Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR contains the following updates:
6.0.1
->6.0.2
1.21.2
->1.22.0
6.7.0
->6.11.0
4.17.3
->4.23.0
1.5.3
->1.6.1
9.2.7
->9.5.39
0.22.0
->0.26.1
5.2.6
->5.4.0
5.0.0
->5.3.0
2.6.3
->2.7.0
Release Notes
yaml/pyyaml (PyYAML)
v6.0.2
Compare Source
What's Changed
Full Changelog: yaml/pyyaml@6.0.1...6.0.2
mtkennerly/dunamai (dunamai)
v1.22.0
Compare Source
--ignore-untracked
CLI flag was ignored.--commit-length
option.python/importlib_metadata (importlib-metadata)
v6.11.0
Compare Source
v6.10.0
Compare Source
v6.9.0
Compare Source
v6.8.0
Compare Source
python-jsonschema/jsonschema (jsonschema)
v4.23.0
Compare Source
=======
v4.22.0
Compare Source
=======
best_match
(and thereby error messages fromjsonschema.validate
) in cases where there are multiple sibling errors from applyinganyOf
/allOf
-- i.e. when multiple elements of a JSON array have errors, we now do prefer showing errors from earlier elements rather than simply showing an error for the full array (#1250).==
would.v4.21.1
Compare Source
=======
contains
keyword by removing some unnecessary validator (re-)creation.v4.21.0
Compare Source
=======
enum
in the presence of0
or1
to properly considerTrue
andFalse
unequal (#1208).{min,max}{Items,Length,Properties}
when they're checking for emptiness rather than true length.v4.20.0
Compare Source
=======
unevaluatedItems
(resp.unevaluatedProperties
) when behind a$dynamicRef
as specified by the 2020 and 2019 specifications.jsonschema.exceptions.ErrorTree.__setitem__
is now deprecated.More broadly, in general users of
jsonschema
should never be mutating objects owned by the library.v4.19.2
Compare Source
=======
additionalItems
keyword into JSON Schema draft 2020-12, where it was replaced byitems
.v4.19.1
Compare Source
=======
hostname
format.This is the behavior specified by the relevant RFC (1123).
IDN hostname behavior was already correct.
v4.19.0
Compare Source
=======
Validator
protocol directly from the package root is deprecated.Import it from
jsonschema.protocols.Validator
instead.Such resources are assumed to be 2020-12 schemas.
This more closely matches the pre-referencing library behavior.
v4.18.6
Compare Source
=======
jsonschema
specific user agent when automatically retrieving remote references (which is deprecated).v4.18.5
Compare Source
=======
v4.18.4
Compare Source
=======
v4.18.3
Compare Source
=======
applicable_validators
in extended validators.Specifically, validators extending early drafts where siblings of
$ref
were ignored will properly ignore siblings in the extended validator.v4.18.2
Compare Source
=======
jsonschema.RefResolver
and pointer resolution.v4.18.1
Compare Source
=======
jsonschema.RefResolver
based resolution when used in combination with a custom validation dialect (viajsonschema.validators.create
).v4.18.0
Compare Source
=======
This release majorly rehauls the way in which JSON Schema reference resolution is configured.
It does so in a way that should be backwards compatible, preserving old behavior whilst emitting deprecation warnings.
jsonschema.RefResolver
is now deprecated in favor of the newreferencing library <https://github.com/python-jsonschema/referencing/>
_.referencing
will begin in beta, but already is more compliant than the existing$ref
support.This change is a culmination of a meaningful chunk of work to make
$ref
resolution more flexible and more correct.Backwards compatibility should be preserved for existing code which uses
RefResolver
, though doing so is again now deprecated, and all such use cases should be doable using the new APIs.Please file issues on the
referencing
tracker if there is functionality missing from it, or here on thejsonschema
issue tracker if you have issues with existing code not functioning the same, or with figuring out how to change it to usereferencing
.In particular, this referencing change includes a change concerning automatic retrieval of remote references (retrieving
http://foo/bar
automatically within a schema).This behavior has always been a potential security risk and counter to the recommendations of the JSON Schema specifications; it has survived this long essentially only for backwards compatibility reasons, and now explicitly produces warnings.
