Don't re-encode already encoded post titles

[notnownikki]

Data that comes from the Posts collection is already encoded, ready to output safely, so we set the inner html to avoid encoding twice.

Fixes #2389

[aduth]

It would be best for us to avoid dangerouslySetInnerHTML if at all possible if alternatives exist (hence its name).

In this case, there are other options available for unescaping entities:

• https://lodash.com/docs/4.17.4#unescape (only handles a few entities)
• https://www.npmjs.com/package/he (a very large module, probably too large)
• https://github.com/Automattic/wp-calypso/blob/master/client/lib/formatting/decode-entities/browser.js (can be an expensive operation, might ought to be memoized or shortcut)

[notnownikki]

@aduth lodash doesn't unescape everything we need, and I agree that he is too big... I'll try creating a textarea to use for decoding, and reuse the element so we're not doing element creation for every post.

[codecov]

Codecov Report

Merging #2407 into master will increase coverage by 0.59%.
The diff coverage is 90%.

@@            Coverage Diff             @@
##           master    #2407      +/-   ##
==========================================
+ Coverage    25.9%   26.49%   +0.59%     
==========================================
  Files         157      161       +4     
  Lines        4853     5170     +317     
  Branches      822      903      +81     
==========================================
+ Hits         1257     1370     +113     
- Misses       3035     3180     +145     
- Partials      561      620      +59

Impacted Files	Coverage Δ
blocks/library/latest-posts/index.js	10% <ø> (ø)	⬆️
utils/entities.js	90% <90%> (ø)
blocks/library/code/index.js	40% <0%> (-10%)	⬇️
blocks/library/paragraph/index.js	39.28% <0%> (-7.78%)	⬇️
blocks/library/preformatted/index.js	37.5% <0%> (-6.95%)	⬇️
blocks/library/table/index.js	30% <0%> (-6.37%)	⬇️
blocks/library/quote/index.js	13.11% <0%> (-1.18%)	⬇️
blocks/library/heading/index.js	22.85% <0%> (-0.96%)	⬇️
blocks/editable/index.js	10.36% <0%> (-0.61%)	⬇️
blocks/library/list/index.js	6.55% <0%> (-0.42%)	⬇️
... and 12 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 77620b2...7433aca. Read the comment docs.

[aduth]

Do you think this might be a utility worth generalizing? Maybe moving to utils/decode-entities.js and including some unit tests? There may even be some overlap between this and what's being proposed by @adamsilverstein as an @wordpress/sanitize utility at WordPress/packages#12 (which takes it a step further to strip HTML tags).

While it wouldn't be wise to generalize before there's a need, we've already seen some use of Lodash's _.unescape that could be better served by something like this:

gutenberg/blocks/library/categories/index.js

Lines 117 to 123 in a0714e2

	renderCategoryName( category ) {
	if ( ! category.name ) {
	return __( '(Untitled)' );
	}
	return unescape( category.name ).trim();
	}

And if Calypso's usage can serve as evidence, it's a very common utility:

https://github.com/Automattic/wp-calypso/search?utf8=%E2%9C%93&q=decodeentities&type=

[notnownikki]

Yeah... ok, happy to generalize this :)

[notnownikki]

Added an 'entities' module that has 'decodeEntities'.

[aduth]

Looks good 👍

You may consider reaching out on WordPress/packages#12 to see if we can reconcile requirements here, since I think the new utility falls well under what's being proposed there.