Changelog.md

Changelog

This document tracks features added, behaviour changes, and bugs fixed in Serialize. Generally, Serialize follows Semver; unless you’ve been relying on buggy behaviour, you should have no problem upgrading to new minor and patch versions.

2.4.0

• Serialize now recognizes when to remove markups while converting elements. For example, <p><em style="font-style: normal;">Foo</em></p> will have no markups once serialized.
• Serialize is now fully subclassable.
• Invalid markups can no longer be added. Invalid markups are those whose start point is before or equal to their end point.
• When adding a link markup, other links will be truncated or removed so as to not overlap with the markup being added. Previously, it was possible to create overlapping links with different hrefs; this is no longer the case.
• Markups are automatically merged after adding them. There is no longer any reason to call Serialize#mergeAdjacent after adding markups.

2.3.0

• Serialize now accounts for styles on the given elements. For example, the serialization of <p style="font-weight: bold;">Foo</p> will have a bold markup spanning the length of the serialization’s text.
• Added the Serialize#addMarkup method.

2.2.2

• Serialize now recognizes only text and element nodes when creating the serialization. All other nodes (e.g. comment nodes) are ignored.

2.2.1

• Serialize#replace now returns a new serialization.
• Fixed a minor bug that caused Serialize#append to return undefined when called with an empty string or other falsy value.

2.2.0

• Defines a custom toString method that returns the outerHTML of Serialize#toElement.

2.1.3

• Fixed a minor bug where Serialize#removeMarkup did not return the context, as it is supposed to.

2.1.2

• Fixed a relatively serious issue whereby markups of type Serialize.types.link did not conserve their href property when calling methods such as substr and append.

2.1.1

• Fixed the href property of Serialize.types.link markups not being preserved when part of such a markup is truncated.

2.1.0

• Allowed passing a string to Serialize#append. The string is appended to the serialization’s text and any markups that previously terminated at the end of the serialization are extended so that they terminate at the end of the newly extended serialization.

2.0.1

• Fixed a bug whereby Serialize#replace didn’t pay attention to special replacement patterns (more).

2.0.0

• The length property of Serializations is now automatically updated when one manually changes a serialization’s text. For example:

  var p = document.createElement('p')
  p.textContent = 'foo'
  
  var s = new Serialize(p)
  console.log(s.length) // 3
  
  s.text += '\n'
  console.log(s.length) // 4

---

1.0.0 – 1.3.0

Versions 1.x.y are backports of their respective 2.x.y patches, respectively. In theory, the 1.0 releases had wider browser support; after turning on cloud testing, however, it was determined that Serialize had some serious compatibility issues with Internet Explorer versions 8 and earlier which were not worth fixing.

1.0.0

• No major changes. This release was mainly to avoid odd behaviour in the Semver spec regarding 0.x.y releases.

0.5.0

• Added the Serialize#removeMarkup method.

0.4.0

• Added the Serialize#equals method to succintly compare two serializations.

0.3.3

• Added the Serialize.fromText function.

0.3.2

• Fixed Serialize’s CommonJS compatibility; previously, the bundled, minified version of Serialize was served when require-ing the package.

0.3.1

• Addresses some style nits.

0.3.0

• Added the Serialize#append method.

0.2.0

• Added the Serialize#substr method.

0.1.1

• Changed Serialize.fromJSON; the only mandatory properties when creating an instance of Serialize from a JSON object are now text and type.

0.1.0

• Adds the Serialize#replace method.