Support for templated impl declarations #2700
Conversation
github-actions
bot
added
dependent
|
This PR/issue depends on:
|
The strategy that we use for now to support template instantiation is to check the impl declaration as if it were a generic, but to defer all checking of the impl definition until we see a use in which all template parameters have arguments. At that point, we clone the impl definition and type-check the whole thing, with constant values set on the template parameters corresponding to the given arguments. No caching of template instantiations is performed yet; each time we form a reference to a template instantiation, we instantiate it afresh.
zygoloid
force-pushed
the
explorer-template
branch
from
3b5816c to
82679b4
Compare
This currently doesn't work properly.
| } | ||
|
|
||
| impl forall [template T:! type] T as GetX { | ||
| // CHECK:STDERR: COMPILATION ERROR: {{.*}}/explorer/testdata/template/fail_no_member.carbon:[[@LINE+1]]: member access, unexpected i32 in self.x |
There was a problem hiding this comment.
This error is kind of confusing; where in self.x is the unexpected i32? This looks like you're falling through to the "unhandled" case of TypeCheckExp, not a deliberately written error message.
There was a problem hiding this comment.
i.e., handling this should probably be part of this PR, since it seems you may be adding code that deliberately reaches an unhandled code path.
There was a problem hiding this comment.
This isn't new in this PR: https://carbon.godbolt.org/z/f87nc4e4f
(It would be nice to include a note saying that we're performing an instantiation with T=i32; there's a TODO in the code for that already.)
There was a problem hiding this comment.
Do you think this would make sense for a "good first issue"?
There was a problem hiding this comment.
Sure, filed as #2705.
Co-authored-by: Richard Smith <richard@metafoo.co.uk>
The strategy that we use for now to support template instantiation is to check the impl declaration as if it were a generic, but to defer all checking of the impl definition until we see a use in which all template parameters have arguments. At that point, we clone the impl definition and type-check the whole thing, with constant values set on the template parameters corresponding to the given arguments.
No caching of template instantiations is performed yet; each time we form a reference to a template instantiation, we instantiate it afresh. We also don't implement the name lookup rule from #949 yet; lookups during template instantiation look only in the actual type and not in the constraint.
Depends on #2699