> limit  # starred assignment is allowed for the fields
[10, 120]

After login to the site and the rate limit will change. The limit user group might be ‘user’:

>>> limit = site.ratelimit('edit')
>>> limit
RateLimit(group='user', hits=90, seconds=60)
>>> limit.ratio
1.5
>>> limit = site.ratelimit('urlshortcode')  # no action limit found
>>> group, *limits = limit
>>> group
'unknown'  # the group is 'unknown' because action was not found
>>> limits
(50, 10)  # hits is maxlimit and seconds is config.put_throttle
>>> site.maxlimit, pywikibot.config.put_throttle
(50, 10)

If a user is logged in and has no rate limit, e.g bot accounts, we always get a default RateLimit namedtuple like this:

>>> site.has_right['noratelimit']
True
>>> limit = site.ratelimit('any_action')  # maxlimit is used
>>> limit
RateLimit(group='noratelimit', hits=500, seconds=0)
>>> limit.delay, limit.ratio
(0.0, inf)

Note

It is not verified whether action parameter has a valid value.

See also

tools.collections.RateLimit for RateLimit examples.
API:Ratelimit

Added in version 9.0.

Parameters:: action (str) – action which might be limited
Returns:: RateLimit tuple with group, hits and seconds fields and properties for delay and ratio.
Return type:: RateLimit

redirects()[source]#

Return a list of localized tags for the site without preceding ‘#’.

Added in version 8.4.

Return type:: list[str]

rollbackpage(page, **kwargs)[source]#

Roll back page to version before last user’s edits.

See also

API:Rollback

The keyword arguments are those supported by the rollback API.

As a precaution against errors, this method will fail unless the page history contains at least two revisions, and at least one that is not by the same user who made the last edit.

Parameters:

page (BasePage) – the Page to be rolled back (must exist)
kwargs (Any)

Keyword Arguments:

user – the last user to be rollbacked; default is page.latest_revision.user

Return type:

None

server_time()[source]#

Return a Timestamp object representing the current server time.

It uses the ‘time’ property of the siteinfo ‘general’. It’ll force a reload before returning the time.

Returns:: the current server time
Return type:: Timestamp

simple_request(**kwargs)[source]#

Create a request by defining all kwargs as parameters.

Added in version 7.1: _simple_request becomes a public method

Parameters:: kwargs (Any)
Return type:: Request

property siteinfo: Siteinfo#: Site information dict.

stash_info(file_key, props=None)[source]#

Get the stash info for a given file key.

See also

API:Stashimageinfo

Parameters:

file_key (str)
props (list[str] | None)

Return type:

dict[str, Any]

property tokens: TokenWallet#

Return the TokenWallet collection.

TokenWallet collection holds all available tokens. The tokens are loaded via get_tokens() method with the first token request and is retained until the TokenWallet is cleared.

Usage:

>>> site = pywikibot.Site()
>>> token = site.tokens['csrf']
>>> token
'df8...9e6+\\'
>>> 'csrf' in site.tokens
... # Check whether the token exists
True
>>> 'invalid' in site.tokens
False
>>> token = site.tokens['invalid']
Traceback (most recent call last):
...
KeyError: "Invalid token 'invalid' for user ...
>>> site.tokens.clear()  # clears the internal cache
>>> site.tokens['csrf']
... # get a new token
'1c8...9d3+\\'
>>> del site.tokens  # another variant to clear the cache

Changed in version 8.0: tokens attribute became a property to enable deleter.

Warning

A deprecation warning is shown if the token name is outdated, see API:Tokens (action).

See also

API:Tokens for valid token types

unblockuser(user, reason=None)[source]#

Remove the block for the user.

See also

API:Block

Parameters:

user (pywikibot.page.User) – The username/IP without a namespace.
reason (str | None) – Reason for the unblock.

Return type:

dict[str, Any]

undelete(page, reason, *, revisions=None, fileids=None)[source]#

Undelete page from the wiki. Requires appropriate privilege level.

See also

API:Undelete

Added in version 6.1: renamed from undelete_page

Changed in version 6.1: fileids parameter was added, keyword argument required for revisions.

Parameters:

page (BasePage) – Page to be deleted.
reason (str) – Undeletion reason.
revisions (list[str] | None) – List of timestamps to restore. If None, restores all revisions.
fileids (list[int | str] | None) – List of fileids to restore.

Return type:

None

upload(filepage, **kwargs)[source]#

Upload a file to the wiki.

See also

API:Upload

Either source_filename or source_url, but not both, must be provided.

Changed in version 6.0: keyword arguments required for all parameters except filepage

Changed in version 6.2:: asynchronous upload is used if asynchronous parameter is set.

For keyword arguments refer pywikibot.site._upload.Uploader

Parameters:

filepage (pywikibot.page.FilePage) – a FilePage object from which the wiki-name of the file will be obtained.
kwargs (Any)

Returns:

It returns True if the upload was successful and False otherwise.

Return type:

bool

property userinfo: dict[str, Any]#

Retrieve userinfo from site and store in _userinfo attribute.

To force retrieving userinfo ignoring cache, just delete this property.

Usage

>>> site = pywikibot.Site('test')
>>> info = site.userinfo
>>> info['id']  # returns 0 if no ip user
...
0
>>> info['name']  # username or ip
...
...
'92.198.174.192'
>>> info['groups']
['*']
>>> info['rights']
['createaccount', 'read', 'edit', 'createpage', 'createtalk', ...]
>>> info['messages']
False
>>> del site.userinfo  # delete userinfo cache
>>> 'blockinfo' in site.userinfo
False
>>> 'anon' in site.userinfo
True

Useful alternatives to userinfo property

has_group() to verify the group membership
has_right() to verify that the user has a given right
logged_in() to verify the user is loggend in to a site

See also

API:Userinfo

Changed in version 8.0: Use API formatversion 2.

Returns:

A dict with the following keys and values:

id: user id (int)
name: username (if user is logged in)
anon: present if user is not logged in
groups: list of groups (could be empty)
rights: list of rights (could be empty)
messages: True if user has a new message on talk page (bool)
blockinfo: present if user is blocked (dict)

validate_tokens(types)[source]#

Validate if requested tokens are acceptable.

Valid tokens may depend on mw version.

Deprecated since version 8.0.

Parameters:: types (list[str])
Return type:: list[str]

version()[source]#

Return live project version number as a string.

Use mw_version to compare MediaWiki versions.

Return type:: str

watch(pages, unwatch=False)[source]#

Add or remove pages from watchlist.

See also

API:Watch

Parameters:

pages (BasePage | str | list[BasePage | str]) – A single page or a sequence of pages.
unwatch (bool) – If True, remove pages from watchlist; if False add them (default).

Returns:

True if API returned expected response; False otherwise

Raises:

KeyError – ‘watch’ isn’t in API response

Return type:

bool

Objects representing API interface to MediaWiki site extensions.

class pywikibot.site._extensions.EchoMixin[source]#

Bases: object

APISite mixin for Echo extension.

notifications(**kwargs)[source]#

Yield Notification objects from the Echo extension.

Keyword Arguments:: format (Optional[str]) – If specified, notifications will be returned formatted this way. Its value is either model, special or None. Default is special.

See also

API:Notifications for other keywords.

notifications_mark_read(**kwargs)[source]#

Mark selected notifications as read.

See also

API:echomarkread

Returns:: whether the action was successful
Return type:: bool

class pywikibot.site._extensions.GeoDataMixin[source]#

Bases: object

APISite mixin for GeoData extension.

loadcoordinfo(page)[source]#

Load [[mw:Extension:GeoData]] info.

Return type:: None

class pywikibot.site._extensions.GlobalUsageMixin[source]#

Bases: object

APISite mixin for Global Usage extension.

globalusage(page, total=None)[source]#

Iterate global image usage for a given FilePage.

Parameters:

page (FilePage) – the page to return global image usage for.
total – iterate no more than this number of pages in total.

Raises:

TypeError – input page is not a FilePage.
pywikibot.exceptions.SiteDefinitionError – Site could not be defined for a returned entry in API response.

class pywikibot.site._extensions.LinterMixin[source]#

Bases: object

APISite mixin for Linter extension.

linter_pages(lint_categories=None, total=None, namespaces=None, pageids=None, lint_from=None)[source]#

Return a generator to pages containing linter errors.

Parameters:

lint_categories (an iterable that returns values (str), or a pipe-separated string of values.) – categories of lint errors
total (int) – if not None, yielding this many items in total
namespaces (iterable of str or Namespace key, or a single instance of those types. May be a '|' separated list of namespace identifiers.) – only iterate pages in these namespaces
pageids (an iterable that returns pageids (str or int), or a comma- or pipe-separated string of pageids (e.g. '945097,1483753, 956608' or '945097|483753|956608')) – only include lint errors from the specified pageids
lint_from (str representing digit or integer) – Lint ID to start querying from

Returns:

pages with Linter errors.

