Commit Graph

426 Commits

Author SHA1 Message Date
Artem Chikin
19c5c201dc [Parseable Output] Generate valid output for files of unknown types
When we encounter an input or output with an unknown extension 'TY_INVALID', still produce valid JSON specifying the type as "unknown" instead of either crashing or producing malformed JSON.

Resolves rdar://94348593
2022-06-03 11:40:32 -07:00
Doug Gregor
6d82448dc8 Add -enable-experimental-feature X for experimental features.
Experimental features can only be enabled in non-production (+Asserts)
builds. They can be detected with `hasFeature` in the same manner as
"future" features.

The `-enable-experimental-feature X` flag will also look for future
features by that name, so that when an experimental feature becomes an
accepted future feature, it will still be enabled in the same manner.

Switch variadic generics over to this approach, eliminating the
specific LangOption for it.
2022-05-26 11:50:39 -07:00
Artem Chikin
0ddd1a9202 [Parseable Output] Emit parseable messages on failure in CompilerInstance.setup
FrontendTool sets up diagnostic infrastructure early and suppresses stdout diagnostics if frontend-parseable-output is enabled. It then calls `CompilerInstance.setup` which may fail - if it fails, we exit early. But that means we have not gotten a chance to emit parseable-output.

This change moves emission of the `began` parseable message to before `CompilerInstance.setup`, and ensures that a corresponding `finished` message is emitted if the setup fails.

Resolves rdar://93187783
2022-05-12 15:18:57 -07:00
Artem Chikin
b66d6cc9ec [Parseable Output] Compute file extensions using full extension strings
Generation of valid JSON output for parseable output depends on being able to determine file types of inputs and outputs of compilation tasks. FileTypes.def defines multiple file kinds with multiple '.' extensions, such as '.abi.json' or '.features.json', but existing code attempted to compute file outputs only using the trailing suffix of the path after the last '.'. This led to some files not being recognized, which led to us not being able to generate valid JSON.

Resolves rdar://92961252
2022-05-12 13:01:09 -07:00
Artem Chikin
9d2aecbde4 Frontend emitting parsable output, as implemented, emitted began/finished messages on a per-primary basis. Which means that -frontend-parseable-output had no effect in contexts where no primary inputs are specified, such as WMO. This change fixes that by also emitting began/finished messages when no primary outputs are specified.
Resolves rdar://91999048
2022-04-28 13:24:03 -07:00
Josh Soref
38c0de0598 Spelling frontend (#42465)
* spelling: calculated

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: compilations

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: containing

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: dependency

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: dependent

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: diagnose

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: explicit

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: explicitly

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: feature

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: fulfill

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: fulfillment

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: invoke

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: module

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: multiple

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: nonexistent

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: omitted

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: optimization

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: performing

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: primaries

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: propagate

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: sacrifices

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: scanned

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: substitution

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: successful

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: typecheck

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: unobtrusive

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

* spelling: utilities

Signed-off-by: Josh Soref <jsoref@users.noreply.github.com>

Co-authored-by: Josh Soref <jsoref@users.noreply.github.com>
2022-04-20 15:18:37 -07:00
Arnold Schwaighofer
609428cad9 IRGen: disable-llvm-optzns should disable the objc-arc-contract pass
Don't run the objc-arc-contract pass when disable-llvm-optzns is passed.

In pipeline setups where the swift-frontend is called twice;
- the first time to create bitcode
- the second time to generate object code from the bitcode (using
  disable-llvm-optzns)
we don't want to run the objc-arc-contract pass twice.

