-
Notifications
You must be signed in to change notification settings - Fork 108
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[WIP] Various IDE improvements and new features #1334
base: master
Are you sure you want to change the base?
Conversation
(this commit is not compiling however!)
… by simply creating a fake error in the case that no_verify and show_ide_info are set.
…ps, even though they introduce some new problems
…ocket. Report quantifier instantiations split by methods.
…is was cached to clients.
In general, when skipping verification, we introduced a so called fake_error to make sure the compiler does not just cache this as successful verification and then skip later verifications even if that's not intended. Same issue occured for selective verification, when verifying a single method that is "correct", this would be cached. Now we throw a fake error in this case too, to avoid this.
Prusti Assistant offers a few new features, and it checks for this version for them to be available.
prusti-utils/src/config.rs
Outdated
settings.set_default("smt_qi_profile", true).unwrap(); | ||
settings.set_default("smt_qi_profile_freq", 10000).unwrap(); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Will these be passed to Z3 when not using the IDE? And in such a case, will the messages not be reported to the user? If so, QI profiling should be disabled by default.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
(After looking at how they are used in the other file:) please make these be Option<_>
, such that they are only passed as arguments when a value is set. The IDE can set these if they are needed.
} | ||
None => { | ||
// function is a Trait (or something else?) | ||
// are there other cases for AssocFns? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't think so.
// can a single generic type have the same traitbound | ||
// more than once with different arguments? | ||
// I would imagine yes.. may be a todo |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Like T: Foo<i32> + Foo<u32>
? Yes, this is possible.
Also add some debug statements.
In 2 places we throw fake errors to avoid caching of successes when we should not. This lead to problems with local dependencies, since they are verified too and the fake_error ended up being thrown too early. This bug is resolved now.
First of all we separated trait bounds from generic args, since for example for a function signature within an impl block, the trait bounds at the function level can still refer to the generic args defined in the impl block. Secondly, we had a problem if a generic had duplicate traitbounds, with different generics, e.g.: `T: Foo<u32> + Foo<String>`, this is now handled correctly, under the assumption that the predicates occurr in a certain order, i.e. projections come after their respective trait. So far this seems to be always the case, if it's not I don't see how this problem can be solved.
Ported change from PR viperproject#1334
Port basic functionality for the new prusti-server work flow according to PR viperproject#1334 .
Querying signatures, spans of call contract collection and call finding does not work yet. PR: viperproject#1334
Defpath only verification does not work since test_entrypoint currently just puts the entire program into the request regardless of the verification tasks at hand. Update the flags in the documentation. PR: viperproject#1334
Changes to support various new features in prusti-assistant.
@Aurel300
@jthomme1