Compare commits

..

No commits in common. "targetable-sites" and "v4.4.1" have entirely different histories.

22 changed files with 52 additions and 233 deletions

View File

@ -1,4 +1,3 @@
azion
builtins
github
hakiri
@ -8,7 +7,6 @@ Microsoft
ssh
Statictastic
statictastic
supranode
ubuntu
Wikipedia
workaround

View File

@ -1,28 +1,3 @@
## HEAD
### Bug Fixes
* Avoid caching resource when called via `include_relative` tag (#9784)
* Fix logs containing IPv6 URLs (#9813)
* Do not treat colons in `url_placeholders` as URI delimiters (#9850)
### Documentation
* Improve documentation on collections in step-by-step walkthrough (#9803)
* Add `https://form.taxi` as a third-party resource for forms (#9802)
* Add Supranode to third-party deployment guide (#9786)
* Document the need for a `Gemfile` in deployment step of step-by-step walkthrough (#9805)
* Add Azion to the 3rd party deployment docs (#9811)
* Add ruby-erb prerequisite for Arch Linux installations (#9832)
### Development Fixes
* Improve readability of `post_url` tag (#9829)
### Minor Enhancements
* feat: Allowing post_url tag to receive liquid variables (#9776)
## 4.4.1 / 2025-01-29
### Bug Fixes

View File

@ -47,9 +47,9 @@ Read this [Jekyll step-by-step guide](https://www.netlify.com/blog/2020/04/02/a-
[Render](https://render.com) provides zero config continuous deployment for static sites. The service is free under 100GB monthly bandwidth.
## Hostman
## Hostman
[Hostman](https://hostman.com) allows you to host websites for free with no configurations. Read [this guide](https://hostman.com/docs/jekyll) to deploy your Jekyll site on Hostman.
[Hostman](https://hostman.com) allows you to host websites for free with no configurations. Read [this guide](https://hostman.com/docs/jekyll) to deploy your Jekyll site on Hostman.
## Static Publisher
@ -70,17 +70,7 @@ Read this [Jekyll step-by-step guide](https://www.21yunbox.com/docs/#/deploy-jek
[Layer0](https://www.layer0.co) is an all-in-one platform to develop, deploy, preview, experiment on, monitor, and run your headless frontend. It is focused on large, dynamic websites and best-in-class performance through EdgeJS (a JavaScript-based Content Delivery Network), predictive prefetching, and performance monitoring. Layer0 offers a free tier. Get started in just a few minutes by following [Layer0's guide to deploying Jekyll](https://docs.layer0.co/guides/jekyll).
## Kinsta Application Hosting
[Kinsta Application Hosting](https://kinsta.com/application-hosting) is a Cloud Platform designed to help your company and dev teams ship web projects faster and more efficiently. You can host your apps, databases, and sites all in one place. Easily connect with GitHub and automate deployments and get 24/7 support for all your favorite languages and frameworks.
Read [this guide](https://kinsta.com/docs/jekyll-static-site-example/) to learn how to deploy Jekyll site on Kinsta.
## Supranode
[Supranode](https://supranode.com) offers customizable continuous deployment for static websites, featuring automatic HTTPS, a high-performance CDN, secret management, deployment previews, password protection, and more.
## Azion
[Azion](https://www.azion.com/en/) is an web platform that provides a wide range of services. It allows you to host static sites, including Jekyll-powered websites, with features like automatic HTTPS, custom domains, and real-time analytics. Azion's platform is designed for performance and scalability, making it an excellent choice for hosting global websites.
In [this guide](https://www.azion.com/en/documentation/products/guides/jekyll-boilerplate/) you can learn how to deploy a Jekyll site on Azion.

