Skip to content

Latest commit

 

History

History
290 lines (170 loc) · 8.94 KB

CHANGELOG.md

File metadata and controls

290 lines (170 loc) · 8.94 KB
Raw

6.5.10

  • Be optimistic in versions of govuk_publishing_components and i18n allowed (PR#200)

6.5.9

  • Adjust footnote markup to accommodate multiple references (PR#198)

6.5.8

  • Customise footnote markup for accessibility (PR#192)

6.5.7

  • Update GOV.UK Publishing components to version to 23.0.0 or greater

6.5.6

  • Update Kramdown version to 2.3.0 or greater

6.5.5

  • Prevent links in table headers being stripped (PR#187)

6.5.4

6.5.3

  • Use button component for buttons (PR#176)

6.5.2

  • Allow data attributes on div tags (PR#173)

6.5.1

  • Change unicode testing characters after external gem change
  • Move from govuk-lint to rubocop-govuk
  • Allow version 6 of actionview

## 6.5.0

  • Allow data attributes on links

6.4.0

  • Add table heading syntax that allows a table cell outside of thead to be marked as a table heading with a scope of row. (PR#161)

6.3.0

  • Unicode characters forbidden in HTML are stripped from input
  • Validation is now more lenient for HTML input

6.2.1

  • Update warning callout label text from 'Help' to 'Warning'

6.2.0

  • Remove experimental status on AttachementLink:attachment-id and Attachement:attachment-id
  • Deprecate embed:attachments:inline:content-id

6.1.1

  • Fix wrapping AttachmentLink:attachment-id in a paragraph when used inline

6.1.0

  • Remove [embed:attachments:content-id] this isn't used by any apps and has never worked
  • Add dependency on govuk_publishing_components
  • Add new AttachementLink:attachment-id extension and mark as experimental
  • Add new Attachement:attachment-id extension and mark as experimental
  • Blockquote quote remover is now more forgiving to spaces before or after quote characters

6.0.0

  • BREAKING CHANGE: Input is sanitized by default, to use unsafe HTML initialize with a sanitize option of false
  • Allow sanitize option on remove invalid HTML from source input
  • BREAKING CHANGE: Remove to_sanitized_html method in favour of sanitize option on initialize
  • BREAKING CHANGE: Remove to_sanitized_html_without_images as no apps use this anymore
  • BREAKING CHANGE: Remove CLI usage

5.9.1

  • Don't render [Image: {file-name}] within a paragraph to avoid invalid HTML

5.9.0

  • Add image credits to embedded images

5.8.0

  • Add new Image:image-id extension and deprecate embed:attachments:image:content-id

5.7.1

  • Include locale, config and asset files in the gem distribution

5.7.0

  • Update ActionView to 5.x
  • Fix translation files not loading when gem is used in an app
  • Change the data inputted into Contacts to match contacts schema #130

5.6.0

  • Update sanitize version to 4.6.x #127

5.5.0

  • Ignore links with blank or missing hrefs when extracting links from a document with Govspeak::Document#extracted_links #124

5.4.0

  • Add an optional website_root argument to Govspeak::Document#extracted_links in order to get all links as fully qualified URLs #122

5.3.0

  • Add a link extraction class for finding links in documents #120

5.2.2

  • Fix rendering buttons with inconsistent linebreaks seen in publishing #118

5.2.1

  • Fix validation to make sure buttons are considered valid
  • Only allow buttons to be used on new lines, not when indented or inline within text (useful for guides) #116

5.2.0

  • Add button component for govspeak #114 see README for usage

5.1.0

  • Update Kramdown version to 1.15.0

5.0.3

  • Fix matching links/attachments/contacts by regex to use equality #105

5.0.2

  • Loosen ActionView dependency to allow use with Rails 5 #99

5.0.1

  • Move presenters into the Govspeak namespace #93
  • Embedded links now will automatically be marked with rel="external" #96

5.0.0

  • Update Kramdown version to 1.12.0
  • Add pry-byebug to development dependencies
  • Ability to run Govspeak as a binary from command line #87
  • Uses hashes the primary interface for options to commands #89
  • Adds the [embed:attachments:image:%content_id%] extension #90
  • Renders incorrect usages of embedding content as empty strings rather than outputting markdown 91

4.0.0

  • Drop support for Ruby 1.9.3
  • Update Ruby to 2.3.1
  • Adds support for the following items for feature parity with whitehall:
    • {barchart}
    • [embed:attachments:%content_id%]
    • [embed:attachments:inline:%content_id%]
    • [embed:link:%content_id%]
    • [Contact:%content_id%]
  • Changes blockquote rendering to match whitehall #81

3.7.0

  • Update Addressable version from 2.3.8 to 2.4.0

3.6.2

  • Fix bug with link parsing introduced in Kramdown 1.6.0 with the "inline attribute lists" feature which clashed with our monkey patch #75

3.6.1

  • Update minimum Kramdown version from 1.5.0 to 1.10.0 (changelog)
  • Allow table columns to be left, right or centre aligned using the standard markdown pattern provided by Kramdown

3.6.0

  • Yanked, see 3.6.1 which includes fix

3.5.2

3.5.1

  • Continue to support non-strict URIs in links on Ruby 2.2.x. See #57

3.5.0

  • Add {stat-headline}*10m* big numbers{/stat-headline} markdown for HTML publications
  • aside elements are now allowed through the sanitization process.
  • Update Ruby from 1.9.3 to 2.1

3.4.0

  • Increase the version of Kramdown to 1.5.0. This allows compatibility with Jekyll and the GitHub pages gem.

3.3.0

  • Relax Nokogiri dependency to 1.5.x rather than 1.5.10. This allows Govspeak to work with Rails 4.2 and greater.

3.2.0

  • span elements are now allowed through the sanitization process.

3.1.1

  • Fix a bug where address information could get mixed in with preceding paragraph text.

3.1.0

  • Allow a call to #valid? on a Govspeak::Document to accept validation options.

3.0.0

  • Add an allowed_image_hosts options to HtmlValidator (and HtmlSanitizer)
  • BREAKING CHANGE: Added the $EndLegislativeList tag which allows line breaks in LegislativeLists.

## 2.0.2

  • Fix a bug with the HtmlValidator to do with kramdown now respecting character encodings of input data.

2.0.1

  • Upgrade to newest kramdown, which fixes a number of bugs, including rendering multiple footnote references and handling underscores in attachment titles

2.0.0

  • Upgrade sanitize dependency to 2.1.0
    • Now allows address, bdi, hr and summary elements by default.
    • Allows colons in IDs
    • BREAKING CHANGE: This changes the validation rules of HtmlSanitizer.
  • Stop duplicate entries in the sanitization config

1.6.2

  • Fix a bug with parsing of $LegislativeList and $PriorityList with \r\n line endings.

1.6.1

  • Fix a bug with parsing of $LegislativeList and $PriorityList so that they are not matched when immediately preceeded by other text.

1.6.0

  • Add $LegislativeList to allow editors to insert lists with custom markers.

1.5.4

  • Stop steps list regex matching newline.

1.5.3

  • Nested Govspeak blocks are now parsed recursively using Govspeak. Among other things, this fixes the display of external links in callout blocks.

1.5.2

  • Fixed over-eager step list matching.

1.5.1

  • Fixed newline matching and operating over multiple list blocks.

1.5.0

  • Added $PriorityList:x construct to class the first x list items with primary-item.
  • Fixed rcov dependency.

1.4.0

  • Added #structured_headers method to provide heirarchically structured headers extracted from markdown text heading tags.