* pathawks-rubocop/misc:
Fix Page#relative_path so that it consistently does NOT have the prepending slash (previously inconsistent)
Rubocop cleanup for lib/jekyll/layout.rb
Rubocop cleanup for lib/jekyll/plugin_manager.rb
Rubocop cleanup for lib/jekyll/page.rb
* master: (38 commits)
Mention where it came from. [ci skip]
Update history to reflect merge of #4944 [ci skip]
Update history to reflect merge of #4943 [ci skip]
Mention where it came from. [ci skip]
Update history to reflect merge of #4942 [ci skip]
Update history to reflect merge of #4941
External: remove &block arg, use block_given?
Update history to reflect merge of #4936 [ci skip]
lib/jekyll.rb: require document_drop to ease our pain
Sort the results of the require_all glob.
Rubocop fixes
Reset {{ layout }} between each render & merge layout data properly
Add failing test for layout data inheritance bug (#4433)
Add failing test for layout bug (#4897)
Fix tests for plugins in configuration.
Define Drop#each so we can use the new frozen/duping behavior
Don't default 'include' and 'exclude' to an empty array
Fix some minor things in the tests
Freeze configuration defaults & duplicate in deep_merge_hashes if need be.
Remove merge conflicts I forgot to fix.
...
Previously you could do, e.g. {{ site | jsonify }}, but with the introduction of Liquid Drops, this didn't work anymore.
This PR adds the ability to render drops as JSON. You can safely run drop.to_json and it should Do the Right Thing.
This commit cleans up rubocop violations for the following files:
test/test_ansi.rb
test/test_cleaner.rb
test/test_coffeescript.rb
test/test_collections.rb
test/test_command.rb
test/test_commands_serve.rb
* Allow users to filter directories by ending their path with "/"
* Allow users to filter with a Regexp, some scenariors can really require it.
* Use Pathutil#in_path? for Symlink verification, it real/expand.
This also requires some downstream work in "jekyll-watch" which at this time is
not very robust, it doesn't recognize the difference either, and should probably
start doing so (what I mean is detecting "/" and using the full path.)
* master: (58 commits)
Update history to reflect merge of #4792 [ci skip]
Update history to reflect merge of #4793 [ci skip]
Update history to reflect merge of #4804 [ci skip]
Update history to reflect merge of #4754 [ci skip]
Update history to reflect merge of #4813 [ci skip]
Added missing single quote on rsync client side command
Add v3.0.4 and v3.1.3 to the history.
Fixed typo
Add jekyll-autoprefixer plugin
Explicitly require Filters rather than implicitly.
Update history to reflect merge of #4786 [ci skip]
Update history to reflect merge of #4789 [ci skip]
updates example domain in config template
Globalize Jekyll's Filters.
Update JRuby to 9.0.5.0; Drop the double digit test.
Update Rack-Jekyll Heroku deployment blog post url
convertible: use Document::YAML_FRONT_MATTER_REGEXP to parse transformable files
Update history to reflect merge of #4734 [ci skip]
Update history to reflect merge of #4478 [ci skip]
Fix rubocop warning.
...
This commit introduces a where_exp filter, which can be used as follows:
`{{ array | where_exp: "item", "item == 10" }}`
`{{ array | where_exp: "item", "item.field > 10" }}`
`{{ site.posts | where_exp: "post", "post contains 'field'" }}`
`{{ site.posts | where_exp: "post", "post.array contains 'giraffes'" }}`
This permits a variety of use cases, such as reported in: jekyll#4467,
jekyll#4385, jekyll#2787.
Update Rake and disable verbosity when running the specs because we have some
deprecated usage (for now -- however, see: jekyll/jekyll#4719) and because
Rouge, and Liquid throw out thousands (probably hyperbolic) of warnigns when the
specs are being ran. We need upstream to fix their problems while we fix ours
or we'll all have a bad day, not that we aren't already.
* origin/master: (65 commits)
Update history to reflect merge of #4703 [ci skip]
Update history to reflect merge of #4712 [ci skip]
Highlight the test code
Update history to reflect merge of #4640 [ci skip]
readded "env=prod"-condition
Update history to reflect merge of #3849 [ci skip]
Update history to reflect merge of #4624 [ci skip]
Update history to reflect merge of #4704 [ci skip]
Update history to reflect merge of #4706 [ci skip]
Checks for link file extension in tests
Updating assets documentation
Fix test teardown for cleaner.
Update history to reflect merge of #4542 [ci skip]
Add explanation of site variables in the example _config.yml
Use double quotes in the gemfile
Add test for creation of Gemfile by 'jekyll new'
Add comment about github-pages
Update history to reflect merge of #4533 [ci skip]
Ensure Rouge closes its div/figure properly after highlighting ends.
Add Site#config= which can be used to set the config
...
