add more comprehensive tests for incr. comp.

[nikomatsakis]

Test various changes one can do to structs.

r? @michaelwoerister

[michaelwoerister]

Interesting...

[michaelwoerister]

@bors r+

Do we have some documentation somewhere that describes the dependency graph node labels? I.e. how do we expect people to know that it's specifically TypeckItemBody (for example) that should be invalidated by a given change?

[bors]

📌 Commit fa64021 has been approved by michaelwoerister

[nikomatsakis]

There is no documentation aside from the code. But obviously these are unit
tests, so familiar with the code seems like a reasonable expectation to me.

On Tue, May 24, 2016 at 6:02 PM, Michael Woerister <notifications@github.com

wrote:

@borshttps://github.com/bors r+

Do we have some documentation somewhere that describes the dependency
graph node labels? I.e. how do we expect people to know that it's
specifically TypeckItemBody (for example) that should be invalidated by a
given change?

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#33851 (comment)

[bors]

💡 This pull request was already approved, no need to approve it again.

• There's another pull request that is currently being tested, blocking this pull request: Save metadata even with -Z no-trans (e.g. for multi-crate cargo check). #33602

[bors]

📌 Commit fa64021 has been approved by nikomatsakis

[Manishearth]

Travis fails

[bors]

⌛ Testing commit fa64021 with merge 2367fc4...

[bors]

💔 Test failed - auto-win-gnu-32-opt-rustbuild

[nikomatsakis]

closing in favor of a new PR that incorporates a fix.