rdar://91908312
2022-04-18 13:47:10 -07:00
Allan Shortlidge
e10ebcab08 ModuleInterface: Adopt swiftinterface verification lit substitutions in emit-interface-macos-canonical-version.swift. Move the test to the ModuleInterface directory to colocate it with other swiftinterface generation tests and rename it to availability-macos-canonical-version.swift. Remove an extraneous -experimental-skip-non-inlinable-function-bodies flag that seemed like it was copypasta. 2022-04-15 20:12:59 -07:00
Alex Lorenz
2e3aa87737 Revert "Revert "Merge pull request #41831 from hyp/unify-header""
This reverts commit 4c9582c295.
2022-03-19 13:36:28 -07:00
Alex Lorenz
4c9582c295 Revert "Merge pull request #41831 from hyp/unify-header"
This reverts commit cd93d23bac, reversing
changes made to f9f5476e9a.
2022-03-18 10:03:07 -07:00
Alex Lorenz
9d52099d5b [cxx-interop] start to emitting a unified header file for a Swift module
This change removes the -emit-cxx-header option, and adds a new -emit-clang-header-path option instead. It's aliased to -emit-objc-header-path for now, but in the future, -emit-objc-header-path will alias to it. After this change Swift can start emitting a single header file that can be expose declarations to C, Objective-C, or C++. For now C++ interface is generated (for all public decls) only when -enable-cxx-interop flag is passed, but that behavior will change once  attribute is supported.
2022-03-17 10:34:47 -07:00
Allan Shortlidge
e25b822f7a Merge pull request #41673 from tshortli/inlining-availability-checking
Add -target-min-inlining-version to aid type checking for inlinable functions in resilient libraries
2022-03-07 17:30:59 -08:00
Becca Royal-Gordon
cdcd726f92 Add fixit alternation to -verify
You can now put `||` between two fix-its to indicate that the test succeeds if either of them is present. This is meant for situations where a fix-it might vary slightly in different subtests or test configurations.

Also fixes a bug in the diagnostic verifier where "expected-whatever" would search beyond the same line for its opening "{{", potentially finding one many lines away and giving a bad diagnostic and poor recovery behavior.
2022-03-04 10:28:42 -08:00
Jonathan Grynspan
34a4d66481 [SR-15938] Error when referencing #dsohandle in a Swift test on Windows 2022-03-04 12:25:49 -05:00
swift-ci
ad903094c0 Merge pull request #41386 from beccadax/your-library-is-overdue
Weaken some type checks for @preconcurrency decls
2022-03-01 16:13:59 -08:00
Becca Royal-Gordon
5330653f3d Improve -verify’s wrong diagnostic kind diagnosis
If, for instance, an error is emitted as a warning instead, the verifier now detects this and emits a single diagnostic saying that the warning was found but had the wrong kind, instead of emitting one diagnostic saying the error was missing and another saying the warning was unexpected.

