auto merge of #4944 : catamorphism/rust/xxx-to-fixme, r=catamorphism

This commit is contained in:
bors 2013-02-14 23:11:13 -08:00
commit 9727008ed0
4 changed files with 19 additions and 41 deletions

View File

@ -62,7 +62,7 @@ pub unsafe fn rt_fail_borrowed() {
}
}
// XXX: Make these signatures agree with exchange_alloc's signatures
// FIXME #4942: Make these signatures agree with exchange_alloc's signatures
#[rt(exchange_malloc)]
#[lang="exchange_malloc"]
pub unsafe fn rt_exchange_malloc(td: *c_char, size: uintptr_t) -> *c_char {

View File

@ -250,7 +250,7 @@ struct ProgRes {
r: ProgRepr,
drop {
unsafe {
// XXX: This is bad.
// FIXME #4943: This is bad.
destroy_repr(cast::transmute(&self.r));
}
}

View File

@ -73,7 +73,7 @@ pub fn loan(bccx: @BorrowckCtxt,
Err(ref e) => return Err((*e)),
Ok(()) => {}
}
// XXX: Workaround for borrow check bug.
// FIXME #4945: Workaround for borrow check bug.
Ok(copy lc.loans)
}

View File

@ -96,8 +96,8 @@ pub struct binding_info {
// Implementation resolution
//
// XXX: This kind of duplicates information kept in ty::method. Maybe it
// should go away.
// FIXME #4946: This kind of duplicates information kept in
// ty::method. Maybe it should go away.
pub type MethodInfo = {
did: def_id,
@ -268,8 +268,8 @@ pub enum MethodSort {
// allows it to reference private names. Currently, this is used for the test
// runner.
//
// XXX: The X-ray flag is kind of questionable in the first place. It might
// be better to introduce an expr_xray_path instead.
// FIXME #4947: The X-ray flag is kind of questionable in the first
// place. It might be better to introduce an expr_xray_path instead.
#[deriving_eq]
pub enum XrayFlag {
@ -469,14 +469,6 @@ pub struct Module {
anonymous_children: HashMap<node_id,@Module>,
// XXX: This is about to be reworked so that exports are on individual
// items, not names.
//
// The ident is the name of the exported item, while the node ID is the
// ID of the export path.
exported_names: HashMap<ident,node_id>,
// The status of resolving each import in this module.
import_resolutions: HashMap<ident,@mut ImportResolution>,
@ -498,7 +490,6 @@ pub fn Module(parent_link: ParentLink,
children: HashMap(),
imports: DVec(),
anonymous_children: HashMap(),
exported_names: HashMap(),
import_resolutions: HashMap(),
glob_count: 0,
resolved_import_count: 0
@ -831,7 +822,7 @@ pub struct Resolver {
mut current_module: @Module,
// The current set of local scopes, for values.
// XXX: Reuse ribs to avoid allocation.
// FIXME #4948: Reuse ribs to avoid allocation.
value_ribs: @DVec<@Rib>,
// The current set of local scopes, for types.
@ -2137,8 +2128,9 @@ fn resolve_single_import(module_: @Module,
// We need to resolve both namespaces for this to succeed.
//
// XXX: See if there's some way of handling namespaces in a more
// generic way. We have two of them; it seems worth doing...
// FIXME #4949: See if there's some way of handling namespaces in
// a more generic way. We have two of them; it seems worth
// doing...
let mut value_result = UnknownResult;
let mut type_result = UnknownResult;
@ -3206,8 +3198,8 @@ fn report_unresolved_imports(module_: @Module) {
// This pass simply determines what all "export" keywords refer to and
// writes the results into the export map.
//
// XXX: This pass will be removed once exports change to per-item. Then
// this operation can simply be performed as part of item (or import)
// FIXME #4953 This pass will be removed once exports change to per-item.
// Then this operation can simply be performed as part of item (or import)
// processing.
fn record_exports() {
@ -3507,8 +3499,8 @@ fn search_ribs(ribs: @DVec<@Rib>, name: ident, span: span,
allow_capturing_self: AllowCapturingSelfFlag)
-> Option<def_like> {
// XXX: This should not use a while loop.
// XXX: Try caching?
// FIXME #4950: This should not use a while loop.
// FIXME #4950: Try caching?
let mut i = (*ribs).len();
while i != 0 {
@ -3649,7 +3641,7 @@ fn resolve_item(item: @item, visitor: ResolveVisitor) {
// Create a new rib for the method-specific type
// parameters.
//
// XXX: Do we need a node ID here?
// FIXME #4951: Do we need a node ID here?
match *method {
required(ref ty_m) => {
@ -4555,7 +4547,7 @@ fn resolve_identifier(identifier: ident,
namespace);
}
// XXX: Merge me with resolve_name_in_module?
// FIXME #4952: Merge me with resolve_name_in_module?
fn resolve_definition_of_name_in_module(containing_module: @Module,
name: ident,
namespace: Namespace,
@ -4873,20 +4865,6 @@ fn resolve_expr(expr: @expr, visitor: ResolveVisitor) {
expr_struct(path, _, _) => {
// Resolve the path to the structure it goes to.
//
// XXX: We might want to support explicit type parameters in
// the path, in which case this gets a little more
// complicated:
//
// 1. Should we go through the ast_path_to_ty() path, which
// handles typedefs and the like?
//
// 2. If so, should programmers be able to write this?
//
// class Foo<A> { ... }
// type Bar<A> = Foo<A>;
// let bar = Bar { ... } // no type parameters
match self.resolve_path(path, TypeNS, false, visitor) {
Some(def_ty(class_id)) | Some(def_struct(class_id))
if self.structs.contains_key(&class_id) => {
@ -5320,7 +5298,7 @@ fn dump_module(module_: @Module) {
None => { value_repr = ~""; }
Some(_) => {
value_repr = ~" value:?";
// XXX
// FIXME #4954
}
}
@ -5329,7 +5307,7 @@ fn dump_module(module_: @Module) {
None => { type_repr = ~""; }
Some(_) => {
type_repr = ~" type:?";
// XXX
// FIXME #4954
}
}