Return type:

Iterable[Page]

class pywikibot.site._extensions.PageImagesMixin[source]#

Bases: object

APISite mixin for PageImages extension.

loadpageimage(page)[source]#

Load [[mw:Extension:PageImages]] info.

Parameters:: page (Page) – The page for which to obtain the image
Raises:: APIError – PageImages extension is not installed
Return type:: None

class pywikibot.site._extensions.ProofreadPageMixin[source]#

Bases: object

APISite mixin for ProofreadPage extension.

loadpageurls(page)[source]#

Load URLs from api and store in page attributes.

Load URLs to images for a given page in the “Page:” namespace. No effect for pages in other namespaces.

Added in version 8.6.

See also

API:imageforpage

Parameters:: page (pywikibot.page.BasePage)
Return type:: None

property proofread_index_ns#: Return Index namespace for the ProofreadPage extension.

property proofread_levels#: Return Quality Levels for the ProofreadPage extension.

property proofread_page_ns#: Return Page namespace for the ProofreadPage extension.

class pywikibot.site._extensions.TextExtractsMixin[source]#

Bases: object

APISite mixin for TextExtracts extension.

Added in version 7.1.

extract(page, *, chars=None, sentences=None, intro=True, plaintext=True)[source]#

Retrieve an extract of a page.

Parameters:

page (Page) – The Page object for which the extract is read
chars (int | None) – How many characters to return. Actual text returned might be slightly longer.
sentences (int | None) – How many sentences to return
intro (bool) – Return only content before the first section
plaintext (bool) – if True, return extracts as plain text instead of limited HTML

Return type:

str

See also

class pywikibot.site._extensions.ThanksMixin[source]#

Bases: object

APISite mixin for Thanks extension.

thank_revision(revid, source=None)[source]#

Corresponding method to the ‘action=thank’ API action.

Parameters:

revid (int) – Revision ID for the revision to be thanked.
source (str) – A source for the thanking operation.

Raises:

APIError – On thanking oneself or other API errors.

Returns:

The API response.

class pywikibot.site._extensions.UrlShortenerMixin[source]#

Bases: object

APISite mixin for UrlShortener extension.

create_short_link(url)[source]#

Return a shortened link.

Note that on Wikimedia wikis only metawiki supports this action, and this wiki can process links to all WM domains.

Parameters:: url (str) – The link to reduce, with propotol prefix.
Returns:: The reduced link, without protocol prefix.
Return type:: str

class pywikibot.site._extensions.WikibaseClientMixin[source]#

Bases: object

APISite mixin for WikibaseClient extension.

unconnected_pages(total=None)[source]#

Yield Page objects from Special:UnconnectedPages.

Warning

The retrieved pages may be connected in meantime.

Parameters:: total – number of pages to return

Objects representing API generators to MediaWiki site.

class pywikibot.site._generators.GeneratorsMixin[source]#

Bases: object

API generators mixin to MediaWiki site.

allcategories(start='!', prefix='', total=None, reverse=False, content=False)[source]#

Iterate categories used (which need not have a Category page).

Iterator yields Category objects. Note that, in practice, links that were found on pages that have been deleted may not have been removed from the database table, so this method can return false positives.

See also

API:Allcategories

Parameters:

start (str) – Start at this category title (category need not exist).
prefix (str) – Only yield categories starting with this string.
reverse (bool) – if True, iterate in reverse Unicode lexicographic order (default: iterate in forward order)
content (bool) – if True, load the current content of each iterated page (default False); note that this means the contents of the category description page, not the pages that are members of the category
total (int | None)

Return type:

Iterable[Category]

alldeletedrevisions(*, namespaces=None, reverse=False, content=False, total=None, **kwargs)[source]#

Yield all deleted revisions.

See also

API:Alldeletedrevisions

Warning

user keyword argument must be given together with start or end.

Parameters:

namespaces (NamespaceArgType) – Only iterate pages in these namespaces
reverse (bool) – Iterate oldest revisions first (default: newest)
content (bool) – If True, retrieve the content of each revision
total (int | None) – Number of revisions to retrieve

Keyword Arguments:

from (str) – Start listing at this title
to (str) – Stop listing at this title
prefix (str) – Search for all page titles that begin with this value
excludeuser (str) – Exclude revisions by this user
tag (str) – Only list revisions tagged with this tag
user (str) – List revisions by this user
start – Iterate revisions starting at this Timestamp
end – Iterate revisions ending at this Timestamp
prop (list[str]) – Which properties to get. Defaults are ids, timestamp, flags, user, and comment (if the bot has the right to view).

Return type:

Generator[dict[str, Any], None, None]

allimages(start='!', prefix='', minsize=None, maxsize=None, reverse=False, sha1=None, sha1base36=None, total=None, content=False)[source]#

Iterate all images, ordered by image title.

Yields FilePages, but these pages need not exist on the wiki.

See also

API:Allimages

Parameters:

start (str) – start at this title (name need not exist)
prefix (str) – only iterate titles starting with this substring
minsize (int | None) – only iterate images of at least this many bytes
maxsize (int | None) – only iterate images of no more than this many bytes
reverse (bool) – if True, iterate in reverse lexicographic order
sha1 (str | None) – only iterate image (it is theoretically possible there could be more than one) with this sha1 hash
sha1base36 (str | None) – same as sha1 but in base 36
content (bool) – if True, load the current content of each iterated page (default False); note that this means the content of the image description page, not the image itself
total (int | None)

Return type:

Iterable[FilePage]

alllinks(start='', prefix='', namespace=0, unique=False, fromids=False, total=None)[source]#

Iterate all links to pages (which need not exist) in one namespace.

Note

In practice, links that were found on pages that have been deleted may not have been removed from the links table, so this method can return false positives.

Caution

unique parameter is no longer supported by MediaWiki 1.43 or higher. Pywikibot uses tools.itertools.filter_unique() in that case which might be memory intensive. Use it with care.

Important

Using namespace option different from 0 needs a lot of time on Wikidata site. You have to increase the read timeout part of socket_timeout in Http Settings in your user-config.py file. Or increase it partially within your code like:

from pywikibot import config
save_timeout = config.socket_timeout  # save the timeout config
config.socket_timeout = save_timeout[0], 60
... # your code here
config.socket_timeout = save_timeout  # restore timeout config

The minimum read timeout value should be 60 seconds in that case.

See also

API:Alllinks
pagebacklinks()
pagelinks()

Parameters:

start (str) – Start at this title (page need not exist).
prefix (str) – Only yield pages starting with this string.
namespace (SingleNamespaceType) – Iterate pages from this (single) namespace
unique (bool) – If True, only iterate each link title once (default: False)
fromids (bool) – if True, include the pageid of the page containing each link (default: False) as the ‘_fromid’ attribute of the Page; cannot be combined with unique
total (int | None)

Raises:

KeyError – the namespace identifier was not resolved
TypeError – the namespace identifier has an inappropriate type such as bool, or an iterable with more than one namespace

Return type:

Generator[Page, None, None]

allpages(start='!', prefix='', namespace=0, filterredir=None, filterlanglinks=None, minsize=None, maxsize=None, protect_type=None, protect_level=None, reverse=False, total=None, content=False)[source]#

Iterate pages in a single namespace.

See also

API:Allpages

Parameters:

start (str) – Start at this title (page need not exist).
prefix (str) – Only yield pages starting with this string.
namespace (SingleNamespaceType) – Iterate pages from this (single) namespace.
filterredir (bool | None) – if True, only yield redirects; if False (and not None), only yield non-redirects (default: yield both).
filterlanglinks (bool | None) – if True, only yield pages with language links; if False (and not None), only yield pages without language links (default: yield both).
minsize (int | None) – if present, only yield pages at least this many bytes in size.
maxsize (int | None) – if present, only yield pages at most this many bytes in size.
protect_type (str | None) – only yield pages that have a protection of the specified type.
protect_level (str | None) – only yield pages that have protection at this level; can only be used if protect_type is specified.
reverse (bool) – if True, iterate in reverse Unicode lexicographic order (default: iterate in forward order).
content (bool) – if True, load the current content of each iterated page (default False).
total (int | None)

Raises:

KeyError – the namespace identifier was not resolved
TypeError – the namespace identifier has an inappropriate type such as bool, or an iterable with more than one namespace or filterredir parameter has an invalid type.

Return type:

Iterable[Page]

allusers(start='!', prefix='', group=None, total=None)[source]#

Iterate registered users, ordered by username.

