The bug(s) seems to be fixed already.
Added a test case for verification.
Change-Id: If42d796ba27694254bdcd79d69aad323763c2105
Signed-off-by: Thomas Corbat <tcorbat@hsr.ch>
Implemented proper handling of nested namespaces for toggle function
refactoring.
Change-Id: I850d3a7c9957dc2e26db4d1ac1aabf9a33bc2223
Signed-off-by: Thomas Corbat <tcorbat@hsr.ch>
only prefer a method to a class if the method belongs to that class
Change-Id: I8cf9aed692ce989d83cc8ffd08672a3ee73a1c8b
Signed-off-by: Nathan Ridge <zeratul976@hotmail.com>
This makes it easier to change arguments for all tests. As a side-effect,
all tests now enable assertion (-ea) and have the same max heap size, etc.
Change-Id: I0965d40ba3283efe7eacdf927f6c302ee0ab30f0
Signed-off-by: Marc-Andre Laperle <marc-andre.laperle@ericsson.com>
specializations
if a binding is a member of a ClassInstance (owner) we should not try to
find a corresponding class template (target).
Change-Id: Ic865161f137ecb2a300afd03761cc26ee758b38b
Signed-off-by: Michi <woskimi@yahoo.de>
- Test case for reproducing the problem.
- The code is lost because a copy of the body, which contains macro
references is not rewritten, but its raw signature is taken. The raw
signature of a copy is empty in this case. I've solved this issue by
using the raw signature of the original node for get raw signature. This
is a fundamental change of ASTNode.getRawSignature(). We could also
solve this in the StatementWriter to get the original node before
accessing the raw signature.
Change-Id: I64b408b09444df818d30d99d99de4a1974eacf93
Signed-off-by: Thomas Corbat <tcorbat@hsr.ch>
Reviewed-on: https://git.eclipse.org/r/39406
Tested-by: Hudson CI
Reviewed-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
class templates of a class template
There are 2 changes. Finding the proper naming scope, which basically is
the same as what is done in CPPUnknownTypeScope.getBindings when
retrieving the bindings themselves.
The second change was picking the right scope for the actual
accessibility check for base templates.
Change-Id: I535c8cdd9d07272d37da9d23a03edb9e6b1b3a7a
Signed-off-by: Michi <woskimi@yahoo.de>
Reviewed-on: https://git.eclipse.org/r/39016
Reviewed-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
Tested-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
class
The problem was with template base classes as in the unit test.
Accessibility check did not find the proper AccessContext.namingClass
and always ended further up in the class tree.
Change-Id: I447567cabd20ad5c57c05a9ffa5f0f12c5321ccf
Signed-off-by: Michi <woskimi@yahoo.de>
Reviewed-on: https://git.eclipse.org/r/38944
Tested-by: Hudson CI
Reviewed-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
Tested-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
a CPPASTQualifiedName that represents some template instance can never
refer to a template of the same type
Change-Id: Iaf9a452e4797fc0e797e5ed5ccaadfb00ff4fb8f
Signed-off-by: Michi <woskimi@yahoo.de>
Reviewed-on: https://git.eclipse.org/r/38730
Reviewed-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
Tested-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
class
structs are now being shown as candidates for base classes
while I was at it I also fixed proposals for typedefs, template
parameters and template aliases.
Change-Id: I319b06a195eed88a12969207a21584dc2477219e
Signed-off-by: Michi <woskimi@yahoo.de>
Reviewed-on: https://git.eclipse.org/r/38379
Reviewed-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
Tested-by: Sergey Prigogin <eclipse.sprigogin@gmail.com>