This removes the following warnings:
- /lib/jekyll/configuration.rb:151: warning: instance variable @default_config_file not initialized
- /lib/jekyll/converter.rb:12: warning: instance variable @highlighter_prefix not initialized
- /lib/jekyll/converter.rb:24: warning: instance variable @highlighter_suffix not initialized
- /lib/jekyll/converters/markdown.rb:9: warning: instance variable @setup not initialized
- /lib/jekyll/converters/markdown/kramdown_parser.rb:60: warning: instance variable @highlighter not initialized
- /lib/jekyll/frontmatter_defaults.rb:97: warning: shadowing outer local variable - path
- /lib/jekyll/plugin.rb:66: warning: instance variable @safe not initialized
- /lib/jekyll/regenerator.rb:147: warning: instance variable @disabled not initialized
- /test/test_convertible.rb:40: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_filters.rb:154: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_new_command.rb:84: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_site.rb:234: warning: assigned but unused variable - site
- /test/test_site.rb:240: warning: assigned but unused variable - site
- /test/test_site.rb:522: warning: assigned but unused variable - source
- /test/test_tags.rb:153: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_tags.rb:425: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_tags.rb:449: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_tags.rb:496: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_tags.rb:496: warning: instance variable @result not initialized
- /test/test_tags.rb:511: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_tags.rb:773: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_tags.rb:773: warning: instance variable @result not initialized
- /test/test_tags.rb:788: warning: ambiguous first argument; put parentheses or a space even after `/' operator
- /test/test_url.rb:66: warning: shadowing outer local variable - doc
- /lib/jekyll/url.rb:119:in `escape_path': warning: URI.escape is obsolete
Previously `titleize` used `capitalize!` which has the side effect of
returning `nil` for anything already starting with a capital letter. This
commit changes it to just `capitalize`.
Example, before:
A file "2016-01-01-This-is-a-title-with-Capitals.markdown" would return "Is A
Title With" for `post.title`
Example, after:
A file "2016-01-01-This-is-a-title-with-Capitals.markdown" will return "This Is A
Title With Capitals" for `post.title`
Tests added for `titleize_slug` in test_utils.rb
Fix problem introduced in 67f842546e
References #4525
Previously, even if the document permalink was a folder, it would look for
an extension on that. For example, if I have:
permalink: "/new-version-jekyll-v3.0.0/"
the output_ext would be ".0". Now, the output_ext honors the trailing
slash and will report based on the converters instead.
This reverts a bit of the work @willnorris had made to support
extensionless permalinks. Using the ‘permalink’ front matter will no
longer work as it must allow non-html extensions to be written.
Occasionally, extra spaces at the end of the YAML front matter prologue are
saved to a file and it goes missing without telling the user why. This
should simply accept those changes without any detriment to the user,
allowing anyone to add as many spaces as they like to the end of their
front matter prologues.
* akoeplinger-doctor-permalink-same-case-warning:
Added tests for new jekyll doctor warning
Incorporate code review feedback
Incorporate code review feedback
Add a Jekyll doctor warning for URLs that only differ by case
* Add support for SSL through command line switches.
* Add suppport for file/index.html > file.html > directory.
* Add support for custom-headers through configuration.
* Modernize and split up the serve.
* Add a few basic tests.
which works the same way as Dir.glob but seperating the input
into two parts ('dir' + '/' + 'pattern') to make sure
the first part('dir') does not act as a pattern.
Disable the feature as it's still not 100% working 100% of the time. Feature
can be re-enabled by specifying `full_rebuild: false` in the configuration
Fix the code coverage reporting when using `.bundle` to store my gems in
by having SimpleCov ignore that directory. Use of `.bundle` to store my
gems consolidates things since since that directory also holds the
bundler config file. It also keeps a `vendor` directory out of the
project tree for non-Rails projects. Simplecov was not ignoring that
directory though, which meant that the code coverage numbers I were
seeing locally were wrong (and very frightening). With this change, all
is right with the world once again. 😃
When a post does not contain an excerpt_separator, meaning the excerpt
includes the entire post, the excerpt should contain exactly the post
content.
This is desirable both from a correctness standpoint, that the excerpt
should not introduce any new content, and more practically to allow fast
and easy detection of whole-post excerpts in Liquid templates using
`post.excerpt == post.content`. A common use-case is deciding whether
to render "Read More" links on a page containing post excerpts.
This commit does exactly that. It avoids adding additional newlines to
the excerpt content when the excerpt includes the whole post and adds
tests to ensure that this behavior is correct and preserved going
forward.
Signed-off-by: Kevin Locke <kevin@kevinlocke.name>
Non-string input was being missed as a result of poor comparison.
Converting inputs to strings ensure numerical and boolean values are
properly compared.
Fixes#3911.
This enables files such as images and PDFs to show up in the same relative
output directory as other HTML and Markdown documents in the same collection.
It also enables static files to be hidden using defaults from _config.yml in
the same way that other documents in the same collection and directories may
be hidden using `published: false`.
Pat Hawks
Parker Moore
jekyllbot
Brint O'Hearn
Jordon Bedwell
Thomas Wood
Jussi Kinnula
Pierre Fenoll
Florian Thomas
Ben Balter
Jeff Kolesky
Shinn Kondo
Tom Fejfar
Saleem Rashid
timwis
Tony Garnock-Jones
Henry Goodman
atomicules
Alfred Xing
rebornix
Zshawn Syed
Pedro Euko
Nick Quaranto
Ducksan Cho
Tim Cuthbertson
Paul Robert Lloyd
XhmikosR
Matt Rogers
Kevin Locke
Fabio Niephaus
Veres Lajos
Robert Papp
Mike Bland
Rafael Picanço
Florian Weingarten