Iterated values are dicts containing ‘name’, ‘editcount’, ‘registration’, and (sometimes) ‘groups’ keys. ‘groups’ will be present only if the user is a member of at least 1 group, and will be a list of str; all the other values are str and should always be present.

See also

API:Allusers

Parameters:

start (str) – start at this username (name need not exist)
prefix (str) – only iterate usernames starting with this substring
group (str | None) – only iterate users that are members of this group
total (int | None)

Return type:

Iterable[dict[str, str | list[str]]]

ancientpages(total=None)[source]#

Yield Pages, datestamps from Special:Ancientpages.

Parameters:: total (int | None) – number of pages to return
Return type:: Generator[tuple[Page, Timestamp], None, None]

blocks(starttime=None, endtime=None, reverse=False, blockids=None, users=None, iprange=None, total=None)[source]#

Iterate all current blocks, in order of creation.

The iterator yields dicts containing keys corresponding to the block properties.

See also

API:Blocks

Note

logevents only logs user blocks, while this method iterates all blocks including IP ranges.

Warning

iprange parameter cannot be used together with users.

Parameters:

starttime (Timestamp | None) – start iterating at this Timestamp
endtime (Timestamp | None) – stop iterating at this Timestamp
reverse (bool) – if True, iterate oldest blocks first (default: newest)
blockids (int | str | Iterable[int | str] | None) – only iterate blocks with these id numbers. Numbers must be separated by ‘|’ if given by a str.
users (str | Iterable[str] | None) – only iterate blocks affecting these usernames or IPs
iprange (str | None) – a single IP or an IP range. Ranges broader than IPv4/16 or IPv6/19 are not accepted.
total (int | None) – total amount of block entries

Return type:

Iterable[dict[str, Any]]

botusers(total=None)[source]#

Iterate bot users.

Iterated values are dicts containing ‘name’, ‘userid’, ‘editcount’, ‘registration’, and ‘groups’ keys. ‘groups’ will be present only if the user is a member of at least 1 group, and will be a list of str; all the other values are str and should always be present.

Parameters:: total (int | None)
Return type:: Generator[dict[str, Any], None, None]

broken_redirects(total=None)[source]#

Yield Pages with broken redirects from Special:BrokenRedirects.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

categorymembers(category, *, namespaces=None, sortby=None, reverse=False, starttime=None, endtime=None, total=None, startprefix=None, endprefix=None, content=False, member_type=None)[source]#

Iterate members of specified category.

You should not use this method directly; instead use one of the following:

Changed in version 4.0: parameters except category are keyword arguments only.

Changed in version 8.0: raises TypeError instead of Error if no Category is specified

See also

API:Categorymembers

Parameters:

category (Category) – The Category to iterate.
namespaces (NamespaceArgType) – If present, only return category members from these namespaces. To yield subcategories or files, use parameter member_type instead.
sortby (str | None) – determines the order in which results are generated, valid values are “sortkey” (default, results ordered by category sort key) or “timestamp” (results ordered by time page was added to the category)
reverse (bool) – if True, generate results in reverse order (default False)
starttime (Timestamp | None) – if provided, only generate pages added after this time; not valid unless sortby=”timestamp”
endtime (Timestamp | None) – if provided, only generate pages added before this time; not valid unless sortby=”timestamp”
startprefix (str | None) – if provided, only generate pages >= this title lexically; not valid if sortby=”timestamp”
endprefix (str | None) – if provided, only generate pages < this title lexically; not valid if sortby=”timestamp”
content (bool) – if True, load the current content of each iterated page (default False)
member_type (str | Iterable[str] | None) – member type; values must be page, subcat, file. If member_type includes page and is used in conjunction with sortby=”timestamp”, the API may limit results to only pages in the first 50 namespaces.
total (int | None)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool
TypeError – no Category is specified
ValueError – invalid values given

Return type:

Iterable[Page]

deadendpages(total=None)[source]#

Yield Page objects retrieved from Special:Deadendpages.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

deletedrevs(titles=None, start=None, end=None, reverse=False, content=False, total=None, **kwargs)[source]#

Iterate deleted revisions.

Each value returned by the iterator will be a dict containing the ‘title’ and ‘ns’ keys for a particular Page and a ‘revisions’ key whose value is a list of revisions in the same format as recentchanges plus a ‘content’ element with key ‘*’ if requested when ‘content’ parameter is set. For older wikis a ‘token’ key is also given with the content request.

See also

API:Deletedrevisions

Parameters:

titles (str | Page | Iterable[str] | Iterable[Page] | None) – The page titles to check for deleted revisions
reverse (bool)
content (bool)
total (int | None)

Keyword Arguments:

revids – Get revisions by their ID

Return type:

Generator[dict[str, Any], None, None]

Note

either titles or revids must be set but not both

Parameters:

start – Iterate revisions starting at this Timestamp
end – Iterate revisions ending at this Timestamp
reverse (bool) – Iterate oldest revisions first (default: newest)
content (bool) – If True, retrieve the content of each revision
total (int | None) – number of revisions to retrieve
titles (str | Page | Iterable[str] | Iterable[Page] | None)

Keyword Arguments:

user – List revisions by this user
excludeuser – Exclude revisions by this user
tag – Only list revision tagged with this tag
prop – Which properties to get. Defaults are ids, user, comment, flags and timestamp

Return type:

Generator[dict[str, Any], None, None]

double_redirects(total=None)[source]#

Yield Pages with double redirects from Special:DoubleRedirects.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

exturlusage(url=None, protocol=None, namespaces=None, total=None, content=False)[source]#

Iterate Pages that contain links to the given URL.

See also

API:Exturlusage

Parameters:

url (str | None) – The URL to search for (with or without the protocol prefix); this may include a ‘*’ as a wildcard, only at the start of the hostname
namespaces (list[int] | None) – list of namespace numbers to fetch contribs from
total (int | None) – Maximum number of pages to retrieve in total
protocol (str | None) – Protocol to search for, likely http or https, http by default. Full list shown on Special:LinkSearch wikipage
content (bool)

Return type:

Iterable[Page]

filearchive(start=None, end=None, reverse=False, total=None, **kwargs)[source]#

Iterate archived files.

Yields dict of file archive information.

See also

API:filearchive

Parameters:

start (str | None) – start at this title (name need not exist)
end (str | None) – end at this title (name need not exist)
reverse (bool) – if True, iterate in reverse lexicographic order
total (int | None) – maximum number of pages to retrieve in total

Keyword Arguments:

prefix – only iterate titles starting with this substring
sha1 – only iterate image with this sha1 hash
sha1base36 – same as sha1 but in base 36
prop – Image information to get. Default is timestamp

Return type:

Iterable[dict[str, Any]]

imageusage(image, *, namespaces=None, filterredir=None, total=None, content=False)[source]#

Iterate Pages that contain links to the given FilePage.

See also

API:Imageusage

Changed in version 7.2: all parameters except image are keyword only.

Parameters:

image (FilePage) – the image to search for (FilePage need not exist on the wiki)
namespaces (NamespaceArgType) – If present, only iterate pages in these namespaces
filterredir (bool | None) – if True, only yield redirects; if False (and not None), only yield non-redirects (default: yield both)
total (int | None) – iterate no more than this number of pages in total
content (bool) – if True, load the current content of each iterated page (default False)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Iterable[Page]

load_pages_from_pageids(pageids)[source]#

Return a page generator from pageids.

Pages are iterated in the same order than in the underlying pageids.

Pageids are filtered and only one page is returned in case of duplicate pageids.

Parameters:: pageids (str | Iterable[int | str]) – an iterable that returns pageids (str or int), or a comma- or pipe-separated string of pageids (e.g. ‘945097,1483753, 956608’ or ‘945097|483753|956608’)
Return type:: Generator[Page, None, None]

loadrevisions(page, *, content=False, section=None, **kwargs)[source]#

Retrieve revision information and store it in page object.

By default, retrieves the last (current) revision of the page, unless any of the optional parameters revids, startid, endid, starttime, endtime, rvdir, user, excludeuser, or total are specified. Unless noted below, all parameters not specified default to False.

If rvdir is False or not specified, startid must be greater than endid if both are specified; likewise, starttime must be greater than endtime. If rvdir is True, these relationships are reversed.

