find places where we do not need to be passing a parameter to a
function, because that function has a default value which matches the
value we are passing.
Change-Id: I04d1fd6275204dd4925e6563282464f461123632
(This is no issue for OUString, as it has two ctors taking a single pointer
argument, so passing a null pointer is ambiguous anyway.)
Change-Id: I36f44b29eb84eb83e284fa080d706eabb4b485d5
...from most rtl/bootstrap.h functions. They are effectively only called from
C++ code (there is no plain C UNO binding), so it should be fine to let std
exceptions (like bad_alloc or length_error) propagate from their implementations
to call sites.
(The exception is rtl_bootstrap_args_close, which is typically called from C++
dtors, so should not throw anyway.)
This would strictly speaking be an [API CHANGE], but it should make no practical
difference whether a process terminates abruptly because an exception cannot
pass through a SAL_THROW_EXTERN_C() nothrow specification or because legacy
client code does not expect exceptions to be thrown from functions from which
SAL_THROW_EXTERN_C() has now been removed.
Change-Id: I08e8479e9c5731e46021aadd6a725c1793024d10
This reverts commit 5cba714b4d03ed54debf71534ad8c8edc383a01e, now including a
workaround for <https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53658> "internal
compiler error -- segmentation fault."
Change-Id: I31f6d9ddcb0b884134703df2b9dc1800ba0a84be
This reverts commit 4d4f3512db0cf0bf47c2ba1b39c3813842903ef7, at least "gcc
version 4.8.3 20140627 [gcc-4_8-branch revision 212064] (SUSE Linux)" fails with
"include/rtl/stringutils.hxx:175:49: internal compiler error: Segmentation
fault."
...by making all places that accept a string literal via ConstCharArrayDetector
also accept an OUStringLiteral1 via ConstCharArrayDetector (which required some
tweaking of the ConstCharArrayDetector internals). That removes the need for
special-purpose OUStringLiteral1 overloads, and will allow OUStringLiteral1 to
be used in more places.
Change-Id: I370de8480e02f8423cde5677dd38479b81bccdb2
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>, modifying the patch to
carefully keep the undefined behavior in the already existing additions that may
potentially overflow, instead of making the static_cast<sal_Int32> introduction
look like end - pData->buffer will be guaranteed to fit into sal_Int32 (which it
is not).
Change-Id: Id2fb64dc4585dae34257be6968a8905254a3b42d
...with numeric values that are not meant to be characters. Inspired by
<https://gerrit.libreoffice.org/#/c/16339/> "OUString: add constructor from
single Unicode code point" and ecaf1cb3861c1b519a1495487fc796884cf37b92 "seems
to be what was meant." Found e46e8d19458fd64ff20b1013e5eeabd07e62379c "Shall
this produce a string representation of a number?" and
f2a0e4032734d6be635f14ade3dea499c29ec6c2 "Don't truncate a UTF-32 code point to
a UTF-16 code unit."
(LIBO_INTERNAL_ONLY mostly because it uses a C++11 delegating ctor for
simplicity, but also because it requires mild modifications to client code, see
additional changes in this commit and some other recent commits.)
Change-Id: I8a156f01d1b772c6052d7afaa088ae13ed234e2b
Document that in README.lifecycle; the problem is that our constructors
currently take and release references left/right on the object being
created, which ... means we need an initial reference.
Change-Id: I5de952b73ac67888c3fbb150d4a7cde2a7bc9abf
Create 4 new formats enums rtl_math_StringFormat:
rtl_math_StringFormat_E1, rtl_math_StringFormat_E2,
rtl_math_StringFormat_G1, rtl_math_StringFormat_G2
to 1 or 2 digits in exponent for scientific notation.
Set General format to use rtl_math_StringFormat_E2.
Set trendline equation in status bar to use rtl_math_StringFormat_E1
Change-Id: I41466a6d4ba808ba5b9b38ba252b37c6b4560f12
Reviewed-on: https://gerrit.libreoffice.org/14562
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
Also remove some now redundant asserts from headers.
Some of these actually trigger on unit tests so are commented out.
Change-Id: I07c6b2b2bd175361691a141f22eec584e3ab8f0b
...by adding some further SAL_DLLPUBLIC_RTTI type annotations (cf.
b4f6b26b5a1a78fecfa95ec2eb7ac8b80495d8aa "SAL_DLLPUBLIC_RTTI for proper RTTI
visibility for LLVM") and by making sure relevant function types do not use
incomplete types in their parameter and return types (which would make the RTTI
hidden).
Change-Id: Id7aadcbc0704b9759968ae36266fc9ce11a2e340
...to harmonize the mechanisms to not define LIBO_INTERNAL_ONLY for extension
code and CppunitTest_odk_checkapi. (It also needs to revert any -Wundef, as the
config_host/config_*.h files will not be included, so the HAVE_* macros will be
undefined.)
This obsoletes the need for RTL_DISABLE_FAST_STRING.
Change-Id: If5eacba80c349efb90762aade8f2ea6d2db7e314
...for cases where ? "a" : "bb" does not work, as well as to work around the
MSVC bug for cases like ? "a" : "b".
Change-Id: Id404716047aca5cc81440f291616d92365379b8f
...to use single ASCII character literals "more directly" in the OUString API
(instead of having to go via an intermediary OUString ctor call). Especially
useful for character literals that are defined as const variables or via macros
("direct" uses of character literals in the OUString API can often simply be
replaced with single-character string literals, for improved readability).
(The functions overloaded for OUStringLiteral1 are those that are actually used
by the existing LO code; more could potentially be added. The asymmetry in the
operator ==/!= parameter types is by design, though---writing code like
'x' == s
is an abomination that shall not be abetted.)
Change-Id: Ic5264714be7439eed56b5dfca6ccaee277306f1f