View File

@ -40,7 +40,7 @@ sudo emerge --ask --verbose jekyll
### ArchLinux
```sh
sudo pacman -S ruby base-devel ruby-erb
sudo pacman -S ruby base-devel
```
### OpenSUSE

View File

@ -188,25 +188,3 @@ You can also use this tag to create a link to a post in Markdown as follows:
[Name of Link]({% post_url 2010-07-21-name-of-post %})
```
{% endraw %}
Now lets say you have a [datafile]({{ '/docs/datafiles/' | relative_url }}) `_data/cool_posts.yaml` used to keep track
of certain posts that you intend to be listed as say *Cool Posts*:
```yaml
- title: "An Awesome Post"
slug: "2010-07-21-name-of-post"
- title: "Another Awesome Post"
slug: "2016-07-26-name-of-post"
```
You may list such posts using the `post_url` tag as well (from {%- include docs_version_badge.html version="4.5.0" -%}):
{% raw %}
```liquid
Cool posts:
{%- for cool_post in site.data.cool_posts %}
- [{{ cool_post.title }}]({% post_url {{ cool_post.slug }} %})
{%- endfor %}
```
{% endraw %}

View File

@ -57,7 +57,7 @@ Ted has been eating fruit since he was baby.
Let's add a page which lists all the authors on the site. Jekyll makes the
collection available at `site.authors`.
Create `staff.html` in the root directory and iterate over `site.authors` to output all the staff:
Create `staff.html` and iterate over `site.authors` to output all the staff:
{% raw %}
```liquid

View File

@ -11,7 +11,7 @@ It's good practice to have a [Gemfile](/docs/ruby-101/#gemfile) for your site.
This ensures the version of Jekyll and other gems remains consistent across
different environments.
If you completed step one in this tutorial, you have already created a Gemfile. If you skipped step one, create a `Gemfile` in the root.
Create a `Gemfile` in the root.
The file should be called 'Gemfile' and should *not* have any extension.
You can create a Gemfile with Bundler and then add the `jekyll` gem:
@ -20,7 +20,7 @@ bundle init
bundle add jekyll
```
Your `Gemfile` should look something like:
Your file should look something like:
```ruby
# frozen_string_literal: true

View File

@ -62,7 +62,6 @@ Use a SaaS service as a backend for functionality on your Jekyll site
- [Formspark](https://formspark.io/)
- [Formspree (open source)](https://formspree.io/)
- [formX](https://formx.stream)
- [Form.taxi](https://form.taxi/en/backend)
- [Simple Form](https://getsimpleform.com/)
- [SmartForms](https://smartforms.dev/)
- [Typeform](https://www.typeform.com/templates/c/forms/)

View File

@ -58,22 +58,3 @@ Feature: include_relative Tag
Then I should get a zero exit status
And the _site directory should exist
And I should see "Welcome back Dear Reader!" in "_site/index.html"
Scenario: Include multiple files relative to a page at root
Given I have an "apple.md" page with foo "bar" that contains "{{ page.path }}, {{ page.foo }}"
And I have an "banana.md" page with content:
"""
{% include_relative apple.md %}
{% include_relative cherry.md %}
{{ page.path }}
"""
And I have an "cherry.md" page with foo "lipsum" that contains "{{ page.path }}, {{ page.foo }}"
When I run jekyll build
Then I should get a zero exit status
And the _site directory should exist
And I should see "<p>apple.md, bar</p>" in "_site/apple.html"
And I should see "<hr />\n<p>foo: bar" in "_site/banana.html"
And I should see "<hr />\n<p>foo: lipsum" in "_site/banana.html"
And I should see "<p>cherry.md, lipsum</p>" in "_site/cherry.html"
But I should not see "foo: lipsum" in "_site/cherry.html"