See also

API:Revisions

Parameters:

page (Page) – retrieve revisions of this Page and hold the data.
content (bool) – if True, retrieve the wiki-text of each revision; otherwise, only retrieve the revision metadata (default)
section (int | None) – if specified, retrieve only this section of the text (content must be True); section must be given by number (top of the article is section 0), not name

Keyword Arguments:

revids – retrieve only the specified revision ids (raise Exception if any of revids does not correspond to page)
startid – retrieve revisions starting with this revid
endid – stop upon retrieving this revid
starttime – retrieve revisions starting at this Timestamp
endtime – stop upon reaching this Timestamp
rvdir – if false, retrieve newest revisions first (default); if true, retrieve oldest first
user – retrieve only revisions authored by this user
excludeuser – retrieve all revisions not authored by this user
total – number of revisions to retrieve

Raises:

ValueError – invalid startid/endid or starttime/endtime values
pywikibot.exceptions.Error – revids belonging to a different page

Return type:

None

logevents(logtype=None, user=None, page=None, namespace=None, start=None, end=None, reverse=False, tag=None, total=None)[source]#

Iterate all log entries.

See also

API:Logevents

Note

logevents with logtype='block' only logs user blocks whereas site.blocks iterates all blocks including IP ranges.

Parameters:

logtype (str | None) – only iterate entries of this type (see mediawiki api documentation for available types)
user (str | None) – only iterate entries that match this user name
page (str | Page | None) – only iterate entries affecting this page
namespace (NamespaceArgType) – namespace(s) to retrieve logevents from
start (str | Timestamp | None)
end (str | Timestamp | None)
reverse (bool)
tag (str | None)
total (int | None)

Return type:

Iterable[pywikibot.logentries.LogEntry]

Note

due to an API limitation, if namespace param contains multiple namespaces, log entries from all namespaces will be fetched from the API and will be filtered later during iteration.

Parameters:

start (str | Timestamp | None) – only iterate entries from and after this Timestamp
end (str | Timestamp | None) – only iterate entries up to and through this Timestamp
reverse (bool) – if True, iterate oldest entries first (default: newest)
tag (str | None) – only iterate entries tagged with this tag
total (int | None) – maximum number of events to iterate
logtype (str | None)
user (str | None)
page (str | Page | None)
namespace (NamespaceArgType)

Raises:

KeyError – the namespace identifier was not resolved
TypeError – the namespace identifier has an inappropriate type such as bool, or an iterable with more than one namespace

Return type:

Iterable[pywikibot.logentries.LogEntry]

lonelypages(total=None)[source]#

Yield Pages retrieved from Special:Lonelypages.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

longpages(total=None)[source]#

Yield Pages and lengths from Special:Longpages.

Yields a tuple of Page object, length(int).

Parameters:: total (int | None) – number of pages to return
Return type:: Generator[tuple[Page, int], None, None]

newpages(user=None, returndict=False, start=None, end=None, reverse=False, bot=False, redirect=False, excludeuser=None, patrolled=None, namespaces=None, total=None)[source]#

Yield new articles (as Page objects) from recent changes.

Starts with the newest article and fetches the number of articles specified in the first argument.

The objects yielded are dependent on parameter returndict. When true, it yields a tuple composed of a Page object and a dict of attributes. When false, it yields a tuple composed of the Page object, timestamp (str), length (int), an empty string, username or IP address (str), comment (str).

Parameters:

namespaces (NamespaceArgType) – only iterate pages in these namespaces
returndict (bool)
reverse (bool)
bot (bool)
redirect (bool)
total (int | None)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Generator[tuple[Page, dict[str, Any]], None, None] | Generator[tuple[Page, str, int, str, str, str], None, None]

page_embeddedin(page, *, filter_redirects=None, namespaces=None, total=None, content=False)[source]#

Iterate all pages that embedded the given page as a template.

See also

API:Embeddedin

Parameters:

page (Page) – The Page to get inclusions for.
filter_redirects – If True, only return redirects that embed the given page. If False, only return non-redirect links. If None, return both (no filtering).
namespaces (NamespaceArgType) – If present, only return links from the namespaces in this list.
content (bool) – if True, load the current content of each iterated page (default False)
total (int | None)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Iterable[Page]

page_extlinks(page, *, total=None)[source]#

Yield all external links on page, yielding URL strings.

See also

API:Extlinks

Parameters:

page (Page)
total (int | None)

Return type:

Generator[str, None, None]

page_redirects(page, *, filter_fragments=None, namespaces=None, total=None, content=False)[source]#

Iterale all redirects to the given page.

See also

API:Redirects

Added in version 7.0.

Parameters:

page (Page) – The Page to get redirects for.
filter_fragments (bool | None) – If True, only return redirects with fragments. If False, only return redirects without fragments. If None, return both (no filtering).
namespaces (NamespaceArgType) – Only return redirects from the namespaces
total (int | None) – maximum number of redirects to retrieve in total
content (bool) – load the current content of each redirect

Return type:

Iterable[Page]

pagebacklinks(page, *, follow_redirects=False, filter_redirects=None, namespaces=None, total=None, content=False)[source]#

Iterate all pages that link to the given page.

See also

API:Backlinks

Parameters:

page (Page) – The Page to get links to.
follow_redirects (bool) – Also return links to redirects pointing to the given page.
filter_redirects – If True, only return redirects to the given page. If False, only return non-redirect links. If None, return both (no filtering).
namespaces (NamespaceArgType) – If present, only return links from the namespaces in this list.
total (int | None) – Maximum number of pages to retrieve in total.
content (bool) – if True, load the current content of each iterated page (default False)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Iterable[Page]

pagecategories(page, *, with_sort_key=False, total=None, content=False)[source]#

Iterate categories to which page belongs.

Added in version 9.6: the with_sort_key parameter.

See also

Note

This method also yields categories which are transcluded.

Parameters:

with_sort_key (bool) – if True, include the sort key in each Category
content (bool) – if True, load the current content of each iterated page default False); note that this means the contents of the category description page, not the pages contained in the category
page (Page)
total (int | None)

Return type:

Iterable[Page]

pageimages(page, *, total=None, content=False)[source]#

Iterate images used (not just linked) on the page.

See also

API:Images

Parameters:

content (bool) – if True, load the current content of each iterated page (default False); note that this means the content of the image description page, not the image itself
page (Page)
total (int | None)

Return type:

Iterable[Page]

pagelanglinks(page, *, total=None, include_obsolete=False, include_empty_titles=False)[source]#

Yield all interlanguage links on page, yielding Link objects.

Changed in version 6.2:: include_empty_titles parameter was added.

See also

API:Langlinks

Parameters:

include_obsolete (bool) – if true, yield even Link objects whose site is obsolete
include_empty_titles (bool) – if true, yield even Link objects whose title is empty but redirects to a site like [[en:]]
page (Page)
total (int | None)

Return type:

Generator[Link, None, None]

pagelinks(page, *, namespaces=None, follow_redirects=False, total=None, content=False)[source]#

Yield internal wikilinks contained (or transcluded) on page.

See also

Parameters:

namespaces (NamespaceArgType) – Only iterate pages in these namespaces (default: all)
follow_redirects (bool) – if True, yields the target of any redirects, rather than the redirect page
total (int | None) – iterate no more than this number of pages in total
content (bool) – if True, load the current content of each iterated page
page (pywikibot.page.BasePage)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Generator[Page, None, None]

pagereferences(page, *, follow_redirects=False, filter_redirects=None, with_template_inclusion=True, only_template_inclusion=False, namespaces=None, total=None, content=False)[source]#

Convenience method combining pagebacklinks and page_embeddedin.

Parameters:

namespaces (NamespaceArgType) – If present, only return links from the namespaces in this list.
follow_redirects (bool)
filter_redirects (bool | None)
with_template_inclusion (bool)
only_template_inclusion (bool)
total (int | None)
content (bool)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Iterable[Page]

pages_with_property(propname, *, total=None)[source]#

Iterate Page objects from Special:PagesWithProp.

See also

API:Pageswithprop

Parameters:

propname (str) – must be a valid property.
total (int | None) – number of pages to return

Returns:

return a generator of Page objects

Return type:

iterator

pagetemplates(page, *, content=False, namespaces=None, total=None)[source]#

Iterate pages transcluded (not just linked) on the page.

See also

Parameters:

content (bool) – if True, load the current content of each iterated page (default False)
namespaces (NamespaceArgType) – Only iterate pages in these namespaces
total (int | None) – maximum number of pages to retrieve in total
page (Page)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool
UnsupportedPageError – a Page object is not supported due to namespace restriction

Return type:

Iterable[Page]

patrol(rcid=None, revid=None, revision=None)[source]#

Return a generator of patrolled pages.

See also

API:Patrol

Pages to be patrolled are identified by rcid, revid or revision. At least one of the parameters is mandatory. See https://www.mediawiki.org/wiki/API:Patrol.

