Changelog

Hickory: parse HTML into Clojure data structures.

0.7.8 (2026-11-09)

• #98: as-hiccup converts the element after an empty script or style element (@jacobemcken)
• #94: parse accepts a String, File or Path without reflection and throws on other input (@erdos)
• Bump jsoup to 1.22.2

0.7.7 (2025-03-13)

• #92: Suppress the reflection warning in parse

0.7.6 (2025-03-11)

• Bump jsoup to 1.19.1

0.7.5 (2024-10-16)

• Add parent and ancestor selectors
• Fix find-in-text inside a child selector

0.7.4 (2023-12-22)

• Remove the quoin dependency
• Split class names on any whitespace
• Fix ClojureScript advanced compilation

0.7.3 (2022-11-20)

• Upgrade ClojureScript

0.7.2 (2022-11-20)

• Move to clj-commons: org.clj-commons/hickory
• Parse deeply nested HTML without a stack overflow
• Support Node.js in ClojureScript
• Bump jsoup to 1.15.2

0.7.1 (2017-03-30)

• Add the up-pred zipper function (@tkocmathla)

0.7.0 (2016-10-30)

Thanks to Ricardo J. Méndez for the following updates.

• Removed dependency on cljx, since it was deprecated in June 2015.
• Converted all files and conditionals to cljc.
• Moved tests to cljs.test with doo, since cemerick.test was deprecated over a year ago.
• Updated Clojure and ClojureScript dependencies to avoid conflicts.
• Updated JSoup to 1.9.2, which should bring improved parsing performance.

0.6.0 (2016-01-04)

• Updated JSoup to version 1.8.3. This version of JSoup contains bug fixes, but slightly changes the way it handles HTML: some parses and output might have different case than before. HTML is still case-insensitive, of course, but Hickory minor version has been increased just in case. API and semantics are otherwise unchanged.

0.5.4

• Fixed project dependencies so ClojureScript is moved to a dev-dependency.

0.5.3

• Minor bug fix to accommodate ClojureScript's new type hinting support.

0.5.2

• Updates the Clojurescript version to use the latest version of Clojurescript (0.0-1934).

0.5.1

• Added has-child and has-descendant selectors. Be careful with has-descendant, as it must do a full subtree search on each node, which is not fast.

0.5.0

• Now works in Clojurescript as well, huge thanks to Julien Eluard for doing the heavy lifting on this.
• Reorganized parts of the API into more granular namespaces for better organization.
• Added functions to convert between Hiccup and Hickory format; note that this conversion is not always exact or roundtripable, and can cause a full HTML reparse.
• Added new selector, element-child, which selects element nodes that are the child of another element node.
• Numerous bug fixes and improvements.

0.4.1

• Adds a number of new selectors and selector combinators, including find-in-text, precede-adjacent, follow-adjacent, precede and follow.

0.4.0

• Adds the hickory.select namespace with many helpful functions for searching through hickory-format HTML documents for specific nodes.

0.3.0

• Provides a more helpful error message when hickory-to-html has an error. Now requires Clojure 1.4.

0.2.3

• Fixes a bug where hickory-to-html was not html-escaping the values of tag attributes.

0.2.2

• Fixes a bug where hickory-to-html was improperly html-escaping the contents of script/style tags.

0.2.1

• hickory-to-html now properly escapes text nodes
• text nodes will now preserve whitespace correctly

0.2.0

• Adds a second parsed data format, explained in the README. To support this, the API for parse and parse-fragment has been changed to allow their return values to be passed to functions as-hiccup or as-hickory to determine the final format. Also added are zippers for both Hiccup and Hickory formats.