# Paste **MarkDown**

[ellatrix]

This adds MarkDown parsing to the paste handling if the posted content is plain text. It won't parse anything if you don't paste from a plain text editor.

This is a WIP... Still need to have a look how to handle inline text better and at the parsers available. Any suggestions? Showdown is is used by Ghost so I picked that for now.

[aduth]

What's the impact on bundle size by pulling in showdown ? Also curious if this is something best left for plugin territory...

[ellatrix]

10.7 kb gzipped... I'm fine either way, I think this is a nice surprise enhancement. I mainly want to show here that it's easily possible to add, and that not everything would run through the parser, only plain text.

Also this may be a place then where we would want to add a filter for plugins.

[dmsnell]

For cross-reference: https://github.com/Automattic/simplenote-electron/blob/77b4c41d6eea1e8200e2eb8de8b8bef7d90203e1/lib/note-detail.jsx#L15

[codecov]

Codecov Report

Merging #2531 into master will decrease coverage by 0.03%.
The diff coverage is 22.22%.

@@            Coverage Diff             @@
##           master    #2531      +/-   ##
==========================================
- Coverage   33.69%   33.65%   -0.04%     
==========================================
  Files         185      185              
  Lines        5594     5615      +21     
  Branches      976      980       +4     
==========================================
+ Hits         1885     1890       +5     
- Misses       3141     3155      +14     
- Partials      568      570       +2

Impacted Files	Coverage Δ
blocks/library/code/index.js	57.14% <ø> (ø)	⬆️
blocks/library/preformatted/index.js	50% <ø> (ø)	⬆️
blocks/api/paste/utils.js	82.97% <0%> (-14.53%)	⬇️
blocks/api/paste/test/integration/index.js	100% <100%> (ø)	⬆️
blocks/editable/index.js	10.5% <20%> (+0.18%)	⬆️
blocks/api/paste/index.js	79.31% <40%> (-8.69%)	⬇️
components/form-file-upload/index.js	66.66% <0%> (-4.77%)	⬇️

---

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 a93c553...aa73265. Read the comment docs.

[ellatrix]

Okay, so this will now do two things.

1. We're going to look at the plain text the clipboard API is giving us if the HTML is plain. If the source (editor) is plain text, this will be more reliable than any HTML made up by the browser.
2. If we have some reliable plain text, then we convert any markdown in it. If the user pastes HTML, this will never run.

The PR uses showdown, but we can easily use anything else later of course.

[ellatrix]

Merged after 👍 form @mtias and @dmsnell.

[aduth]

We should have updated package-lock.json here from the added dependency. npm install in a clean clone while running npm 5+ will introduce changes.

[aduth]

Updated in ff1ef2b

[ellatrix]

Thanks @aduth! Sorry about that, wasn't aware of package-lock.

[aduth]

No worries, since we don't explicitly target the version of Node that comes bundled with npm5+ (until LTS release in October), we maybe shouldn't have the package-lock.json at all, but the niceties of 5+ are enough for me to want to run it anyways 😄

[youknowriad]

@aduth The package-lock.json locks only the dependencies of the dependencies since 5.1.0 which makes it useless IMO. (it's not the same as shrinkwrap or yarn.lock)

[aduth]

package-lock.json locks only the dependencies of the dependencies since 5.1.0 which makes it useless IMO

That still seems like a pretty good benefit on its own, if you ask me.

Part of it is that it's automatic, so it becomes a nuisance to be merely running npm install and having changes introduced. If we really didn't care, I'd want to find a way to turn that off.

But then, I saw what you'd referred to come past, but I'd have to think it still provides some benefit; surely they wouldn't make the thing useless? 😄 I'll have to do some reading over the weekend.