797472 - Content model for live samples

Status: RESOLVED FIXED

(developer.mozilla.org Graveyard :: Wiki pages, defect)

Description

[Les Orchard [:lorchard]]

Need to create new Django content models for live samples. Going from the jsFiddle model, we need at least these fields:

* parent document (since sheppy told me we can associate samples with a page)
* caption
* html
* css
* js
* author
* modified, created

Thinking we also want to track revisions, so need to somewhat mirror the doc/rev relationship too.

Comment 2

[Jean-Yves Perrier [:teoli]]

I confirm that we will need to track revisions (and author should be linked to revision).

Comment 3

[Les Orchard [:lorchard]]

Rethinking this in context of louisremi's experimental PR:

https://github.com/mozilla/kuma/pull/644

Rather than develop a completely separate collection of content objects devoted to code samples, maybe a live sample could be authored right inline on a wiki document as a group of code listings (ie. a <div> with HTML / CSS / JS sub-tabs).

Then, I can build the server-side tooling to extract the cluster of code listings for a given sample ID, combine them into a working HTML document, and serve that up at an iframe-ready URL (bug 797474)

louisremi's code does the above entirely with client-side JS, a sandboxed iframe, and a data: URL. But, there's some concern about browser compatibility, which is what makes me think we might need a server-side solution instead

Comment 4

[Les Orchard [:lorchard]]

(Oh, but one of the really important things: Managing code samples inline and extracting the code to build a page means that edits to a code sample would appear just like an edit to prose, with respect to change feeds and dashboards.)

Comment 5

[Les Orchard [:lorchard]]

Almost merged: https://github.com/mozilla/kuma/pull/645

Comment 6

[[github robot]]

Commits pushed to master at https://github.com/mozilla/kuma

https://github.com/mozilla/kuma/commit/1ec9f1f45632e1133d26d606ab6094b1ea0d94ee
bug 797472: Tooling to extract a code sample and parts from a document

https://github.com/mozilla/kuma/commit/084ccaca0eab72c086d350bf6ef13805c41fa2c2
bug 797472: Model additions to extract code sample parts from a Document object

https://github.com/mozilla/kuma/commit/0ccc3501b1e3ffede05f6e1164c97b6df2d49bf9
bug 797472: Tweak to use jinja2.escape in tests

Comment 7

[Les Orchard [:lorchard]]

Going to say that covers the initial stab at content models for live samples (or lack thereof). Code listings are managed as inline content on documents, and extracted to produce a live sample.