Constraint checks C727 to C730 and most constraints related to attributes #1084
Conversation
lib/Semantics/tools.cpp
Outdated
| static constexpr std::size_t intrinsicNameCount{6}; | ||
| static const std::string intrinsicNames[intrinsicNameCount]{ | ||
| "integer", "real", "doubleprecision", "complex", "character", "logical"}; |
There was a problem hiding this comment.
| static constexpr std::size_t intrinsicNameCount{6}; | |
| static const std::string intrinsicNames[intrinsicNameCount]{ | |
| "integer", "real", "doubleprecision", "complex", "character", "logical"}; | |
| static const std::string intrinsicNames[]{ | |
| "integer", "real", "doubleprecision", "complex", "character", "logical"}; | |
| std::size_t intrinsicNameCount{ | |
| sizeof(intrinsicNames) / sizeof(intrinsicNames[0])}; |
This avoids the magic number 6.
There was a problem hiding this comment.
Using std::array would be even nicer. Or null termination, which I tend to use myself.
There was a problem hiding this comment.
Thanks, Tim. I couldn't figure out how to do this.
test/Semantics/resolve76.f90
Outdated
| !ERROR: No intrinsic or user-defined ASSIGNMENT(=) matches operand types LOGICAL(4) and REAL(4) | ||
| f = 1.0 | ||
| end | ||
| end |
There was a problem hiding this comment.
What happened to the test previously in resolve76.f90?
There was a problem hiding this comment.
Thanks for catching this. I mistakenly overwrote it when I merged from master into my branch. I'll restore it.
| !ERROR: Attribute 'TARGET' cannot be used more than once | ||
| real, target, target :: targetVar | ||
| !ERROR: Attribute 'VOLATILE' cannot be used more than once | ||
| real, volatile, volatile :: volatileVar |
There was a problem hiding this comment.
There should be some tests where the duplicate attributes are not in the same statement.
There was a problem hiding this comment.
Wow, I didn't even realize that this was possible. I'll add some tests.
lib/Semantics/resolve-names.cpp
Outdated
| bool AttrsVisitor::IsDuplicateAttr(Attr attrName) { | ||
| if (attrs_->test(attrName)) { | ||
| Say(currStmtSource().value(), | ||
| "Attribute '%s' cannot be used more than once"_err_en_US, |
There was a problem hiding this comment.
Why a fatal error? It's not going to hurt anything.
There was a problem hiding this comment.
Thanks for pointing this out. I agree with you. I just checked, and duplicate attributes don't seem to be noticed by pgf90, but they're fatal errors for both IBM and Intel.
I'll change the duplicates to warnings.
There was a problem hiding this comment.
Actually, some attributes such as CODIMENSION, EXTENDS, DIMENSION convey extra information. It seems appropriate to me that duplicating these attributes should be an error.
There was a problem hiding this comment.
EXTENDS is an optional part of the definition of a derived type, not an attribute of an entity.
I'm not saying that we should allow all possible duplicated attributes (besides the ones that the standard says we have to allow).
…ted to attributes The full list of constraints is C727, C728, C729, C730, C743, C755, C759, C778, and C1543. I added a function to tools.cpp to check to see if a symbol name is the name of an intrinsic type. The biggest change was to resolve-names.cpp to check to see if attributes were either duplicated or in conflict with each other. I changed all locations where attributes were set to check for duplicates or conflicts. I also added tests for all checks and annotated the tests and code with the numbers of the constraints being tested/checked.
psteinfeld
force-pushed
the
ps-typechecks
branch
from
4271a0b
to
3f30e8a
Compare
…nts related to attributes The full list of constraints is C727, C728, C729, C730, C743, C755, C759, C778, and C1543. I added a function to tools.cpp to check to see if a symbol name is the name of an intrinsic type. The biggest change was to resolve-names.cpp to check to see if attributes were either duplicated or in conflict with each other. I changed all locations where attributes were set to check for duplicates or conflicts. I also added tests for all checks and annotated the tests and code with the numbers of the constraints being tested/checked. Original-commit: flang-compiler/f18@3f30e8a Reviewed-on: flang-compiler/f18#1084
…r/ps-typechecks Constraint checks C727 to C730 and most constraints related to attributes Original-commit: flang-compiler/f18@6dbfb80 Reviewed-on: flang-compiler/f18#1084
…nts related to attributes The full list of constraints is C727, C728, C729, C730, C743, C755, C759, C778, and C1543. I added a function to tools.cpp to check to see if a symbol name is the name of an intrinsic type. The biggest change was to resolve-names.cpp to check to see if attributes were either duplicated or in conflict with each other. I changed all locations where attributes were set to check for duplicates or conflicts. I also added tests for all checks and annotated the tests and code with the numbers of the constraints being tested/checked. Original-commit: flang-compiler/f18@3f30e8a Reviewed-on: flang-compiler/f18#1084
The full list of checked constraints is C727, C728, C729, C730, C743, C755, C759, C778,
and C1543.
I added a function to tools.cpp to check to see if a symbol name is the name
of an intrinsic type.
The biggest change was to resolve-names.cpp to check to see if attributes were
either duplicated or in conflict with each other. I changed all locations
where attributes were set to check for duplicates or conflicts.
I also added tests for all checks and annotated the tests and code with the
numbers of the constraints being tested/checked.