View File

@ -157,37 +157,3 @@ Feature: PostUrl Tag
But the _site directory should exist
And I should see "<p><a href=\"/cats%20and%20dogs/2019/02/04/hello-world.html\">Post 1</a></p>" in "_site/index.html"
And I should see "<p><a href=\"/2019/02/05/hello-again.html\">Post 2</a></p>" in "_site/index.html"
Scenario: Calling for a post via a liquid variable
Given I have a _posts directory
And I have the following post:
| title | date | content |
| Hello World | 2019-02-04 | Lorem ipsum dolor |
And I have an "index.md" page with content:
"""
{% assign value='2019-02-04-hello-world' %}
[Welcome]({% post_url {{ value }} %})
"""
When I run jekyll build
Then I should get a zero exit status
And the _site directory should exist
And I should see "<p><a href=\"/2019/02/04/hello-world.html\">Welcome</a></p>" in "_site/index.html"
Scenario: Calling for posts via a liquid variable in a for tag
Given I have a _posts directory
And I have the following post:
| title | date | content |
| Hello World | 2019-02-04 | Lorem ipsum dolor |
| We Meet Again | 2019-02-05 | Alpha beta gamma |
And I have an "index.md" page with content:
"""
{% assign posts = '2019-02-04-hello-world;2019-02-05-we-meet-again' | split: ';' %}
{%- for slug in posts -%}
[{{ slug }}]({% post_url {{ slug }} %})
{%- endfor %}
"""
When I run jekyll build
Then I should get a zero exit status
And the _site directory should exist
And I should see "<a href=\"/2019/02/04/hello-world.html\">2019-02-04-hello-world</a>" in "_site/index.html"
And I should see "<a href=\"/2019/02/05/we-meet-again.html\">2019-02-05-we-meet-again</a>" in "_site/index.html"

View File

@ -185,8 +185,7 @@ module Jekyll
#
# Returns true if the 'write' metadata is true, false otherwise.
def write?
!!metadata.fetch("output", false) &&
(site.target == metadata.fetch("target", site.target))
!!metadata.fetch("output", false)
end
# The URL template to render collection's documents at.

View File

@ -243,7 +243,7 @@ module Jekyll
def format_url(ssl_enabled, address, port, baseurl = nil)
format("%<prefix>s://%<address>s:%<port>i%<baseurl>s",
:prefix => ssl_enabled ? "https" : "http",
:address => address.include?(":") ? "[#{address}]" : address,
:address => address,
:port => port,
:baseurl => baseurl ? "#{baseurl}/" : "")
end

View File

@ -55,9 +55,8 @@ module Jekyll
EM.schedule { @started_event.set }
EM.add_shutdown_hook { @stopped_event.set }
host = opts["host"].include?(":") ? "[#{opts["host"]}]" : opts["host"]
Jekyll.logger.info "LiveReload address:",
"http://#{host}:#{opts["livereload_port"]}"
"http://#{opts["host"]}:#{opts["livereload_port"]}"
end
end
@thread.abort_on_exception = true

View File

@ -30,7 +30,6 @@ module Jekyll
"limit_posts" => 0,
"future" => false,
"unpublished" => false,
"target" => "default",
# Plugins
"whitelist" => [],

View File

@ -348,20 +348,17 @@ module Jekyll
end
# Determine whether this document should be written.
# Based on the Collection to which it belongs
# and site and document target.
# Based on the Collection to which it belongs.
#
# True if the document has a collection and if that collection's #write?
# method returns true, and if the site's Publisher will publish the document,
# and if the document's target matches the site target or is undefined.
# method returns true, and if the site's Publisher will publish the document.
# False otherwise.
#
# rubocop:disable Naming/MemoizedInstanceVariableName
def write?
return @write_p if defined?(@write_p)
@write_p = collection&.write? && site.publisher.publish?(self) &&
(site.target == data.fetch("target", site.target))
@write_p = collection&.write? && site.publisher.publish?(self)
end
# rubocop:enable Naming/MemoizedInstanceVariableName

View File

@ -15,7 +15,7 @@ module Jekyll
private delegate_method_as :data, :fallback_data
delegate_methods :id, :output, :content, :to_s, :relative_path, :url, :date
data_delegators "title", "categories", "tags", :target
data_delegators "title", "categories", "tags"
def collection
@obj.collection.label

View File

@ -180,7 +180,7 @@ module Jekyll
end
def write?
site.config["target"] == data.fetch("target", site.config["target"])
true
end
def excerpt_separator