The
referencing
library itself will not automatically retrieve references if you interact directly with it, so the deprecated behavior is only triggered if you fully rely on the default$ref
resolution behavior and also include remote references in your schema, which will still be retrieved during the deprecation period (after which they will become an error).This should not be a "visible" change in the sense that
requires-python
has been updated, so users using 3.7 should still receivev4.17.3
when installing the library.unevaluatedItems
now properly does not consider items to be evaluated by anadditionalItems
schema ifitems
is missing from the schema, as the specification says in this case thatadditionalItems
must be completely ignored.date
format checker on Python 3.11 (when format assertion behavior is enabled), where it was too liberal (#1076).unevaluatedProperties
(#1075).Deprecations
jsonschema.RefResolver
-- see above for details on the replacementjsonschema.RefResolutionError
-- see above for details on the replacementjsonschema.ErrorTree
-- instead import it viajsonschema.exceptions.ErrorTree
jsonschema.FormatError
-- instead import it viajsonschema.exceptions.FormatError
mkdocs/mkdocs (mkdocs)
v1.6.1
Compare Source
Version 1.6.1 (Friday 30th August, 2024)
Fixed
SOURCE_DATE_EPOCH=0
is set. #3795mkdocs_theme.yml
config is empty. #3700python -W
andPYTHONWARNINGS
instead of overriding the configuration. #38090.0.0.0
dev server warning. #3784changefreq
fromsitemap.xml
. #3629Added
v1.6.0
Compare Source
Local preview
mkdocs serve
no longer locks up the browser when more than 5 tabs are open. This is achieved by closing the polling connection whenever a tab becomes inactive. Background tabs will no longer auto-reload either - that will instead happen as soon the tab is opened again. Context: #3391New flag
serve --open
to open the site in a browser.After the first build is finished, this flag will cause the default OS Web browser to be opened at the home page of the local site.
Context: #3500
Drafts
The
exclude_docs
config no longer has any special behavior formkdocs serve
- it now always completely excludes the listed documents from the site.If you wish to use the "drafts" functionality like the
exclude_docs
key used to do in MkDocs 1.5, please switch to the new config keydraft_docs
.See documentation.
Other changes:
Update to deduction of page titles
MkDocs 1.5 had a change in behavior in deducing the page titles from the first heading. Unfortunately this could cause unescaped HTML tags or entities to appear in edge cases.
Now tags are always fully sanitized from the title. Though it still remains the case that
Page.title
is expected to contain HTML entities and is passed directly to the themes.Images (notably, emojis in some extensions) get preserved in the title only through their
alt
attribute's value.Context: #3564, #3578
Themes
"readthedocs" theme
Fix: "readthedocs" theme can now correctly handle deeply nested nav configurations (over 2 levels deep), without confusedly expanding all sections and jumping around vertically. (#3464)
Fix: "readthedocs" theme now shows a link to the repository (with a generic logo) even when isn't one of the 3 known hosters. (#3435)
"readthedocs" theme now also has translation for the word "theme" in the footer that mistakenly always remained in English. (#3613, #3625)
"mkdocs" theme
The "mkdocs" theme got a big update to a newer version of Bootstrap, meaning a slight overhaul of styles. Colors (most notably of admonitions) have much better contrast.
The "mkdocs" theme now has support for dark mode - both automatic (based on the OS/browser setting) and with a manual toggle. Both of these options are not enabled by default and need to be configured explicitly.
See
color_mode
,user_color_mode_toggle
in documentation.Context: #3493, #3649
Configuration
New "
enabled
" setting for all pluginsYou may have seen some plugins take up the convention of having a setting
enabled: false
(or usually controlled through an environment variable) to make the plugin do nothing.Now every plugin has this setting. Plugins can still choose to implement this config themselves and decide how it behaves (and unless they drop older versions of MkDocs, they still should for now), but now there's always a fallback for every plugin.
See documentation. Context: #3395
Validation
Validation of hyperlinks between pages
Absolute links
If you dislike having to always use relative links, now you can opt into absolute links and have them work correctly.
If you set the setting
validation.links.absolute_links
to the new valuerelative_to_docs
, all Markdown links starting with/
will be understood as being relative to thedocs_dir
root. The links will then be validated for correctness according to all the other rules that were already working for relative links in prior versions of MkDocs. For the HTML output, these links will still be turned relative so that the site still works reliably.So, now any document (e.g. "dir1/foo.md") can link to the document "dir2/bar.md" as
[link](/dir2/bar.md)
, in addition to the previously only correct way[link](../dir2/bar.md)
.You have to enable the setting, though. The default is still to just skip any processing of such links.
