Dialyzer won't work

[si14]

There are at least 2 problems:

• mvar(_) -> pid() | atom() won't work; mvar(_T) is better
• resource_pool.erl:25: Polymorphic opaque types not supported yet

[rvirding]

There was a discussion on the first point recently on the mailing and bugs list. I don't agree with how dialyzer does it.

[si14]

@rvirding did that discussion ended up to something? Dialyzer is a great tool, but there are 2 things that killed all my attempts to verify somewhat large project: the one that is mentioned in this issue and the fact that Dialyzer can't handle polymorphic opaque types.

[rvirding]

Not really any more than the replies I got in the mailing-list. They
said that just having variables like I did meant that they were type
variables and type variables shouldn't be singletons. Why they allow
them as singletons if they start with '_' I don't know. I will
probably revert to having my "spec" in the comments the information
I need is the variable name, dialyzer doesn't need the specs and I
don't like cluttering up my code with unnecessary data.

This is not the main-stream view. :-)

Robert

On 23 February 2012 12:17, Dmitry Groshev
reply@reply.github.com
wrote:

@rvirding did that discussion ended up to something? Dialyzer is a great tool, but there are 2 things that killed all my attempts to verify somewhat large project: the one that is mentioned in this issue and the fact that Dialyzer can't handle polymorphic opaque types.

---

Reply to this email directly or view it on GitHub:
#14 (comment)