Parameters:

rcid (int | str | Iterable[int] | Iterable[str] | None) – an int/string/iterable/iterator providing rcid of pages to be patrolled.
revid (int | str | Iterable[int] | Iterable[str] | None) – an int/string/iterable/iterator providing revid of pages to be patrolled.
revision (pywikibot.page.Revision | Iterable[pywikibot.page.Revision] | None) – an Revision/iterable/iterator providing Revision object of pages to be patrolled.

Return type:

Generator[dict[str, int | str], None, None]

preloadpages(pagelist, *, groupsize=None, templates=False, langlinks=False, pageprops=False, categories=False, content=True, quiet=True)[source]#

Return a generator to a list of preloaded pages.

Pages are iterated in the same order than in the underlying pagelist. In case of duplicates in a groupsize batch, return the first entry.

Changed in version 7.6: content parameter was added.

Changed in version 7.7: categories parameter was added.

Changed in version 8.1: groupsize is maxlimit by default. quiet parameter was added. No longer show the “Retrieving pages from site” message by default.

Parameters:

pagelist (Iterable[Page]) – an iterable that returns Page objects
groupsize (int | None) – how many Pages to query at a time. If None (default), maxlimit is used.
templates (bool) – preload pages (typically templates) transcluded in the provided pages
langlinks (bool) – preload all language links from the provided pages to other languages
pageprops (bool) – preload various properties defined in page content
categories (bool) – preload page categories
content (bool) – preload page content
quiet (bool) – If True (default), do not show the “Retrieving pages” message

Return type:

Generator[Page, None, None]

protectedpages(namespace=0, protect_type='edit', level=False, total=None, type='[deprecated name of protect_type]')[source]#

Return protected pages depending on protection level and type.

For protection types which aren’t ‘create’ it uses APISite.allpages, while it uses for ‘create’ the ‘query+protectedtitles’ module.

Changed in version 9.0: type parameter was renamed to protect_type.

See also

API:Protectedtitles

Parameters:

namespace (NamespaceArgType) – The searched namespace.
protect_type (str) – The protection type to search for (default ‘edit’).
level (str | bool) – The protection level (like ‘autoconfirmed’). If False it shows all protection levels.
total (int | None)

Returns:

The pages which are protected.

querypage(special_page, total=None)[source]#

Iterate Page objects retrieved from Special:{special_page}.

Generic function for all special pages supported by the site MW API.

See also

API:Querypage

Changed in version 9.0: Raises ValueError instead of AssertionError if special_page is invalid.

Parameters:

special_page (str) – Special page to query
total (int | None) – number of pages to return

Raises:

ValueError – special_page is not supported in SpecialPages.

Return type:

Iterable[Page]

randompages(total=None, namespaces=None, redirects=False, content=False)[source]#

Iterate a number of random pages.

Pages are listed in a fixed sequence, only the starting point is random.

See also

API:Random

Changed in version 9.0: Raises TypeError instead of AssertionError if redirects is invalid.

Parameters:

total (int | None) – the maximum number of pages to iterate
namespaces (NamespaceArgType) – only iterate pages in these namespaces.
redirects (bool | None) – if True, include only redirect pages in results, False does not include redirects and None include both types (default: False).
content (bool) – if True, load the current content of each iterated page (default False).

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool
TypeError – unsupported redirects parameter

Return type:

Iterable[Page]

recentchanges(*, start=None, end=None, reverse=False, namespaces=None, changetype=None, minor=None, bot=None, anon=None, redirect=None, patrolled=None, top_only=False, total=None, user=None, excludeuser=None, tag=None)[source]#

Iterate recent changes.

See also

API:RecentChanges

Parameters:

start (Timestamp) – Timestamp to start listing from
end (Timestamp) – Timestamp to end listing at
reverse (bool) – if True, start with oldest changes (default: newest)
namespaces (NamespaceArgType) – only iterate pages in these namespaces
changetype (str | None) – only iterate changes of this type (“edit” for edits to existing pages, “new” for new pages, “log” for log entries)
minor (bool | None) – if True, only list minor edits; if False, only list non-minor edits; if None, list all
bot (bool | None) – if True, only list bot edits; if False, only list non-bot edits; if None, list all
anon (bool | None) – if True, only list anon edits; if False, only list non-anon edits; if None, list all
redirect (bool | None) – if True, only list edits to redirect pages; if False, only list edits to non-redirect pages; if None, list all
patrolled (bool | None) – if True, only list patrolled edits; if False, only list non-patrolled edits; if None, list all
top_only (bool) – if True, only list changes that are the latest revision (default False)
user (str | list[str] | None) – if not None, only list edits by this user or users
excludeuser (str | list[str] | None) – if not None, exclude edits by this user or users
tag (str | None) – a recent changes tag
total (int | None)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Iterable[dict[str, Any]]

redirectpages(total=None)[source]#

Yield redirect pages from Special:ListRedirects.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

search(searchstring, *, namespaces=None, where=None, total=None, content=False, sort='relevance')[source]#

Iterate Pages that contain the searchstring.

Note that this may include non-existing Pages if the wiki’s database table contains outdated entries.

Changed in version 7.0: Default of where parameter has been changed from ‘text’ to None. The behaviour depends on the installed search engine which is ‘text’ on CirrusSearch’. raises APIError instead of Error if searchstring is not set or what parameter is wrong.

Changed in version 10.0: The sort parameter was added.

See also

API:Search

Parameters:

searchstring (str) – the text to search for
namespaces (NamespaceArgType) – search only in these namespaces (defaults to all)
where (Literal['text', 'title', 'nearmatch'] | None) – Where to search; value must be one of the given literals or None (many wikis do not support all search types)
total (int | None) – limit result to this number of pages
content (bool) – if True, load the current content of each iterated page (default False)
sort (Literal['create_timestamp_asc', 'create_timestamp_desc', 'incoming_links_asc', 'incoming_links_desc', 'just_match', 'last_edit_asc', 'last_edit_desc', 'none', 'random', 'relevance', 'user_random'] | None) – Set the sort order of returned results. If None is given, ‘none’ is used. Default is sort by relevance.

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool
APIError – The “gsrsearch” parameter must be set: searchstring parameter is not set
APIError – Unrecognized value for parameter “gsrwhat”: wrong where parameter is given

Return type:

Iterable[Page]

shortpages(total=None)[source]#

Yield Pages and lengths from Special:Shortpages.

Yields a tuple of Page object, length(int).

Parameters:: total (int | None) – number of pages to return
Return type:: Generator[tuple[Page, int], None, None]

uncategorizedcategories(total=None)[source]#

Yield Categories from Special:Uncategorizedcategories.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

uncategorizedfiles(total=None)#

Yield FilePages from Special:Uncategorizedimages.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

uncategorizedimages(total=None)[source]#

Yield FilePages from Special:Uncategorizedimages.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

uncategorizedpages(total=None)[source]#

Yield Pages from Special:Uncategorizedpages.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

uncategorizedtemplates(total=None)[source]#

Yield Pages from Special:Uncategorizedtemplates.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

unusedcategories(total=None)[source]#

Yield Category objects from Special:Unusedcategories.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

unusedfiles(total=None)[source]#

Yield FilePage objects from Special:Unusedimages.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

unwatchedpages(total=None)[source]#

Yield Pages from Special:Unwatchedpages (requires Admin privileges).

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

usercontribs(user=None, userprefix=None, start=None, end=None, reverse=False, namespaces=None, minor=None, total=None, top_only=False)[source]#

Iterate contributions by a particular user.

Iterated values are in the same format as recentchanges.

See also

Parameters:

user (str | None) – Iterate contributions by this user (name or IP)
userprefix (str | None) – Iterate contributions by all users whose names or IPs start with this substring
start – Iterate contributions starting at this Timestamp
end – Iterate contributions ending at this Timestamp
reverse (bool) – Iterate oldest contributions first (default: newest)
namespaces (NamespaceArgType) – only iterate pages in these namespaces
minor (bool | None) – if True, iterate only minor edits; if False and not None, iterate only non-minor edits (default: iterate both)
total (int | None) – limit result to this number of pages
top_only (bool) – if True, iterate only edits which are the latest revision (default: False)

Raises:

pywikibot.exceptions.Error – either user or userprefix must be non-empty
KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Iterable[dict[str, Any]]

users(usernames)[source]#

Iterate info about a list of users by name or IP.

See also

API:Users

Parameters:: usernames (Iterable[str]) – a list of user names
Return type:: Iterable[dict[str, Any]]

wantedcategories(total=None)[source]#

Yield Pages from Special:Wantedcategories.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

wantedfiles(total=None)[source]#

Yield Pages from Special:Wantedfiles.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

wantedpages(total=None)[source]#

Yield Pages from Special:Wantedpages.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

wantedtemplates(total=None)[source]#

Yield Pages from Special:Wantedtemplates.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

watched_pages(force=False, total=None, *, with_talkpage=True)[source]#

Return watchlist.

