by TranslationUnit. This list existed solely to allow name lookup of
an unbound IdentifierType to know its DeclContext. Instead of indirecting
through this list, just store the DeclContext in the IdentifierType in its
uninitialized state.
This eliminates a really terrible performance fixme about scanning the list,
eliminates the management fiddling around with this list in the parser, and
is generally much cleaner.
Swift SVN r5246
mode for normal .swift files. We basically parse batches of non-sil function
decls, type check them as a batch, then process any SIL functions. This allows
us to have mutually recursive types and other things that are fully sema'd and
that are referenced by SIL functions, without involving SIL functions too
intimately with type checking.
This does mean that SIL functions can't forward reference types, oh well.
Swift SVN r5243
and infinitely loop because it didn't realize that skipUntil doesn't always
consume something.
This fixes an infinite loop on this testcase:
func isSpace(c : Char) -> Bool {
return (c == '\v' ||
c == '\f')
}
which comes from rdar://11936003. I'm not adding it because it only works
as the last thing in a file, and isn't likely to regress. The diagnostics
produced are also still really really awful for this.
Swift SVN r5241
'|' is part of the character set for operators, but within the
signature of a closure we need to treat the first non-nested '|' as
the closing delimiter for the closure parameter list. For example,
{ |x = 1| 2 + x}
parses with the default value of '1' for x, with the body 2 + x. If
the '|' operator is needed in the default value, it can be wrapped in
parentheses:
{ |x = (1|2)| x }
Note that we have problems with both name binding and type checking
for default values in closures (<rdar://problem/13372694>), so they
aren't actually enabled. However, this allows us to parse them and
recover better in their presence.
Swift SVN r5202
This simple refactor makes pattern-tuple-element available for re-use
in closure expressions. As a drive-by, diagnose non-final ellipses via
diagnose() rather than via assert(), the latter being considered
rather unfriendly. Also, take pains to restore AST invariants after
such an error.
Swift SVN r5163
Original message:
SIL Parsing: add plumbing to know when we're parsing a .sil file
Enhance the lexer to lex "sil" as a keyword in sil mode.
Swift SVN r4988
Per Chris's feedback and suggestions on the verbose fix-it API, convert
diagnostics over to using the builder pattern instead of Clang's streaming
pattern (<<) for fix-its and ranges. Ranges are included because
otherwise it's syntactically difficult to add a fix-it after a range.
New syntax:
diagnose(Loc, diag::warn_problem)
.highlight(E->getRange())
.fixItRemove(E->getLHS()->getRange())
.fixItInsert(E->getRHS()->getLoc(), "&")
.fixItReplace(E->getOp()->getRange(), "++");
These builder functions only exist on InFlightDiagnostic; while you can
still modify a plain Diagnostic, you have to do it with plain accessors
and a raw DiagnosticInfo::FixIt.
Swift SVN r4894
Fix array/dictionary literal parsing robustness by consolidating and improving
the parsing of lists in general (tuples/array/dictionary literals, attribute
lists, statement lists, declaration lists, etc).
Missing commas in tuple/array/dictionary literals or declaration attributes
are now detected, reported, and recovered from.
Premature ellipsis in tuples are now detected, reported, and recovered from.
Swift SVN r4631
own TLCD. This is important to preserve the ordering of stmt and expr w.r.t.
PatternBindingDecls that initialize the decls.
We keep the BraceStmt wrapping it to make it more similar to other decls
though.
Swift SVN r4626
Unfortunately, this regresses the repl when expressions like (1,2) are entered. This is because the repl is violating some invariants (forming dags out of ASTs, making ASDAG's which upset the type checker). I'm going to fix this next, but can't bring myself to do it in the same commit.
Swift SVN r4617
Now that we enforce semicolon or newline separation between statements, we can relax the whitespace requirements on '(' and '[' tokens. A "following" token is now just a token that isn't at the start of a line, and any token can be a "starting" token. This allows for:
a(b)
a (b)
a[b]
a [b]
to parse as applications and subscripts, and:
a
(b)
a
[b]
to parse as an expr followed by a tuple or an expr followed by a container literal.
