coq-club AT inria.fr
Subject: The Coq mailing list
List archive
- From: Johannes Waldmann <johannes AT albapasser.de>
- To: frederic.blanqui AT loria.fr
- Cc: Coq Club <coq-club AT pauillac.inria.fr>
- Subject: Re: [Coq-Club] hierarchical library names and directory layout
- Date: Mon, 09 Jun 2008 21:26:43 +0200
- List-archive: <http://pauillac.inria.fr/pipermail/coq-club/>
I was starting with the naive expectation that "Require Foo.Bar.Baz"
leads to reading the file Foo/Bar/Baz.v(o)
note that such a policy may create new ambiguities [...]
I just noticed a similar problem in Java:
new Foo.Bar.Baz () could mean either of
package Foo; class Bar { static class Baz { } }; // in Foo/Bar.java
package Foo.Bar; class Baz { }; // in Foo/Bar/Baz.java
it is in fact an error (flagged by the compiler)
to compile the second version if the first version is present.
(but not the other way round - strangely)
I think this refusal to compile is right (much better
than compiling and then complaining about ambiguous imports)
Best regards, Johannes Waldmann.
- [Coq-Club] hierarchical library names and directory layout, Johannes Waldmann
- Re: [Coq-Club] hierarchical library names and directory layout,
frederic . blanqui
- Re: [Coq-Club] hierarchical library names and directory layout, Johannes Waldmann
- Is there -pack option? (was [Coq-Club] hierarchical library names and directory layout),
Keiko Nakata
- Re: Is there -pack option? (was [Coq-Club] hierarchical library names and directory layout),
frederic . blanqui
- [Coq-Club] Re: Is there -pack option?, Keiko Nakata
- Re: Is there -pack option? (was [Coq-Club] hierarchical library names and directory layout),
frederic . blanqui
- Re: [Coq-Club] hierarchical library names and directory layout,
frederic . blanqui
Archive powered by MhonArc 2.6.16.