Note

watched_pages is a restartable generator. See tools.collections.GeneratorWrapper for its usage.

See also

API:Watchlistraw

Added in version 8.1: the with_talkpage parameter.

Parameters:

force (bool) – Reload watchlist
total (int | None) – if not None, limit the generator to yielding this many items in total
with_talkpage (bool) – if false, ignore talk pages and special pages

Returns:

generator of pages in watchlist

Return type:

Iterable[Page]

watchlist_revs(start=None, end=None, reverse=False, namespaces=None, minor=None, bot=None, anon=None, total=None)[source]#

Iterate revisions to pages on the bot user’s watchlist.

Iterated values will be in same format as recentchanges.

See also

API:Watchlist

Parameters:

start – Iterate revisions starting at this Timestamp
end – Iterate revisions ending at this Timestamp
reverse (bool) – Iterate oldest revisions first (default: newest)
namespaces (NamespaceArgType) – only iterate pages in these namespaces
minor (bool | None) – if True, only list minor edits; if False (and not None), only list non-minor edits
bot (bool | None) – if True, only list bot edits; if False (and not None), only list non-bot edits
anon (bool | None) – if True, only list anon edits; if False (and not None), only list non-anon edits
total (int | None)

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

Iterable[dict[str, Any]]

withoutinterwiki(total=None)[source]#

Yield Pages without language links from Special:Withoutinterwiki.

Parameters:: total (int | None) – number of pages to return
Return type:: Iterable[Page]

`DataSite` — API Interface for Wikibase#

Objects representing API interface to Wikibase site.

class pywikibot.site._datasite.DataSite(*args, **kwargs)[source]#

Bases: APISite

Wikibase data capable site.

addClaim(entity, claim, bot=True, summary=None, tags=None)[source]#

Add a claim.

Changed in version 9.4: tags parameter was added

Parameters:

entity (pywikibot.page.WikibaseEntity) – Entity to modify
claim (pywikibot.page.Claim) – Claim to be added
bot (bool) – Whether to mark the edit as a bot edit
summary (str | None) – Edit summary
tags (str | None) – Change tags to apply to the revision

Return type:

None

add_form(lexeme, form, *, bot=True, baserevid=None)[source]#

Add a form.

Parameters:

lexeme (LexemePage) – Lexeme to modify
form (LexemeForm) – Form to be added
bot (bool)

Keyword Arguments:

bot – Whether to mark the edit as a bot edit
baserevid – Base revision id override, used to detect conflicts.

Return type:

dict

changeClaimTarget(claim, snaktype='value', bot=True, summary=None, tags=None)[source]#

Set the claim target to the value of the provided claim target.

Changed in version 9.4: tags parameter was added

Parameters:

claim (Claim) – The source of the claim target value
snaktype (str) – An optional snaktype (‘value’, ‘novalue’ or ‘somevalue’). Default: ‘value’
bot (bool) – Whether to mark the edit as a bot edit
summary (str | None) – Edit summary
tags (str | None) – Change tags to apply to the revision

property concept_base_uri#

Return the base uri for concepts/entities.

Returns:: concept base uri
Return type:: str

editEntity(entity, data, bot=True, **kwargs)[source]#

Edit entity.

Note

This method is unable to create entities other than item if dict with API parameters was passed to entity parameter.

Changed in version 9.4: tags keyword argument was added

Parameters:

entity (pywikibot.page.WikibaseEntity | dict) – Page to edit, or dict with API parameters to use for entity identification.
data (dict) – data updates
bot (bool) – Whether to mark the edit as a bot edit.

Keyword Arguments:

baserevid (int) – The numeric identifier for the revision to base the modification on. This is used for detecting conflicts during save.
clear (bool) – If set, the complete entity is emptied before proceeding. The entity will not be saved before it is filled with the data, possibly with parts excluded.
summary (str) – Summary for the edit. Will be prepended by an automatically generated comment. The length limit of the autocomment together with the summary is 260 characters. Be aware that everything above that limit will be cut off.
tags (Iterable[str] | str) – Change tags to apply to the revision.

Returns:

New entity data

Return type:

dict

editQualifier(claim, qualifier, new=False, bot=True, summary=None, tags=None)[source]#

Create/Edit a qualifier.

Changed in version 9.4: tags parameter was added

Changed in version 10.0: deprecated baserevid parameter was removed

Parameters:

claim (Claim) – A Claim object to add the qualifier to
qualifier (Claim) – A Claim object to be used as a qualifier
new (bool) – Whether to create a new one if the qualifier already exists
bot (bool) – Whether to mark the edit as a bot edit
summary (str | None) – Edit summary
tags (str | None) – Change tags to apply to the revision

Raises:

ValueError – The claim cannot have a qualifier.

editSource(claim, source, new=False, bot=True, summary=None, tags=None)[source]#

Create/Edit a source.

Changed in version 9.4: tags parameter was added

Changed in version 10.0: deprecated baserevid parameter was removed

Parameters:

claim (Claim) – A Claim object to add the source to.
source (Claim) – A Claim object to be used as a source.
new (bool) – Whether to create a new one if the “source” already exists.
bot (bool) – Whether to mark the edit as a bot edit.
summary (str | None) – Edit summary.
tags (str | None) – Change tags to apply to the revision.

Raises:

ValueError – The claim cannot have a source.

edit_form_elements(form, data, *, bot=True, baserevid=None)[source]#

Edit lexeme form elements.

Parameters:

form (LexemeForm) – Form
data (dict) – data updates
bot (bool)

Keyword Arguments:

bot – Whether to mark the edit as a bot edit
baserevid – Base revision id override, used to detect conflicts.

Returns:

New form data

Return type:

dict

geo_shape_repository()[source]#: Return Site object for the geo-shapes repository e.g. commons.

getPropertyType(prop)[source]#: Obtain the type of a property.

Deprecated since version 9.5: Use get_property_type() instead.

get_entity_for_entity_id(entity_id)[source]#

Return a new instance for given entity id.

Raises:: pywikibot.exceptions.NoWikibaseEntityError – there is no entity with the id
Returns:: a WikibaseEntity subclass
Return type:: WikibaseEntity

get_namespace_for_entity_type(entity_type)[source]#

Return namespace for given entity type.

Returns:: corresponding namespace
Return type:: Namespace

get_property_type(prop)[source]#

Obtain the type of a property.

This is used specifically because we can cache the value for a much longer time (near infinite).

Added in version 9.5.

Raises:: NoWikibaseEntityError – prop does not exist
Parameters:: prop (Property)
Return type:: str

get_repo_for_entity_type(entity_type)[source]#

Get the data repository for the entity type.

When no foreign repository is defined for the entity type, the method returns this repository itself even if it does not support that entity type either.

Added in version 8.0.

Raises:: ValueError – when invalid entity type was provided
Parameters:: entity_type (str)
Return type:: DataSite

property item_namespace#

Return namespace for items.

Returns:: item namespace
Return type:: Namespace

linkTitles(page1, page2, bot=True)[source]#

Link two pages together.

Changed in version 9.4: tags parameter was added

Parameters:

page1 (Page) – First page to link
page2 (Page) – Second page to link
bot (bool) – Whether to mark the edit as a bot edit

Returns:

dict API output

Return type:

dict

loadcontent(identification, *props)[source]#

Fetch the current content of a Wikibase item.

This is called loadcontent since wbgetentities does not support fetching old revisions. Eventually this will get replaced by an actual loadrevisions.

Parameters:

identification (dict) – Parameters used to identify the page(s)
props – the optional properties to fetch.

mergeItems(from_item, to_item, ignore_conflicts=None, summary=None, bot=True, tags=None)[source]#

Merge two items together.

Changed in version 9.4: tags parameter was added

Parameters:

from_item (ItemPage) – Item to merge from
to_item (ItemPage) – Item to merge into
ignore_conflicts (list[str] | None) – Which type of conflicts (‘description’, ‘sitelink’, and ‘statement’) should be ignored
summary (str | None) – Edit summary
bot (bool) – Whether to mark the edit as a bot edit
tags (str | None) – Change tags to apply to the revision

Returns:

dict API output

Return type:

dict

mergeLexemes(from_lexeme, to_lexeme, summary=None, *, bot=True)[source]#

Merge two lexemes together.

Parameters:

from_lexeme (LexemePage) – Lexeme to merge from
to_lexeme (LexemePage) – Lexeme to merge into
summary (str) – Edit summary
bot (bool)

Keyword Arguments:

bot – Whether to mark the edit as a bot edit

Returns:

dict API output

Return type:

dict

parsevalue(datatype, values, options=None, language=None, validate=False)[source]#

Send data values to the wikibase parser for interpretation.