See documentation. Context: #3485
Absolute links within nav
Absolute links within the
nav:
config were also always skipped. It is now possible to also validate them in the same way withvalidation.nav.absolute_links
. Though it makes a bit less sense because then the syntax is simply redundant with the syntax that comes without the leading slash.Anchors
There is a new config setting that is recommended to enable warnings for:
Example of a warning that this can produce:
Any of the below methods of declaring an anchor will be detected by MkDocs:
Plugins and extensions that insert anchors, in order to be compatible with this, need to be developed as treeprocessors that insert
etree
elements as their mode of operation, rather than raw HTML which is undetectable for this purpose.If you as a user are dealing with falsely reported missing anchors and there's no way to resolve this, you can choose to disable these messages by setting this option to
ignore
(and they are at INFO level by default anyway).See documentation. Context: #3463
Other changes:
When the
nav
config is not specified at all, thenot_in_nav
setting (originally added in 1.5.0) gains an additional behavior: documents covered bynot_in_nav
will not be part of the automatically deduced navigation. Context: #3443Fix: the
!relative
YAML tag formarkdown_extensions
(originally added in 1.5.0) - it was broken in many typical use cases.See documentation. Context: #3466
Config validation now exits on first error, to avoid showing bizarre secondary errors. Context: #3437
MkDocs used to shorten error messages for unexpected errors such as "file not found", but that is no longer the case, the full error message and stack trace will be possible to see (unless the error has a proper handler, of course). Context: #3445
Upgrades for plugin developers
Plugins can add multiple handlers for the same event type, at multiple priorities
See
mkdocs.plugins.CombinedEvent
in documentation. Context: #3448Enabling true generated files and expanding the
File
APISee documentation.
There is a new pair of attributes
File.content_string
that becomes the official API for obtaining the content of a file and is used by MkDocs itself.This replaces the old approach where one had to manually read the file located at
File.abs_src_path
, although that is still the primary action that these new attributes do under the hood.The content of a
File
can be backed by a string and no longer has to be a real existing file atabs_src_path
.It is possible to set the attribute
File.content_string
orFile.content_bytes
and it will take precedence overabs_src_path
.Further,
abs_src_path
is no longer guaranteed to be present and can beNone
instead. MkDocs itself still uses physical files in all cases, but eventually plugins will appear that don't populate this attribute.There is a new constructor
File.generated()
that should be used by plugins instead of theFile()
constructor. It is much more convenient because one doesn't need to manually look up the values such asdocs_dir
anduse_directory_urls
. Its signature is one of:This way, it is now extremely easy to add a virtual file even from a hook:
For large content it is still best to use physical files, but one no longer needs to manipulate the path by providing a fake unused
docs_dir
.There is a new attribute
File.generated_by
that arose by convention - for generated files it should be set to the name of the plugin (the key in theplugins:
collection) that produced this file. This attribute is populated automatically when using theFile.generated()
constructor.It is possible to set the
edit_uri
attribute of aFile
, for example from a plugin or hook, to make it different from the default (equal tosrc_uri
), and this will be reflected in the edit link of the document. This can be useful because some pages aren't backed by a real file and are instead created dynamically from some other source file or script. So a hook could set theedit_uri
to that source file or script accordingly.The
File
object now stores its originalsrc_dir
,dest_dir
,use_directory_urls
values as attributes.Fields of
File
are computed on demand but cached. Only the three above attributes are primary ones, and partly alsodest_uri
. This way, it is possible to, for example, overwritedest_uri
of aFile
, andabs_dest_path
will be calculated based on it. However you need to clear the attribute first usingdel f.abs_dest_path
, because the values are cached.File
instances are now hashable (can be used as keys of adict
). Two files can no longer be considered "equal" unless it's the exact same instance ofFile
.Other changes:
The internal storage of
File
objects inside aFiles
object has been reworked, so any plugins that choose to accessFiles._files
will get a deprecation warning.The order of
File
objects inside aFiles
collection is no longer significant when automatically inferring thenav
. They get forcibly sorted according to the default alphabetic order.Context: #3451, #3463
Hooks and debugging
Hook files can now import adjacent *.py files using the
import
statement. Previously this was possible to achieve only through asys.path
workaround. See the new mention in documentation. Context: #3568Verbose
-v
log shows the sequence of plugin events in more detail - shows each invoked plugin one by one, not only the event type. Context: #3444Deprecations
Python 3.7 is no longer supported, Python 3.12 is officially supported. Context: #3429
The theme config file
mkdocs_theme.yml
no longer executes YAML tags. Context: #3465The plugin event
on_page_read_source
is soft-deprecated because there is always a better alternative to it (see the newFile
API or juston_page_markdown
, depending on the desired interaction).When multiple plugins/hooks apply this event handler, they trample over each other, so now there is a warning in that case.