View File

@ -6,7 +6,7 @@ module Jekyll
:file_read_opts, :future, :gems, :generators, :highlighter,
:include, :inclusions, :keep_files, :layouts, :limit_posts,
:lsi, :pages, :permalink_style, :plugin_manager, :plugins,
:reader, :safe, :target, :show_drafts, :static_files, :theme, :time,
:reader, :safe, :show_drafts, :static_files, :theme, :time,
:unpublished
attr_reader :cache_dir, :config, :dest, :filter_cache, :includes_load_paths,
@ -47,7 +47,7 @@ module Jekyll
def config=(config)
@config = config.clone
%w(safe lsi highlighter baseurl exclude include future unpublished target
%w(safe lsi highlighter baseurl exclude include future unpublished
show_drafts limit_posts keep_files).each do |opt|
send(:"#{opt}=", config[opt])
end
@ -360,7 +360,7 @@ module Jekyll
end
def each_site_file
pages.each { |page| yield (page) if page.write? }
pages.each { |page| yield page }
static_files.each { |file| yield(file) if file.write? }
collections.each_value { |coll| coll.docs.each { |doc| yield(doc) if doc.write? } }
end

View File

@ -249,11 +249,6 @@ module Jekyll
end
class IncludeRelativeTag < IncludeTag
def load_cached_partial(path, context)
context.registers[:cached_partials] ||= {}
context.registers[:cached_partials][path] ||= parse_partial(path, context)
end
def tag_includes_dirs(context)
Array(page_path(context)).freeze
end
@ -272,17 +267,6 @@ module Jekyll
path = File.join(site.config["collections_dir"], path) if page["collection"]
path.delete_suffix("/#excerpt")
end
# Since Jekyll 4 caches convertibles based on their path within the only instance of
# `LiquidRenderer`, initialize a new LiquidRenderer instance on every render of this
# tag to bypass caching rendered output of page / document.
def parse_partial(path, context)
LiquidRenderer.new(context.registers[:site]).file(path).parse(read_file(path, context))
rescue Liquid::Error => e
e.template_name = path
e.markup_context = "included " if e.markup_context.nil?
raise e
end
end
end
end

