duplicated by the InVarOrLetPattern state in the Parser object. Beef
InVarOrLetPattern up so that we can remove it.
NFC except that we now reject pointless let patterns in foreach loops,
similar to how we reject var patterns inside of let patterns.
Swift SVN r26163
This changes 'if let' conditions to take general refutable patterns, instead of
taking a irrefutable pattern and implicitly matching against an optional.
Where before you might have written:
if let x = foo() {
you now need to write:
if let x? = foo() {
The upshot of this is that you can write anything in an 'if let' that you can
write in a 'case let' in a switch statement, which is pretty general.
To aid with migration, this special cases certain really common patterns like
the above (and any other irrefutable cases, like "if let (a,b) = foo()", and
tells you where to insert the ?. It also special cases type annotations like
"if let x : AnyObject = " since they are no longer allowed.
For transitional purposes, I have intentionally downgraded the most common
diagnostic into a warning instead of an error. This means that you'll get:
t.swift:26:10: warning: condition requires a refutable pattern match; did you mean to match an optional?
if let a = f() {
^
?
I think this is important to stage in, because this is a pretty significant
source breaking change and not everyone internally may want to deal with it
at the same time. I filed 20166013 to remember to upgrade this to an error.
In addition to being a nice user feature, this is a nice cleanup of the guts
of the compiler, since it eliminates the "isConditional()" bit from
PatternBindingDecl, along with the special case logic in the compiler to handle
it (which variously added and removed Optional around these things).
Swift SVN r26150
For now, we assume that 'while' after the braces starts
a do/while rather than being an independent statement.
We should disambiguate this, or better, remove do/while.
Tests later.
Swift SVN r26079
This doesn't allow 'continue' out of an if statement for the same reason we don't
allow it on switch: we'd prefer people to write loops more explicitly.
Swift SVN r25565
We didn't have a consistent way to utter attributes in diagnostics, sometimes saying the
'foo' attribute is not allowed
@foo attribute is not allowed
'foo' is not allowed
@foo is not allowed
etc. Standardize on the last one, since it is clear (with the @ sign, with no quotes, with no
'attribute' word in the diagnostic) that we're talking about an attribute. Move a bunch of
diagnostics inline with this.
Swift SVN r25524
un-type-annotated AnyObject binding in "if let", and allows general
patterns in if-let.
This also reverts some unrelated QoI improvements that went in with those
patches, but I'll add those back next.
Swift SVN r25507
two logically independent but related patches conflated together:
- Improve error recovery for malformed if/let conditions, particularly
when the user uses "," instead of &&. Add testcases for error recovery
requested by Jordan.
- Add a syntactic requirement that the pattern of an if/let condition be
a simple identifier or _. We allow slightly broader patterns here now,
but they will change in the future when refutable patterns are allowed.
It is best to be narrow and then open it up later so we can do a great
job of QoI then.
This includes the changes to the stdlib directory that I forgot to commit
with the patch the previous time.
Swift SVN r25408
- Improve error recovery for malformed if/let conditions, particularly
when the user uses "," instead of &&. Add testcases for error recovery
requested by Jordan.
- Add a syntactic requirement that the pattern of an if/let condition be
a simple identifier or _. We allow slightly broader patterns here now,
but they will change in the future when refutable patterns are allowed.
It is best to be narrow and then open it up later so we can do a great
job of QoI then.
Swift SVN r25371
The previous commit enabled this; now it's just about removing the
restriction in the parser and tightening up code completion.
Using 'super' in a closure where 'self' is captured weak or unowned still
doesn't work; the reference to 'self' within the closure is treated as
strong regardless of how it's declared. Fixing this requires a cascade of
effort, so instead I just cloned rdar://problem/19755221.
rdar://problem/14883824
Swift SVN r25065
Change all the existing addressors to the unsafe variant.
Update the addressor mangling to include the variant.
The addressor and mutable-addressor may be any of the
variants, independent of the choice for the other.