See documentation. Context: #3503
API deprecations
It is no longer allowed to set
File.page
to a type other thanPage
or a subclass thereof. Context: #3443 - following the deprecation in version 1.5.3 and #3381.Theme._vars
is deprecated - usetheme['foo']
instead oftheme._vars['foo']
utils
:modified_time()
,get_html_path()
,get_url_path()
,is_html_file()
,is_template_file()
are removed.path_to_url()
is deprecated.LiveReloadServer.watch()
no longer accepts a custom callback.Context: #3429
Misc
sitemap.xml.gz
file is slightly more reproducible and no longer changes on every build, but instead only once per day (upon a date change). Context: #3460Other small improvements; see commit log.
squidfunk/mkdocs-material (mkdocs-material)
v9.5.39
: mkdocs-material-9.5.39Compare Source
Thanks go to @ilyagr for their contributions
v9.5.38
: mkdocs-material-9.5.38Compare Source
Thanks go to @gerardkraja for their contributions
v9.5.37
: mkdocs-material-9.5.37Compare Source
v9.5.36
: mkdocs-material-9.5.36Compare Source
v9.5.35
: mkdocs-material-9.5.35Compare Source
v9.5.34
: mkdocs-material-9.5.34Compare Source
Thanks go to @manselmi for their contributions
v9.5.33
: mkdocs-material-9.5.33Compare Source
v9.5.32
: mkdocs-material-9.5.32Compare Source
Thanks go to @joaopalmeiro for their contributions
v9.5.31
: mkdocs-material-9.5.31Compare Source
v9.5.30
: mkdocs-material-9.5.30Compare Source
Thanks go to @sisp for their contributions
v9.5.29
: mkdocs-material-9.5.29Compare Source
Thanks go to @mmontes11 for their contributions
v9.5.28
: mkdocs-material-9.5.28Compare Source
v9.5.27
: mkdocs-material-9.5.27Compare Source
Thanks to @Eilyre for their contributions
v9.5.26
: mkdocs-material-9.5.26Compare Source
v9.5.25
: mkdocs-material-9.5.25Compare Source
v9.5.24
: mkdocs-material-9.5.24Compare Source
v9.5.23
: mkdocs-material-9.5.23Compare Source
v9.5.22
: mkdocs-material-9.5.22Compare Source
v9.5.21
: mkdocs-material-9.5.21Compare Source
Thanks to @kamilkrzyskow for their contributions
v9.5.20
: mkdocs-material-9.5.20Compare Source
v9.5.19
: mkdocs-material-9.5.19Compare Source
setup.py
, now that GitHub fully understandspyproject.toml
v9.5.18
: mkdocs-material-9.5.18Compare Source
v9.5.17
: mkdocs-material-9.5.17Compare Source
Thanks to @stcksmsh for their contributions.
v9.5.16
: mkdocs-material-9.5.16Compare Source
Thanks to @kamilkrzyskow, @Guts, @szg-alex-payne and @natakazakova for their contributions
v9.5.15
: mkdocs-material-9.5.15Compare Source
v9.5.14
: mkdocs-material-9.5.14Compare Source
Thanks to @kamilkrzyskow, @yubiuser and @todeveni for their contributions
v9.5.13
: mkdocs-material-9.5.13Compare Source
Thanks to @kamilkrzyskow and @scepka for their contributions
v9.5.12
: mkdocs-material-9.5.12Compare Source
v9.5.11
: mkdocs-material-9.5.11Compare Source
v9.5.10
: mkdocs-material-9.5.10Compare Source
v9.5.9
: mkdocs-material-9.5.9Compare Source
v9.5.8
: mkdocs-material-9.5.8Compare Source
v9.5.7
: mkdocs-material-9.5.7Compare Source
v9.5.6
: mkdocs-material-9.5.6Compare Source
v9.5.5
: mkdocs-material-9.5.5Compare Source
onclick
handlersv9.5.4
: mkdocs-material-9.5.4Compare Source
v9.5.3
: mkdocs-material-9.5.3Compare Source
v9.5.2
: mkdocs-material-9.5.2Compare Source
slugify
settings in blog plugin configv9.5.1
: mkdocs-material-9.5.1Compare Source
v9.5.0
: mkdocs-material-9.5.0Compare Source
Merged Insiders features of 'Goat's Horn' funding goal
v9.4.14
: mkdocs-material-9.4.14Compare Source
v9.4.13
: mkdocs-material-9.4.13Compare Source
Configuration
📅 Schedule: Branch creation - "every 2 weeks on Sunday" (UTC), Automerge - At any time (no schedule defined).
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.
👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.
This PR was generated by Mend Renovate. View the repository job log.