Swift SVN r4573
Implement the syntax 'if x then y else z', which evaluates to 'y' if 'x' is true or 'z' if 'x' is false. 'x' must be a valid logic value, and 'y' and 'z' must be implicitly convertible to a common type.
Swift SVN r4407
Walk backward from a '.' through balanced () [] <> to find context. Fix a problem with parseCompletionContextExpr where UnresolvedIdentifierTypes wasn't getting properly transferred to the TranslationUnit after parsing, causing 'Foo<T>.' completion to blow up.
Swift SVN r4075
If the completion prefix has a '.' behind it, guesstimate a context expression by lexing backward through an identifier(.identifier)* dotted path, then attempt to parse and typecheck that expression to decide on a base type in which to find completions.
Swift SVN r4063
The lexer now models tuples, patterns, subscripting, function calls, and
field access robustly. The output tokens are now better named as well:
l_paren and l_paren_call, and l_square and l_square_subscript. It
should be much more clear now which one to use. Also, the use of
l_paren or l_square will not arbitrarily flip flop if the token before
it is a keyword or if the token before it was the trailing ']' of an
attribute list. Similarly, tuples will always cause the lexer to produce
l_paren, regardless if the user typed '((x,y))' or '( (x,y))'.
When we someday add array literals, the right token is now naturally
falling out of the lexer.
Swift SVN r3840
rdar://12315571
Allow a function to be defined with this syntax:
func doThing(a:Thing) withItem(b:Item) -> Result { ... }
This allows the keyword names in the function type (in this case
`(_:Thing, withItem:Item) -> Result`) to differ from the names bound in the
function body (in this case `(a:Thing, b:Item) -> Result`, which allows
for Cocoa-style `verbingNoun` keyword idioms to be used without requiring
those keywords to also be used as awkward variable names. In addition
to modifying the parser, this patch extends the FuncExpr type by replacing
the former `getParamPatterns` accessor with separate `getArgParamPatterns`
and `getBodyParamPatterns`, which retrieve the argument name patterns and
body parameter binding patterns respectively.
Swift SVN r3098
resulting token goes back through the lexer to get the appropriate
token kind. Thanks to Chris for spotting this.
Also, document the '<' and '>' splitting behavior in LangRef.
Swift SVN r2192
and use this information as cues in the language. Right now,
we do not accept things like "-- *i" because the prefix
operator is not correctly right-bound; instead you have to
write "--(*i)". I'm okay with that; I did add a specialized
diagnostic recognizing operator-binary in a place where we're
expecting a potential operator-prefix.
Swift SVN r2161
protocol conformance types, e.g., 'protocol<P, Q>'. A few things
people *might* want to scream about, or at least scrutinize:
- The parsing of the '<' and '>' is odd, because '<' and '>' aren't
tokens, but are part of the operator grammar. Neither are '>>',
'>>>', '<>', etc., which also come up and need to be parsed
here. Rather than turning anything starting with '<' or '>' into a
different kind of token, I instead parse the initial '<' or '>'
from an operator token and leave the rest of the token as the
remaining operator.
- The canonical form of a protocol-composition type is minimized by
removing any protocols in the list that were inherited by other
protocols in the list, then sorting it. If a singleton list is
left, then the canonical type is simply that protocol type.
- It's a little unfortunate that we now have two existential types
in the system (ProtocolType and ProtocolCompositionType), because
many places will have to check both. Once ProtocolCompositionTypes
are working, we should consider whether it makes sense to remove
ProtocolType.
Still to come: name lookup, coercions.
Swift SVN r2066
I haven't really carefully considered whether existing type-checking etc. is correct for the main module (I think the name-binding rules need to be a bit different?), but it seems to work well enough for the obvious cases.
This is enough to get the one-liner "println(10)" to print "10" in "swift -i" mode, although the path to swift.swift still needs to be explicitly provided with -I.
Swift SVN r1325
The later could represent semantic errors, but we'd rather represent those with an
ExprError node instead. This simplifies the code and allows the parser to build a more
fully-formed AST that IDE clients will like.
Swift SVN r1141