View File

@ -3,39 +3,34 @@
module Jekyll
module Tags
class PostComparer
# Deprecated (soft; No interpreter warnings).
# To be removed in v5.0.
# Use private constant `POST_PATH_MATCHER` instead.
MATCHER = %r!^(.+/)*(\d+-\d+-\d+)-(.*)$!.freeze
POST_PATH_MATCHER = %r!\A(.+/)*?(\d{2,4}-\d{1,2}-\d{1,2})-([^/]*)\z!.freeze
private_constant :POST_PATH_MATCHER
attr_reader :path, :date, :slug, :name
def initialize(name)
@name = name
all, @path, @date, @slug = *name.delete_prefix("/").match(POST_PATH_MATCHER)
all, @path, @date, @slug = *name.sub(%r!^/!, "").match(MATCHER)
unless all
raise Jekyll::Errors::InvalidPostNameError,
"'#{name}' does not contain valid date and/or title."
end
basename_pattern = "#{date}-#{Regexp.escape(slug)}\\.[^.]+"
@name_regex = %r!\A_posts/#{path}#{basename_pattern}|\A#{path}_posts/?#{basename_pattern}!
@name_regex = %r!^_posts/#{path}#{basename_pattern}|^#{path}_posts/?#{basename_pattern}!
end
def post_date
@post_date ||= Utils.parse_date(date, "Path '#{name}' does not contain valid date.")
@post_date ||= Utils.parse_date(
date,
"'#{date}' does not contain valid date and/or title."
)
end
# Returns `MatchData` or `nil`.
def ==(other)
other.relative_path.match(@name_regex)
end
# Deprecated. To be removed in v5.0.
def deprecated_equality(other)
slug == post_slug(other) &&
post_date.year == other.date.year &&
@ -45,9 +40,9 @@ module Jekyll
private
# Construct the directory-aware post slug for a Jekyll::Document object.
# Construct the directory-aware post slug for a Jekyll::Post
#
# other - the Jekyll::Document object.
# other - the Jekyll::Post
#
# Returns the post slug with the subdirectory (relative to _posts)
def post_slug(other)
@ -63,71 +58,47 @@ module Jekyll
class PostUrl < Liquid::Tag
include Jekyll::Filters::URLFilters
def initialize(tag_name, markup, tokens)
def initialize(tag_name, post, tokens)
super
@markup = markup.strip
@template = Liquid::Template.parse(@markup) if @markup.include?("{{")
# Deprecated instance_variables.
# To be removed in Jekyll v5.0.
@orig_post = @markup
@post = nil
@orig_post = post.strip
begin
@post = PostComparer.new(@orig_post)
rescue StandardError => e
raise Jekyll::Errors::PostURLError, <<~MSG
Could not parse name of post "#{@orig_post}" in tag 'post_url'.
Make sure the post exists and the name is correct.
#{e.class}: #{e.message}
MSG
end
end
def render(context)
@context = context
@resolved_markup = @template&.render(@context) || @markup
site = context.registers[:site]
begin
@post_comparer = PostComparer.new(@resolved_markup)
rescue StandardError
raise_markup_parse_error
end
# For backwards compatibility only; deprecated instance_variable.
# To be removed in Jekyll v5.0.
@post = @post_comparer
# First pass-through.
site.posts.docs.each do |post|
return relative_url(post) if @post_comparer == post
site.posts.docs.each do |document|
return relative_url(document) if @post == document
end
# First pass-through did not yield the requested post. Search again using legacy matching
# method. Log deprecation warning if a post is detected via this round.
site.posts.docs.each do |post|
next unless @post_comparer.deprecated_equality(post)
# New matching method did not match, fall back to old method
# with deprecation warning if this matches
log_legacy_usage_deprecation
return relative_url(post)
site.posts.docs.each do |document|
next unless @post.deprecated_equality document
Jekyll::Deprecator.deprecation_message(
"A call to '{% post_url #{@post.name} %}' did not match a post using the new " \
"matching method of checking name (path-date-slug) equality. Please make sure " \
"that you change this tag to match the post's name exactly."
)
return relative_url(document)
end
raise_post_not_found_error
end
private
def raise_markup_parse_error
raise Jekyll::Errors::PostURLError, <<~MSG
Could not parse name of post #{@resolved_markup.inspect} in tag 'post_url'.
Make sure the correct name is given to the tag.
Could not find post "#{@orig_post}" in tag 'post_url'.
Make sure the post exists and the name is correct.
MSG
end
def raise_post_not_found_error
raise Jekyll::Errors::PostURLError, <<~MSG
Could not find post #{@resolved_markup.inspect} in tag 'post_url'.
Make sure the post exists and the correct name is given to the tag.
MSG
end
def log_legacy_usage_deprecation
Jekyll::Deprecator.deprecation_message(
"A call to '{% post_url #{@resolved_markup} %}' did not match a post using the new " \
"matching method of checking name (path-date-slug) equality. Please make sure that " \
"you change this tag to match the post's name exactly."
)
end
end
end
end

View File

@ -144,13 +144,7 @@ module Jekyll
# pct-encoded = "%" HEXDIG HEXDIG
# sub-delims = "!" / "$" / "&" / "'" / "(" / ")"
# / "*" / "+" / "," / ";" / "="
#
# `Addressable::URI::CharacterClassesRegexps::PATH` is used to encode
# non-alphanumeric characters such as "[", "]", etc.
Addressable::URI.encode_component(
path,
Addressable::URI::CharacterClassesRegexps::PATH
).encode("utf-8").sub("#", "%23")
Addressable::URI.encode(path).encode("utf-8").sub("#", "%23")
end
# Unescapes a URL path segment

View File

@ -80,16 +80,5 @@ class TestURL < JekyllUnitTest
).to_s
end
end
should "not treat colons in placeholders as uri delimiters" do
assert_equal "/foo/foo%20bar:foobar/", URL.new(
:template => "/:x/:y/",
:placeholders => { :x => "foo", :y => "foo bar:foobar" }
).to_s
end
should "unescape urls with colons" do
assert_equal "/foo/foo bar:foobar/", Jekyll::URL.unescape_path("/foo/foo%20bar:foobar/")
end
end
end