[WIP] Generic script crate

[hrvolapeter]

Separation is based on adding types to TypeHolderTrait which contains
associated types with traits that script crate actually needs. This allows
using static methods and Sized types.

Affects servo in two ways:

• generic structs do not compile in script crate but are left for later monomorphization
• dom_structs itself are moved to separate crate

TODO:

• tests
• performance
• polyshing

Fixes #1799

r? @jdm

---

• ./mach build -d does not report any errors
• ./mach test-tidy does not report any errors
• These changes fix Split up the script crate for build speed #1799 (github issue number if applicable).

• There are tests for these changes OR
• These changes do not require tests because _____

---

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 @jdm (or someone else) soon.

[highfive]

Heads up! This PR modifies the following files:

• @asajeffrey: components/script/test.rs, components/script/dom/textencoder.rs, components/script/dom/htmlbrelement.rs, components/script/dom/audiodestinationnode.rs, components/script/dom/abstractworkerglobalscope.rs and 345 more
• @wafflespeanut: python/tidy/servo_tidy_tests/ban-domrefcell.rs
• @edunham: python/tidy/servo_tidy_tests/ban-domrefcell.rs
• @paulrouget: components/servo/lib.rs, components/servo/Cargo.toml
• @KiChjang: components/script/test.rs, components/script/dom/textencoder.rs, components/script/dom/htmlbrelement.rs, components/script/dom/audiodestinationnode.rs, components/script/dom/abstractworkerglobalscope.rs and 345 more

[highfive]

Warning

• These commits modify unsafe code. Please review it carefully!

[jdm]

@bors-servo try

[bors-servo]

⌛ Trying commit e668042 with merge 32f598e...

[bors-servo]

💔 Test failed - linux-rel-css

[jdm]

Results from local measurements using non-incremental builds:

master:
./mach build -d; touch components/script/dom/servoparser/html.rs && ./mach build -d: 3:49m

PR:
./mach build -d; touch components/script/fetch.rs && ./mach build -d: 4:18m
./mach build -d; touch components/script_servoparser/html.rs && ./mach build -d: 1:59m

[jdm]

Results from local measurements using incremental builds:

master:
./mach build -d; touch components/script/dom/servoparser/html.rs && ./mach build -d: 3:19m

PR:
./mach build -d; touch components/script/fetch.rs && ./mach build -d: 2:56m
./mach build -d; touch components/script_servoparser/html.rs && ./mach build -d: 1:48m

[bors-servo]

☔ The latest upstream changes (presumably #20678) made this pull request unmergeable. Please resolve the merge conflicts.

[jdm]

@retep007 Does this PR contain the most recent changes you have been making? I might try to keep investigating this week.

[hrvolapeter]

@jdm It contains all latest changes, however I can rebase it for you. Also I don't really know how to solve the ICE problem, no response from rust-lang so far. Those const functions are used to generate const variables and one of them can't be replaced so easily.

[jdm]

Ah, that's good to know. Have you tried making duplicating the function and making a non-const version?

[hrvolapeter]

Last static fn which I wasn't able to remove is Guard::newwhich I think is beyond manual replacement. But maybe we could try lazy_static?

[jdm]

I'm not sure I understand - what's the problem with Guard::new, exactly?

[hrvolapeter]

@jdm I solved the const fn issue and rebased I get window is undefined. I guess that it origniates from mozjs which is checking for null pointer provided by servo. Maybe something wrong with initialization?