In theory there are some edge cases we could handle better by doing two separate passes—one to detect exact expectation matches and remove them, another to detect near-misses and diagnose them—but in practice, I think the text + diagnostic location is likely to be unique enough to keep this from being a problem. (I would hesitate to do wrong-line diagnostics in the same pass like this, though.)
2022-02-18 13:27:59 -08:00
Anthony Latsis
2128678d56 DiagnosticVerifier: Support line offsets in fix-it verification ranges 2022-02-18 04:42:57 +03:00
Anthony Latsis
91fedd9b8a Verifier: Support line numbers in fix-it verification 2022-01-31 05:04:49 +03:00
Slava Pestov
d57f1aaa06 Merge pull request #40919 from slavapestov/debug-generic-signatures
AST: -debug-generic-signatures protocol-qualifies DependentMemberTypes
2022-01-20 10:37:13 -05:00
Slava Pestov
e4973158d8 AST: -debug-generic-signatures protocol-qualifies DependentMemberTypes 2022-01-19 22:36:15 -05:00
Alex Lorenz
6486c3c1d8 [interop] add an option to emit C++ header interface for a module 2022-01-19 18:55:14 -08:00
Erik Eckstein
408cf02bc8 rework cross-module-optimization
* rename the CrossModuleSerializationSetup pass to simply CrossModuleOptimization
* remove the CMO specific serializer pass. Instead run the CrossModuleSerializationSetup pass directly before the standard serializer pass.
* correctly handle shared functions (e.g. specializations)
* refactoring
2021-12-20 11:33:02 +01:00
Xi Ge
28eef324d0 Frontend: support emitting ABI descriptor when specifying the path in supplementary output map. rdar://86559811 2021-12-15 22:42:15 -08:00
Argyrios Kyrtzidis
7ad553f981 Merge pull request #40273 from bnbarham/pch-allow-errors
[ClangImporter] Handle allowing PCM/PCH errors
2021-12-03 18:13:46 -08:00
Ellie Shin
48a483a9f5 Update tests 2021-12-02 20:57:55 -08:00
Ellie Shin
b82e7bf4e1 Add explicit builds mode to tests 2021-12-02 15:51:07 -08:00
Ellie Shin
5da881256f Module aliasing: Use real module names with -scan-dependencies
Resolves rdar://85991587
2021-12-02 15:45:30 -08:00
Ben Barham
efcb5e38ca [Frontend] Add whether allowing compiler errors to the pretty stacktrace
While this will be in the compiler arguments, it's easy to miss when
skimming over the pretty stacktrace. Add an explicit message to make it
easier to see while looking over crashes.
2021-11-30 15:46:52 +10:00
Ellie Shin
e2631c9ca8 Move getRealImportPath to cpp
Remove unnecessary setter
Update tests
2021-11-10 00:30:26 -08:00
Ellie Shin
c0766177ce Update test 2021-11-10 00:30:25 -08:00
Ellie Shin
4814ea483f Update dump-ast test 2021-11-10 00:30:25 -08:00
Ellie Shin
09093e89dc Pass in out param to ImportPath getters
Modify ASTDumper to print module real names
Add doc comments
2021-11-10 00:30:25 -08:00
Ellie Shin
fd5aaa5bcb Add a test with -emit-imported-module 2021-11-10 00:30:25 -08:00
Ellie Shin
7cabcc913e Update dump ast test 2021-11-10 00:30:25 -08:00
Ellie Shin
c08e7b9cd2 Module aliasing: modify ImportPath getters to return real module names
Resolves rdar://83632921, rdar://83633109
2021-11-10 00:30:25 -08:00
Ellie Shin
b294fe94f2 Merge pull request #39929 from apple/es-src
Module Aliasing: do not allow module real names to appear in source files (only allow module aliases).
Resolves rdar://83592084
2021-11-01 17:50:49 -07:00
Ellie Shin
66d64b610d Updates params to ASTContext::getRealModuleName 2021-10-29 01:52:33 -07:00
elsh
10a96dff57 Module Aliasing: do not allow module real names to appear in source files / only allow aliases
Resolves rdar://83592084
2021-10-28 02:36:59 -07:00
elsh
477c7a6c4b Fix tests: remove platform specific entries 2021-10-19 17:32:16 -07:00
elsh
fe29ae027e Disable failing tests temporarily 2021-10-19 17:01:53 -07:00
Ellie Shin
6f34844b49 Merge pull request #39705 from apple/es-module-alias-serialize
[Module aliasing] Serialize SIL and binaries with module real names for referenced or imported modules. Resolves rdar://83632529
2021-10-18 21:55:45 -07:00
elsh
9143bd4805 Update test 2021-10-18 17:37:26 -07:00
elsh
7a3a032ce8 Use module real name in ASTPrinter 2021-10-15 03:57:21 -07:00
elsh
e3d5ae0172 Add SIL re-ingest test 2021-10-14 19:03:31 -07:00
elsh
e1dfdba1e6 Add doc comments and tests for AST / SIL 2021-10-14 16:18:02 -07:00
elsh
4b3ee94c99 Cleanup tests 2021-10-13 14:52:52 -07:00
elsh
1d0c84c690 Update tests 2021-10-13 13:54:10 -07:00
elsh
adf33e0bb9 Add doc comments
Rename test
2021-10-12 15:03:30 -07:00
elsh
96fb3d6112 [Module aliasing] Serialize binary with module real name
Resolves rdar://83632529
2021-10-12 13:59:34 -07:00
Ellie Shin
5c92ccadc9 Merge pull request #39628 from apple/es-m4
[Module aliasing] Mangle symbols using the module real name (binary name), which can be different from the module name appearing in source files if '-module-alias' is used. Resolves rdar://83632604.
2021-10-10 13:37:46 -07:00