Added in version 7.5.

See also

wbparsevalue API

Parameters:

datatype (str) – datatype of the values being parsed. Refer the API for a valid datatype.
values (list[str]) – list of values to be parsed
options (dict[str, Any] | None) – any additional options for wikibase parser (for time, ‘precision’ should be specified)
language (str | None) – code of the language to parse the value in
validate (bool) – whether parser should provide data validation as well as parsing

Returns:

list of parsed values

Raises:

ValueError – parsing failed due to some invalid input values

Return type:

list[Any]

preload_entities(pagelist, groupsize=50)[source]#

Yield subclasses of WikibaseEntity’s with content prefilled.

Note

Pages will be iterated in a different order than in the underlying pagelist.

Parameters:

pagelist (Iterable[WikibaseEntity | Page]) – an iterable that yields either WikibaseEntity objects, or Page objects linked to an ItemPage.
groupsize (int) – how many pages to query at a time

Return type:

Generator[WikibaseEntity, None, None]

property property_namespace#

Return namespace for properties.

Returns:: property namespace
Return type:: Namespace

removeClaims(claims, bot=True, summary=None, tags=None)[source]#

Remove claims.

Changed in version 9.4: tags parameter was added

Changed in version 10.0: deprecated baserevid parameter was removed

Parameters:

claims (list[Claim]) – Claims to be removed
bot (bool) – Whether to mark the edit as a bot edit
summary (str | None) – Edit summary
tags (str | None) – Change tags to apply to the revision

removeSources(claim, sources, bot=True, summary=None, tags=None)[source]#

Remove sources.

Changed in version 9.4: tags parameter was added

Changed in version 10.0: deprecated baserevid parameter was removed

Parameters:

claim (Claim) – A Claim object to remove the sources from
sources (list[Claim]) – A list of Claim objects that are sources
bot (bool) – Whether to mark the edit as a bot edit
summary (str | None) – Edit summary
tags (str | None) – Change tags to apply to the revision

remove_form(form, *, bot=True, baserevid=None)[source]#

Remove a form.

Parameters:

form (LexemeForm) – Form to be removed
bot (bool)

Keyword Arguments:

bot – Whether to mark the edit as a bot edit
baserevid – Base revision id override, used to detect conflicts.

Return type:

dict

remove_qualifiers(claim, qualifiers, bot=True, summary=None, tags=None)[source]#

Remove qualifiers.

Changed in version 9.4: tags parameter was added

Changed in version 10.0: deprecated baserevid parameter was removed

Parameters:

claim (Claim) – A Claim object to remove the qualifier from
qualifiers (list[Claim]) – Claim objects currently used as a qualifiers
bot (bool) – Whether to mark the edit as a bot edit
summary (str | None) – Edit summary
tags (str | None) – Change tags to apply to the revision

save_claim(claim, summary=None, bot=True, tags=None)[source]#

Save the whole claim to the wikibase site.

Changed in version 9.4: tags parameter was added

Parameters:

claim (pywikibot.page.Claim) – The claim to save
bot (bool) – Whether to mark the edit as a bot edit
summary (str | None) – Edit summary
tags (str | None) – Change tags to apply to the revision

Raises:

NoPageError – missing the the snak value
NotImplementedError – claim.isReference or claim.isQualifier is given

search_entities(search, language, total=None, **kwargs)[source]#

Search for pages or properties that contain the given text.

Parameters:

search (str) – Text to find.
language (str) – Language to search in.
total (int | None) – Maximum number of pages to retrieve in total, or None in case of no limit.

Returns:

‘search’ list from API output.

Return type:

Generator

set_redirect_target(from_item, to_item, bot=True)[source]#

Make a redirect to another item.

Parameters:

to_item (ItemPage) – title of target item.
from_item (ItemPage) – Title of the item to be redirected.
bot (bool) – Whether to mark the edit as a bot edit

property sparql_endpoint#

Return the sparql endpoint url, if any has been set.

Returns:: sparql endpoint url
Return type:: str|None

tabular_data_repository()[source]#: Return Site object for the tabular-data repository e.g. commons.

wbsetaliases(itemdef, aliases, **kwargs)[source]#

Set aliases for a single Wikibase entity.

See self._wbset_action() for parameters

wbsetdescription(itemdef, description, **kwargs)[source]#

Set description for a single Wikibase entity.

See self._wbset_action()

wbsetlabel(itemdef, label, **kwargs)[source]#

Set label for a single Wikibase entity.

See self._wbset_action() for parameters

wbsetsitelink(itemdef, sitelink, **kwargs)[source]#

Set, remove or modify a sitelink on a Wikibase item.

See self._wbset_action() for parameters

`Obsolete Sites` — Outdated Sites#

Objects representing obsolete MediaWiki sites.

class pywikibot.site._obsoletesites.ClosedSite(code, fam=None, user=None)[source]#

Bases: APISite

Site closed to read-only mode.

Parameters:

code (str)
fam (str | pywikibot.family.Family | None)
user (str | None)

is_uploaddisabled()[source]#

Upload is disabled on site.

Return type:: bool

newpages(**kwargs)[source]#

An error instead of pointless API call.

Return type:: None

page_restrictions(page)[source]#

Return a dictionary reflecting page protections.

Parameters:: page (Page)
Return type:: dict[str, tuple[str, str]]

recentchanges(**kwargs)[source]#

An error instead of pointless API call.

Return type:: None

class pywikibot.site._obsoletesites.RemovedSite(code, fam=None, user=None)[source]#

Bases: BaseSite

Site removed from a family.

Parameters:

code (str) – the site’s language code
fam (str or pywikibot.family.Family) – wiki family name (optional)
user (str) – bot user name (optional)

`Siteinfo` — Site Info Container#

Objects representing site info data contents.

class pywikibot.site._siteinfo.Siteinfo(site)[source]#

Bases: Container

A ‘dictionary’ like container for siteinfo.

This class queries the server to get the requested siteinfo property. Optionally it can cache this directly in the instance so that later requests don’t need to query the server.

All values of the siteinfo property ‘general’ are directly available.

Initialise it with an empty cache.

BOOLEAN_PROPS = {'general': ['imagewhitelistenabled', 'langconversion', 'titleconversion', 'rtl', 'readonly', 'writeapi', 'variantarticlepath', 'misermode', 'uploadsenabled'], 'magicwords': ['case-sensitive'], 'namespaces': ['subpages', 'content', 'nonincludable']}#

WARNING_REGEX = re.compile('Unrecognized values? for parameter ["\\\']siprop["\\\']: (.+?)\\.?')#

clear()[source]#

Remove all items from Siteinfo.

Added in version 7.1.

Return type:: None

get(key, get_default=True, cache=True, expiry=False)[source]#

Return a siteinfo property.

It will never throw an APIError if it only stated, that the siteinfo property doesn’t exist. Instead it will use the default value.

See also

_get_siteinfo

Parameters:

key (str) – The name of the siteinfo property.
get_default (bool) – Whether to throw an KeyError if the key is invalid.
cache (bool) – Caches the result internally so that future accesses via this method won’t query the server.
expiry (datetime | float | bool) – If the cache is older than the expiry it ignores the cache and queries the server to get the newest value.

Returns:

The gathered property

Raises:

KeyError – If the key is not a valid siteinfo property and the get_default option is set to False.

Return type:

Any

get_requested_time(key)[source]#

Return when ‘key’ was successfully requested from the server.

If the property is actually in the siprop ‘general’ it returns the last request from the ‘general’ siprop.

Parameters:: key (str) – The siprop value or a property of ‘general’.
Returns:: The last time the siprop of ‘key’ was requested.
Return type:: None (never), False (default), datetime.datetime (cached)

is_cached(key)[source]#

Return whether the value is cached.

Added in version 7.1.

Parameters:: key (str)
Return type:: bool

is_recognised(key)[source]#

Return if ‘key’ is a valid property name. ‘None’ if not cached.

Parameters:: key (str)
Return type:: bool | None

`Namespace` — Namespace Object#

Objects representing Namespaces of MediaWiki site.

class pywikibot.site._namespace.BuiltinNamespace(*values)[source]#

Bases: IntEnum

Builtin namespace enum.

CATEGORY = 14#

CATEGORY_TALK = 15#

FILE = 6#

FILE_TALK = 7#

HELP = 12#

HELP_TALK = 13#

MAIN = 0#

MEDIA = -2#

MEDIAWIKI = 8#

MEDIAWIKI_TALK = 9#

PROJECT = 4#

PROJECT_TALK = 5#

SPECIAL = -1#

TALK = 1#

TEMPLATE = 10#

TEMPLATE_TALK = 11#

USER = 2#

USER_TALK = 3#

property canonical: str#: Canonical form of MediaWiki built-in namespace.