SILGen and code synthesis for the new variants is still
untested.
Swift SVN r24387
if-let statements (also while and var, of course) that include multiple bindings
and where clauses.
SILGen support still remains, it currently just asserts on the new constructs.
Swift SVN r24239
SILMetadata is the base class with a single enum member (MDKind).
SILBranchNode is the derived class with additional members:
unsigned NumOperands
an array of uint32_t
A static member function SILBranchNode::get is implemented to get or create
SILBranchNode. All SILMetadata created are uniqued and saved in SILModule's
member variable:
llvm::FoldingSet<SILMetadata> Metadatas
Usage of SILMetadta by SILInstruction is captured in SILModule's member variable:
llvm::DenseMap<const SILInstruction *, SILMetadata *> MetadataStore
This is similar to LLVM's Metadata. Another option is to add a SILMetadata* to
SILInstruction. The disadvantage is the waste of space when we don't have PGO on.
This commit also enables parsing and printing of SILMetadata.
We add keyword sil_metadata to define SILMetadata:
sil_metadata !0 = {"branch_weights", 3, 5}
For parsing, we add a map in SILModule
llvm::DenseMap<unsigned, SILMetadata *> NumberedMetadata
that maps from ID to SILMetadata* to help matching usage of "!id" in SILFunction
with definition of "!id" in sil_metadata section.
For printing, we assign IDs to SILMetadata at SILModule scope, we then pass in
an optional argument of
llvm::DenseMap<const SILMetadata *, unsigned> *MetadataMap
to SILFunction::print in order to get the ID of SILMetadata used in
SILInstruction.
Post-commit review will be appreciated.
rdar://18269754
Swift SVN r23713
The issue is reproducible in erroneous code scenario when a string literal
has invalid stuff inside interpolation segment, like: “... \( abc } ) ...”.
In this case we used to do some magic for switching context inside the string
and parse the interior of \(...) as regular expression list, but expected that
the parsing finishes at closing “)” which does not necessarily true in case
the code has errors.
The assertion was replaced with an error diagnostics.
Swift SVN r23296
define properties and subscripts with the
get+mutableAddress combination of accessors.
Fix a couple of simple problems this exposes.
rdar://17270560
Swift SVN r22419
This lets us reliably print and parse opened archetypes across different compiler invocations. Using a source-related locator would be ideal, but that's complicated by the need to manufacture, print, and parse these things during SIL passes, so cop out and burn a UUID for now.
Swift SVN r22385
This patch extends the AST and parsing of #os(...) queries to permit queries for
multiple platforms, e.g., #os(OSX >= 10.10, iOS >= 8.0). It also improves
parsing error recovery.
Swift SVN r22154
This patch adds a new 'pound_os' token, a new case for it in parseExprPostfix, and parsing of platform version constraints, e.g., OSX >= 10.10.
It also adds enough type checking and SILGen to get the parsing tests to run without triggering "Unimplemented" assertions.
Swift SVN r21865
This already can't happen in most circumstances because of trailing closures, but we didn't explicitly disallow it at the beginning of a BraceStmt or following a statement production. Fixes the parser part of rdar://problem/17850752 (though there's a type checker bug there too).
Swift SVN r21663
The old message, "deinitializer does not have a parameter clause," was
confusing because it would be issued for a deinit that clearly *had* a
parameter clause. Also, one out of three messages called deinitializers
"'deinitializer' functions." Since "deinitializer" does not appear in
the program text, it doesn't make sense to quote it, either.
Swift SVN r21468
This moves a crude parser check for stored properties in enums & extensions into Sema, and changes it
to be correct (rejecting the case that caused the crash in the radar) as well as more descriptive.
Swift SVN r20695
to emit fixit's when we rename something, e.g.:
t.swift:6:9: error: 'float' has been renamed to Float
var y : float
^~~~~
Float
Adopt this in the stdlib.
Swift SVN r20549
