Parse srcset algo

[akhan7]

Implementing parse a sourceset algorithm

---

This change is

[highfive]

Thanks for the pull request, and welcome! The Servo team is excited to review your changes, and you should hear from @mbrubeck (or someone else) soon.

[highfive]

Heads up! This PR modifies the following files:

• @KiChjang: components/net_traits/image_cache_thread.rs, components/net_traits/image_cache_thread.rs, components/script/dom/webidls/HTMLImageElement.webidl, components/script/dom/htmlimageelement.rs, components/net/image_cache_thread.rs

[highfive]

Warning

• These commits modify net and script code, but no tests are modified. Please consider adding a test!

[mbrubeck]

r? @jdm

[jdm]

Note: I keep changing the labels because I use them as a TODO list for myself. When it's time for me to actually review the code, please speak up :)

[bors-servo]

☔ The latest upstream changes (presumably #10767) made this pull request unmergeable. Please resolve the merge conflicts.

[jdm]

I strongly recommend writing unit tests in the same manner as I described at #10827 (comment), by the way!

[akhan7]

@jdm : We've reached till step 9 of parse a source set algorithm, the test tidy passes and the build is successful. Can you please check if the algo written is conceptually correct up till this step?

[jdm]

As far as I can tell, the algorithm makes sense as written. Tests would verify that gut feeling, of course ;) You can implement the EOF condition by matching against the current state after the loop is complete, since that means that the final character was encountered.

[jdm]

I've noted this work as part of #11416. Thank you for doing it; we have learned more about what is required for implementing this. The lack of tests means that this isn't appropriate to merge right now; we'll return to it as part of #11416.