Added in version 7.1.

class pywikibot.site._namespace.MetaNamespace(name, bases, dic)[source]#

Bases: ABCMeta

Metaclass for Namespace attribute settings.

Added in version 9.0.

Set Namespace.FOO to BuiltinNamespace.FOO for each builtin ns.

class pywikibot.site._namespace.Namespace(id, canonical_name=None, custom_name=None, aliases=None, **kwargs)[source]#

Bases: Iterable, ComparableMixin

Namespace site data object.

This is backwards compatible with the structure of entries in site._namespaces which were a list of:

[customised namespace,
 canonical namespace name?,
 namespace alias*]

If the canonical_name is not provided for a namespace between -2 and 15, the MediaWiki built-in names are used. Image and File are aliases of each other by default.

If only one of canonical_name and custom_name are available, both properties will have the same value.

Changed in version 9.0: metaclass from MetaNamespace

Parameters:

canonical_name (str | None) – Canonical name
custom_name (str | None) – Name defined in server LocalSettings.php
aliases (list[str] | None) – Aliases

CATEGORY = 14#

CATEGORY_TALK = 15#

FILE = 6#

FILE_TALK = 7#

HELP = 12#

HELP_TALK = 13#

MAIN = 0#

MEDIA = -2#

MEDIAWIKI = 8#

MEDIAWIKI_TALK = 9#

PROJECT = 4#

PROJECT_TALK = 5#

SPECIAL = -1#

TALK = 1#

TEMPLATE = 10#

TEMPLATE_TALK = 11#

USER = 2#

USER_TALK = 3#

classmethod builtin_namespaces(case='first-letter')[source]#

Return a dict of the builtin namespaces.

Parameters:: case (str)

canonical_namespaces#

classproperty Return the canonical forms of MediaWiki built-in namespaces.

Changed in version 7.1: implemented as classproperty using BuiltinNamespace IntEnum.

Return type:: dict[int, str]

canonical_prefix()[source]#: Return the canonical name with required colons.

custom_prefix()[source]#: Return the custom name with required colons.

static default_case(id, default_case=None)[source]#: Return the default fixed case value for the namespace ID.

static normalize_name(name)[source]#

Remove an optional colon before and after name.

TODO: reject illegal characters.

class pywikibot.site._namespace.NamespacesDict(namespaces)[source]#

Bases: Mapping

An immutable dictionary containing the Namespace instances.

It adds a deprecation message when called as the ‘namespaces’ property of APISite was callable.

Create new dict using the given namespaces.

lookup_name(name)[source]#

Find the Namespace for a name also checking aliases.

Parameters:: name (str) – Name of the namespace.
Return type:: Namespace | None

lookup_normalized_name(name)[source]#

Find the Namespace for a name also checking aliases.

The name has to be normalized and must be lower case.

Parameters:: name (str) – Name of the namespace.
Return type:: Namespace | None

resolve(identifiers)[source]#

Resolve namespace identifiers to obtain Namespace objects.

Identifiers may be any value for which int() produces a valid namespace id, except bool, or any string which Namespace.lookup_name successfully finds. A numerical string is resolved as an integer.

Parameters:

identifiers (iterable of str or Namespace key, or a single instance of those types) – namespace identifiers

Returns:

list of Namespace objects in the same order as the identifiers

Raises:

KeyError – a namespace identifier was not resolved
TypeError – a namespace identifier has an inappropriate type such as NoneType or bool

Return type:

list[Namespace]

`TokenWallet` — Token Wallet#

Objects representing api tokens.

class pywikibot.site._tokenwallet.TokenWallet(site)[source]#

Bases: Container

Container for tokens.

You should not use this container class directly; use APISite.tokens instead which gives access to the site’s TokenWallet instance.

Parameters:: site (APISite)

clear()[source]#: Clear the self._tokens cache. Tokens are reloaded when needed.

Added in version 8.0.

load_tokens(*args, **kwargs)[source]#

Clear cache to lazy load tokens when needed.

Deprecated since version 8.0: Use clear() instead.

Changed in version 8.0: Clear the cache instead of loading tokens. All parameters are ignored.

Parameters:

args (Any)
kwargs (Any)

Return type:

None

update_tokens(tokens)[source]#

Return a list of new tokens for a given list of tokens.

This method can be used if a token is outdated and has to be renewed but the token type is unknown and we only have the old token. It first gets the token names from all given tokens, clears the cache and returns fresh new tokens of the found types.

Usage:

>>> import pywikibot
>>> site = pywikibot.Site()
>>> tokens = [site.tokens['csrf']]
>>> new_tokens = site.tokens.update_tokens(tokens)

Listing 1 An example for replacing request token parameters#

r._params['token'] = r.site.tokens.update_tokens(r._params['token'])

Added in version 8.0.

Parameters:: tokens (list[str])
Return type:: list[str]

`Uploader` — Uploader Interface#

Objects representing API upload to MediaWiki site.

class pywikibot.site._upload.Uploader(site, filepage, *, source_filename=None, source_url=None, comment=None, text=None, watch=False, chunk_size=0, asynchronous=False, ignore_warnings=False, report_success=None)[source]#

Bases: object

Uploader class to upload a file to the wiki.

Added in version 7.1.

Parameters:

site (pywikibot.site.APISite) – The current site to work on
filepage (FilePage) – a FilePage object from which the wiki-name of the file will be obtained.
source_filename (str | None) – path to the file to be uploaded
source_url (str | None) – URL of the file to be uploaded
comment (str | None) – Edit summary; if this is not provided, then filepage.text will be used. An empty summary is not permitted. This may also serve as the initial page text (see below).
text (str | None) – Initial page text; if this is not set, then filepage.text will be used, or comment.
watch (bool) – If true, add filepage to the bot user’s watchlist
chunk_size (int) – The chunk size in bytes for chunked uploading (see API:Upload#Chunked_uploading). It will only upload in chunks, if the chunk size is positive but lower than the file size.
asynchronous (bool) – Make potentially large file operations asynchronous on the server side when possible.
ignore_warnings (bool or callable or iterable of str) –
It may be a static boolean, a callable returning a boolean or an iterable. The callable gets a list of UploadError instances and the iterable should contain the warning codes for which an equivalent callable would return True if all UploadError codes are in that list. If the result is False it’ll not continue uploading the file and otherwise disable any warning and reattempt to upload the file.

Note

If report_success is True or None it’ll raise an UploadError exception if the static boolean is False.
report_success (bool | None) – If the upload was successful it’ll print a success message and if ignore_warnings is set to False it’ll raise an UploadError if a warning occurred. If it’s None (default) it’ll be True if ignore_warnings is a bool and False otherwise. If it’s True or None ignore_warnings must be a bool.

classmethod create_warnings_list(response, file_key)[source]#: Create a list of upload errors.

submit(request, result, data_result, ignore_warnings, ignore_all_warnings, report_success, file_key)[source]#

Submit request and return whether upload was successful.

Parameters:: data_result (str | None)
Return type:: bool

upload()[source]#

Check for required parameters to upload and run the job.

Returns:: Whether the upload was successful.
Return type:: bool

upload_warnings = {'bad-prefix': 'Target filename has a bad prefix {msg}.', 'badfilename': 'Target filename is invalid.', 'duplicate': 'Uploaded file is a duplicate of {msg}.', 'duplicate-archive': 'The file is a duplicate of a deleted file {msg}.', 'duplicate-version': 'The upload is an exact duplicate of older version(s) of this file.', 'empty-file': 'File {msg} is empty.', 'exists': 'File {msg} already exists.', 'exists-normalized': 'File exists with different extension as {msg!r}.', 'filetype-unwanted-type': 'File {msg} type is unwanted type.', 'no-change': 'The upload is an exact duplicate of the current version of this file.', 'page-exists': 'Target filename exists but with a different file {msg}.', 'was-deleted': 'The file {msg} was previously deleted.'}#

site — MediaWiki sites#

BaseSite — Base Class for Sites#

APISite — API Interface for Sites#

DataSite — API Interface for Wikibase#

Obsolete Sites — Outdated Sites#

Siteinfo — Site Info Container#

Namespace — Namespace Object#

TokenWallet — Token Wallet#

Uploader — Uploader Interface#

`site` — MediaWiki sites#

`BaseSite` — Base Class for Sites#

`APISite` — API Interface for Sites#

`DataSite` — API Interface for Wikibase#

`Obsolete Sites` — Outdated Sites#

`Siteinfo` — Site Info Container#

`Namespace` — Namespace Object#

`TokenWallet` — Token Wallet#

`Uploader` — Uploader Interface#