3 Commits

Author SHA1 Message Date
Alex Crichton
4e9e091e91 syntax: Tighten search paths for inner modules
This is an implementation of RFC 16. A module can now only be loaded if the
module declaring `mod name;` "owns" the current directory. A module is
considered as owning its directory if it meets one of the following criteria:

* It is the top-level crate file
* It is a `mod.rs` file
* It was loaded via `#[path]`
* It was loaded via `include!`
* The module was declared via an inline `mod foo { ... }` statement

For example, this directory structure is now invalid

    // lib.rs
    mod foo;

    // foo.rs
    mod bar;

    // bar.rs;
    fn bar() {}

With this change `foo.rs` must be renamed to `foo/mod.rs`, and `bar.rs` must be
renamed to `foo/bar.rs`. This makes it clear that `bar` is a submodule of `foo`,
and can only be accessed through `foo`.

RFC: 0016-module-file-system-hierarchy
Closes #14180

[breaking-change]
2014-05-17 01:01:47 -07:00
Alex Crichton
de7d143176 Fix existing privacy/visibility violations
This commit fixes all of the fallout of the previous commit which is an attempt
to refine privacy. There were a few unfortunate leaks which now must be plugged,
and the most horrible one is the current `shouldnt_be_public` module now inside
`std::rt`. I think that this either needs a slight reorganization of the
runtime, or otherwise it needs to just wait for the external users of these
modules to get replaced with their `rt` implementations.

Other fixes involve making things pub which should be pub, and otherwise
updating error messages that now reference privacy instead of referencing an
"unresolved name" (yay!).
2013-10-07 13:00:52 -07:00
Brian Anderson
6e38e334de Relate the module hierarchy to directory paths in the parser
Introduces a temporary 'path2' attribute that will replace 'path' after
a snapshot
2012-12-11 15:00:23 -08:00