...warning about (for now only) functions and variables with external linkage
that likely don't need it.
The problems with moving entities into unnamed namespacs and breaking ADL
(as alluded to in comments in compilerplugins/clang/external.cxx) are
illustrated by the fact that while
struct S1 { int f() { return 0; } };
int f(S1 s) { return s.f(); }
namespace N {
struct S2: S1 { int f() { return 1; } };
int f(S2 s) { return s.f(); }
}
int main() { return f(N::S2()); }
returns 1, both moving just the struct S2 into an nunnamed namespace,
struct S1 { int f() { return 0; } };
int f(S1 s) { return s.f(); }
namespace N {
namespace { struct S2: S1 { int f() { return 1; } }; }
int f(S2 s) { return s.f(); }
}
int main() { return f(N::S2()); }
as well as moving just the function f overload into an unnamed namespace,
struct S1 { int f() { return 0; } };
int f(S1 s) { return s.f(); }
namespace N {
struct S2: S1 { int f() { return 1; } };
namespace { int f(S2 s) { return s.f(); } }
}
int main() { return f(N::S2()); }
would each change the program to return 0 instead.
Change-Id: I4d09f7ac5e8f9bcd6e6bde4712608444b642265c
Reviewed-on: https://gerrit.libreoffice.org/60539
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
comment from sberg:
aren't these changes broken in general, when the called function
may throw an exception before it takes ownership of the passed-in pointer?
So revert, except for
(a) PlainTextFilterDetect::detect, which was definitely a leak
(b) SwCursor::FindAll, where unique_ptr was being unnecessarily used
This reverts commit 7764ae70b0.
Change-Id: I555e651b44e245b031729013d2ce88d26e8a357e
Reviewed-on: https://gerrit.libreoffice.org/60301
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
most of these changes just make the change of ownership when using
std::unique_ptr clearer, but there is one definite leak fix in
PlainTextFilterDetect::detect
Change-Id: I8282a68007222a4fee84004f394bde0cca8569e9
Reviewed-on: https://gerrit.libreoffice.org/60159
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
...by explicitly defaulting the copy/move functions (and, where needed in turn,
also a default ctor) for classes that have a user-declared dtor that does
nothing other than an implicitly-defined one would do, but needs to be user-
declared because it is virtual and potentially serves as a key function to
emit the vtable, or is non-public, etc. (For LIBO_INTERNAL_ONLY, in cppumaker-
genered code.)
Change-Id: Ia13d945c10600f5793a3247f85a464170ede483d
Reviewed-on: https://gerrit.libreoffice.org/58116
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
m_unsignedLongDependency was already set but never used.
Behavior of Includes::Includes ctor is now in consistency with
the related switch cases in Includes::add (see: m_includeSalTypesH)
Change-Id: I4cca1bab014ac280b73b7532aadfd5ff2b0b9894
Reviewed-on: https://gerrit.libreoffice.org/57792
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Jenkins
rtl/string.hxx and rtl/ustring.hxx both unnecessarily #include <sal/log.hxx>
(and don't make use of it themselves), but many other files happen to depend on it.
This is a continuation of commit 6ff2d84ade
to be able to remove those unneeded includes.
This commit adds missing headers to every file found by:
grep -FwL sal/log.hxx $(git grep -Elw 'SAL_INFO|SAL_INFO_IF|SAL_WARN|SAL_WARN_IF|SAL_DETAIL_LOG_STREAM|SAL_WHERE|SAL_STREAM|SAL_DEBUG')
to directories from a* to configmgr
Change-Id: I6ea1a7f992b1f835f5bac7a725e1135abee3f85a
Reviewed-on: https://gerrit.libreoffice.org/57170
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
auto-rewrite with <https://gerrit.libreoffice.org/#/c/47798/> "Enable
loplugin:cstylecast for some more cases" plus
solenv/clang-format/reformat-formatted-files
Change-Id: If1c6b7a83f6f64fbcbced4f036c0f859a1408d0c
...that are not composed of multiple tokens, like ("foo" "bar"). Also don't yet
warn about Boolean literals, which are sometimes wrapped in parentheses to
silence unreachable-code warnings.
To avoid multiple warnings about code like
f((0))
switch to generally using a set of ParenExpr to keep track of which occurrences
have already been handled.
Change-Id: I036a25a92836ec6ab6c56ea848f71bc6d63822bc
Reviewed-on: https://gerrit.libreoffice.org/45317
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
The types 'long' and 'unsigned long' are different sizes on different
platforms, making them wholy unsuitable for portable code.
And when I mean different sizes, I mean 64bit Linux and 64bit Windows
have different bit sizes.
Change-Id: Id4d8cd709bf71a3f85ed6515fae2b0a21ae8c25d
Reviewed-on: https://gerrit.libreoffice.org/41130
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
this modifies codemaker so that, for an UNO enum, we generate code
that effectively looks like:
#ifdef LIBO_INTERNAL_ONLY && HAVE_CX11_CONSTEXPR
enum class XXX {
ONE = 1
};
constexpr auto ONE = XXX_ONE;
#else
...the old normal way..
#endif
which means that for LO internal code, the enums are scoped.
The "constexpr auto" trick acts like an alias so we don't have to
use scoped naming everywhere.
