This is a port of @eddyb's `const-fn` branch. I rebased it, tweaked a few things, and added tests as well as a feature gate. The set of tests is still pretty rudimentary, I'd appreciate suggestions on new tests to write. Also, a double-check that the feature-gate covers all necessary cases.
One question: currently, the feature-gate allows the *use* of const functions from stable code, just not the definition. This seems to fit our usual strategy, and implies that we might (perhaps) allow some constant functions in libstd someday, even before stabilizing const-fn, if we were willing to commit to the existence of const fns but found some details of their impl unsatisfactory.
r? @pnkfelix
The recent MSVC patch made the build system pass explicit linkers to
rustc, but did not set that up for anything other than MSVC.
This is blocking nightlies.
fixes#25683
I have a very nonscientific measurement of the data via valgrind/massif [here](https://gist.github.com/Manishearth/4c47f15f6835cb3957c4)
I measured the memory usage for both --pretty=expanded and -Z no-trans
It *seems* like there's a 20-25MB decrease during expansion on stage2 librustc; but I'm not quite sure.
r? @eddyb
(have not yet run tests, but it compiles fine, might want to wait before giving r+)
cc @nrc @huon
Reduces pre-trans heap baseline (tested on `fn main() {}`) from 1.2MB to 696kB.
Most of the remaining allocations are used by imported macros.
While it might be possible to also load macros lazily, doing so is non-trivial and this PR doesn't tackle that.
This minimally changes rustdoc's search as described in #25167.
Additionally, I also cleaned up some parts of the JS code.
There is one more change I made: After each result for a primitive type, I added "(Overview of primitive type)". This further differentiates the result from the module (previously, the only difference was that the module's link was blue). I'm not this is the way to go (this seems to be the only place where we do this) and it's no problem for me to remove that commit.
cc @steveklabnik (it concerns docs) and @alexcrichton (who made changes to rustdoc previously)
Make the error message about missing dependencies clearer, since "need program file" is pretty confusing, and make full commands get logged, since that seems to have been the intention.
The change to split up soft_link to OS-specific symlink, symlink_file,
and symlink_dir didn't actually land in 1.0.0. Update the stability and
deprecation attributes to correctly indicate that these changes happend
in 1.1.0.
This "fast path" in `DirEntry::file_type` on Unix wasn't turning out to be so
much of a fast path as the `DT_DIR` case wasn't handled, so directories fell
back to using `lstat` instead. This commit adds the missing case to return
quickly if a path is a directory and `DirEntry::file_type` is used.
This was motivated by http://www.evanmiller.org/a-taste-of-rust.html.
A common problem when working with FFI right now is converting from raw
C strings into `&str` or `String`. Right now you're required to say
something like
let cstr = unsafe { CStr::from_ptr(ptr) };
let result = str::from_utf8(cstr.to_bytes());
This is slightly awkward, and is not particularly intuitive for people
who haven't used the ffi module before. We can do a bit better by
providing some convenience methods on CStr:
fn to_str(&self) -> Result<&str, str::Utf8Error>
fn to_string_lossy(&self) -> Cow<str>
This will make it immediately apparent to new users of CStr how to get a
string from a raw C string, so they can say:
let s = unsafe { CStr::from_ptr(ptr).to_string_lossy() };
The `debug_builders` feature is up for 1.1 stabilization in #24028. This commit stabilizes the API as-is with no changes.
Some nits that @alexcrichton mentioned that may be worth discussing now if anyone cares:
* Should `debug_tuple_struct` and `DebugTupleStruct` be used instead of `debug_tuple` and `DebugTuple`? It's more typing but is a technically more correct name.
* `DebugStruct` and `DebugTuple` have `field` methods while `DebugSet`, `DebugMap` and `DebugList` have `entry` methods. Should we switch those to something else for consistency?
cc @alexcrichton @aturon
The install target depends on compiler-docs but 'all' does not.
This means that running 'make && make install' will run additional
doc builds and tests during installation, which hides bugs in
the build.
For now this just unconditionally stops building compiler docs.
The install target depends on compiler-docs but 'all' does not.
This means that running 'make && make install' will run additional
doc builds and tests during installation, which hides bugs in
the build.
For now this just unconditionally stops building compiler docs.
A common problem when working with FFI right now is converting from raw
C strings into `&str` or `String`. Right now you're required to say
something like
let cstr = unsafe { CStr::from_ptr(ptr) };
let result = str::from_utf8(cstr.to_bytes());
This is slightly awkward, and is not particularly intuitive for people
who haven't used the ffi module before. We can do a bit better by
providing some convenience methods on CStr:
fn to_str(&self) -> Result<&str, str::Utf8Error>
fn to_string_lossy(&self) -> Cow<str>
This will make it immediately apparent to new users of CStr how to get a
string from a raw C string, so they can say:
let s = unsafe { CStr::from_ptr(ptr).to_string_lossy() };
There are more possible optimizations left (cached length in loops) as
well as some possible bugs (shadowed variables) to fix. This is mostly
syntactic.
The `run` function passed its argument to `msg` via `"$@"`, but `msg`
only printed its first argument. I think the intention was for `msg` to
print all its arguments. (If not, `run` should only `msg "$1"`.)
Took me a moment to figure out that the appropriate response to
"need program file" was to install the program named "file", not
to think "If I didn't need the program file, why would I be
compiling things?".
