...where CompilerTest_compilerplugins_clang failed in
compilerplugins/clang/test/useuniqueptr.cxx due to Foo24's
HTMLAttrs::const_iterator it = m_aSetAttrTab.begin();
and either the old compiler lacked Clang's recent
<https://reviews.llvm.org/D50666> "Fix Stmt::ignoreImplicit" (and the above
initialization expression happens to include a CXXBindTemporaryExpr, at least
with libstdc++), or an even older compiler was used in pre-C++17 mode, so the
above initialization expression happens to include an elidable CXXConstructExpr
copy constructor call.
Change-Id: I757a9ad76829e399b4fe2da1c82863909b8c9657
Reviewed-on: https://gerrit.libreoffice.org/61531
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
...where "inline" (in its meaning of "this function can be defined in multiple
translation units") thus doesn't make much sense. (As discussed in
compilerplugins/clang/redundantinline.cxx, exempt such "static inline" functions
in include files for now.)
All the rewriting has been done automatically by the plugin, except for one
instance in sw/source/ui/frmdlg/column.cxx that used to involve an #if), plus
some subsequent solenv/clang-format/reformat-formatted-files.
Change-Id: Ib8b996b651aeafc03bbdc8890faa05ed50517224
Reviewed-on: https://gerrit.libreoffice.org/61573
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
...after 05a337e297 "loplugin:useuniqueptr look
for deleting in loops with iterators", where it didn't emit the warning for
Foo24 in compilerplugins/clang/test/useuniqueptr.cxx during
CompilerTest_compilerplugins_clang, because in the initialization of
HTMLAttrs::const_iterator it = m_aSetAttrTab.begin();
the HTMLAttrs::const_iterator CXXConstructExpr happens to have a second,
defaulted argument.
Change-Id: I882a6dfb5cab1b147f790072f2545b13172c0f9a
Reviewed-on: https://gerrit.libreoffice.org/61530
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
implemeent a reduction approach, which is good at finding virtual
methods that only themselves or their virtual partners.
The accessibility GetVisArea stuff is dead since
commit 891e41fac8
Date: Wed Apr 4 11:23:22 2018 +0200
dead code in AccessibleTextHelper_Impl::UpdateVisibleChildren
Change-Id: I78d9d8bca585ecec8394f2c3fe2baa93db0e58f5
Reviewed-on: https://gerrit.libreoffice.org/60912
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Including:
* expanding STDAPI to its definition (as per
<https://msdn.microsoft.com/library/ms686631(vs.85).aspx> "STDAPI"), to add
__declspec(dllexport) into its middle, in
extensions/source/activex/so_activex.cxx; as discussed in the comments at
<https://gerrit.libreoffice.org/#/c/60691/> "Get rid of Windows .def files in
setup_native, use __declspec(dllexport)", having a function both listed in a
.def file EXPORTS and marking it dllexport is OK, and the latter helps the
heuristics of loplugin:external; however, the relevant functions in
extensions/source/activex/so_activex.cxx probably don't even need to be
exported in the first place?
* follow-up loplugin:salcall in sal/osl/w32/file-impl.hxx
Change-Id: Ida6e17eba19cfa3d7e5c72dda57409005c0a0191
Reviewed-on: https://gerrit.libreoffice.org/60938
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
bdade7e3fc "tdf#105444 DOCX import: don't put
extra paragraphs in comments" caused
> /home/sbergman/lo2/core/writerfilter/source/dmapper/DomainMapper_Impl.cxx:441:22: error: expression always evaluates to zero, lhs=0 rhs=unknown [loplugin:expressionalwayszero]
> (sizeof(SAL_NEWLINE_STRING)-1 == 2 && xCursor->getString() == "\n"))
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
now with sufficiently new Clang, and the code looks reasonable, and there's no
apparent way to avoid such false positives in the plugin. (It could check for a
sub-expression of the problematic expression being an object-like macro, but
SAL_NEWLINE_STRING could just as well be a variable instead of a macro. That
variable would need to be defined in some #if to have different values on
different platforms, so the plugin could theoretically check for such
conditional inclusion, but it's not clear whether that's worth it and would even
be a useful heuristic to not produce neither too many false positives nor too
many false negatives.) So just disable the plugin for good.
Change-Id: I85dc8573735ccac4e19be20ab7443cbaa85a3164
Reviewed-on: https://gerrit.libreoffice.org/60907
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
look for closed cycles of methods i.e. unused code that would not
otherwise be found by the unusedmethods loplugin
Change-Id: I3fb052132d97aabca573bb8e9fc424201b1e2042
Reviewed-on: https://gerrit.libreoffice.org/60875
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
look for collection-like fields that are never added to, and are
therefore effectively unused
Change-Id: Id52c5500ea5e3d2436fb5915aebb86278bf2d925
Reviewed-on: https://gerrit.libreoffice.org/60661
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
...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>
no need to init smart pointers with nullptr, they all have default
constructors that do this already
Change-Id: Ief20c060daa0def8c1aa82f1cf8dc4bc696761e9
Reviewed-on: https://gerrit.libreoffice.org/59818
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
look for fields which are only assigned to in the constructor, so they
can be made const
Change-Id: I0b76817c2181227b04f6a29d6a808f5e31999765
Reviewed-on: https://gerrit.libreoffice.org/60393
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
revert the part of
commit 9f4d23c151
Date: Mon Aug 13 17:24:26 2018 +0200
filter out some of the AST in the plugins
that applied to this plugin. Turns out it really needs to see __all__
the code in order to produce good results.
Change-Id: If580a701049d2570f2a833327b2189641090079b
Reviewed-on: https://gerrit.libreoffice.org/60279
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
look for places where we are appending the temporary result of adding
strings together, to an OUStringBuffer, where we could rather call
append repeatedly and avoid the temporary creation
Change-Id: I481435124291ac7fb54b91a78344a9fe5b379a82
Reviewed-on: https://gerrit.libreoffice.org/59708
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
I seem to have missed quite a few in
commit 9f4d23c151
filter out some of the AST in the plugins
This nets me another 14% improvement
Change-Id: I39b980b49ced560f768045dbedd3ddfef29306c1
Reviewed-on: https://gerrit.libreoffice.org/59501
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