Change-Id: I3054ecb230e8666ce98b4a9cb87b384df5f64fb4
Reviewed-on: https://gerrit.libreoffice.org/34546
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
...after e57ca02849 "Remove dynamic exception
specifications" dropped uses of UNO exception types from those generated files.
This needed first adapting client code that implicitly depended on those
#includes in the generated files it included ("Add missing #includes":
6dce9c6757,
a34d9150b4,
fed3783e77).
Change-Id: Ia1defda2d586b69092f53feef3644875192cbea4
Reviewed-on: https://gerrit.libreoffice.org/34014
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
just the simple and obvious case for now, of a local var being allocated
and deleted inside a single local block, and the delete happening at the
end of the block
Change-Id: I3a7a094da543debdcd2374737c2ecff91d644625
Reviewed-on: https://gerrit.libreoffice.org/33749
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
...(for now, from LIBO_INTERNAL_CODE only). See the mail thread starting at
<https://lists.freedesktop.org/archives/libreoffice/2017-January/076665.html>
"Dynamic Exception Specifications" for details.
Most changes have been done automatically by the rewriting loplugin:dynexcspec
(after enabling the rewriting mode, to be committed shortly). The way it only
removes exception specs from declarations if it also sees a definition, it
identified some dead declarations-w/o-definitions (that have been removed
manually) and some cases where a definition appeared in multiple include files
(which have also been cleaned up manually). There's also been cases of macro
paramters (that were used to abstract over exception specs) that have become
unused now (and been removed).
Furthermore, some code needed to be cleaned up manually
(avmedia/source/quicktime/ and connectivity/source/drivers/kab/), as I had no
configurations available that would actually build that code. Missing @throws
documentation has not been applied in such manual clean-up.
Change-Id: I3408691256c9b0c12bc5332de976743626e13960
Reviewed-on: https://gerrit.libreoffice.org/33574
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
See the mail thread starting at
<https://lists.freedesktop.org/archives/libreoffice/2017-January/076665.html>
"Dynamic Exception Specifications" for details.
This leaves some #include <...Exception.hpp> in the generated code that are no
longer needed by the generated files themselves. But C++ implementation classes
still use dynamic exception specifications and usually don't repeat the relevant
#includes, so postpone removal until the implementation classes are cleaned up
in a later step.
Change-Id: Ifb7df3bb12f7f10c655b85cb0626afe5050479a4
i.e., css::uno::Any function template specializations
Any::has<Any>() const
Any::get(Any const &) const
operator >>=(Any const &, Any &)
operator <<=(Any &, Any const &)
that don't make much sense (the first is always true, the rest can be replaced
with operator =, which additionally supports move semantics). For 3rd-party
compatibility, do this only for LIBO_INTERNAL_ONLY, however.
However, some generic template code did benefit from operator >>= working also
for Any, so make up for that with a new (LIBO_INTERNAL_ONLY, given that
operator >>= still covers if fine for !LIBO_INTERNAL_ONLY) fromAny,
complementing the existing toAny.
Change-Id: I8b1b5f803f0b909808159916366d53c948206a88
Reviewed-on: https://gerrit.libreoffice.org/30022
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Stephan Bergmann <sbergman@redhat.com>
...as
OStringBuffer b("foo"); b = "bar" + b;
doesn't work as one might expect (see the mail thread starting at
<https://lists.freedesktop.org/archives/libreoffice/2016-October/075464.html>
"concat of OUStringBuffer". That feature was LIBO_INTERNAL_ONLY, anyway. And
of the affected places, MethodDescriptor::getSignature
(codemaker/source/javamaker/javatype.cxx) was the only one that would actually
have benefitted.
Change-Id: Ib84266f43e40c42c2e428f0c0616db8cfa90adff
For UNO ABI reasons, these classes need to be CPPU_GCC_DLLPUBLIC_EXPORT (so
their RTTI symbols get exported), so they are careful to explicitly declare any
special member functions that would otherwise be declared implicitly, to mark
them CPPU_GCC_DLLPRIVATE. But for LIBO_INTERNAL_ONLY, we always use
-fvisibility-inlines-hidden (except for MSVC, where it doesn't matter, as
CPPU_GCC_DLLPRIVATE expands to nothing there, either), so can leave those
functions implicitly declared, so not to prevent move ctor and move assign op
from being implicitly declared.
This covers cid#s 1371143, 1371162, 1371185, 1371189, 1371290, and 1371308.
Change-Id: I468088750a02f85a4790d956c37f4c30de03f00c
The issue of 362d4f0cd4 "Explicitly mark
overriding destructors as 'virtual'" appears to no longer be a problem with
MSVC 2013.
(The little change in the rewriting code of compilerplugins/clang/override.cxx
was necessary to prevent an endless loop when adding "override" to
OOO_DLLPUBLIC_CHARTTOOLS virtual ~CloseableLifeTimeManager();
in chart2/source/inc/LifeTime.hxx, getting stuck in the leading
OOO_DLLPUBLIC_CHARTTOOLS macro. Can't remember what that
isAtEndOfImmediateMacroExpansion thing was originally necessary for, anyway.)
Change-Id: I534c634504d7216b9bb632c2775c04eaf27e927e
