Age | Commit message (Collapse) | Author |
|
...where the get member function is defined on a std::__shared_ptr base class,
so loplugin:simplifypointertobool used to miss those until now. (While e.g.
using libc++ on macOS found those cases.)
366d08f2f6d4de922f6099c62bb81b49d89e0a68 "new loplugin:simplifypointertobool"
was mistaken in breaking isSmartPointerType(const clang::Type* t) out of
isSmartPointerType(const Expr* e); c874294ad9fb178df47c66875bfbdec466e39763 "Fix
detection of std::unique_ptr/shared_ptr in loplugin:redundantpointerops" had
introduced that indivisible two-step algorithm on purpose.
The amount of additional hits (on Linux) apparently asked for turning
loplugin:simplifypointertobool into a rewriting plugin. Which in turn showed
that the naive adivce to just "drop the get()" is not sufficient in places that
are not contextually converted to bool, as those places need to be wrapped in a
bool(...) functional cast now. If the expression was already wrapped in
parentheses, those could be reused as part of the functional cast, but
implementing that showed that such cases are not yet found at all by the
existing loplugin:simplifypointertobool. Lets leave that TODO for another
commit.
Besides the changes to compilerplugins/ itself, this change has been generated
fully automatically with the rewriting plugin on Linux.
Change-Id: I83107d6f634fc9ac232986f49044d7017df83e2a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94888
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Jenkins
|
|
As https://docs.microsoft.com/fr-fr/windows/win32/api/objbase/nf-objbase-coinitialize?redirectedfrom=MSDN
advised.
Moreover, it'll make concurrency model explicit
Change-Id: I247a49248bf45beb64a7985a7c0d078ee48efe2d
Change-Id: I6e755fd516d9fb2dd2a9df0c66d9afd6070447ea
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94669
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I09806869f2fdbae61f4c5d5c9db6859202bb63b5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94609
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I7e70614ea5a1cb1a1dc0ef8e9fb6fd48e85c3562
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/93904
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Found by:
run-clang-tidy-10 -checks=-*,misc-unused-using-decls
Change-Id: I3e95791e223ef01e140a6217e29a9efae428a784
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90876
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
RT_RCDATA is defined as MAKEINTRESOURCE(10) in WinUser.h, and MAKEINTRESOURCE is
defined as MAKEINTRESOURCEA in WinUser.h absent a definition of UNICODE.
649854544e28f388774492f27eb1b75485202148 "WinAPI: Fix some leftovers having
UNICODE-define-dependency" had changed this
> - HRSRC hrc = FindResourceW(g_hModule, L"#2000", RT_RCDATA);
> + HRSRC hrc = FindResourceW(g_hModule, L"#2000", MAKEINTRESOURCEW(RT_RCDATA));
before 558956dc811a1f0f07411e348f7081a467bbc3b5 "Drop UNICODE/_UNICODE defines"
dropped defining UNICODE. But that change was technically wrong, as now flagged
by LLVM 10 trunk clang-cl:
> C:/lo-clang/core/shell/source/win32/shlxthandler/thumbviewer/thumbviewer.cxx(50,56): error: cast to smaller integer type 'unsigned short' from 'char *' [-Werror,-Wpointer-to-int-cast]
> HRSRC hrc = FindResourceW(g_hModule, L"#2000", MAKEINTRESOURCEW(RT_RCDATA));
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~
> C:/PROGRA~2/WI3CF2~1/10/Include/10.0.18362.0/um\winuser.h(236,51): note: expanded from macro 'MAKEINTRESOURCEW'
> #define MAKEINTRESOURCEW(i) ((LPWSTR)((ULONG_PTR)((WORD)(i))))
> ^~~~~~~~~
There appears to be no explicit define for MAKEINTRESOURCEW(10), and the fix
presented here seems like a cleaner way than replacing the use of RC_RCDATA with
MAKEINTRESOURCEW(10).
Change-Id: I36f60a8647ac264085656fc9ec3b129c2d81f480
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90799
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Jenkins
|
|
This reverts commit ec7e3609917092ee00667c78ff40cd4c2f0adec1.
Change-Id: I11aa028f6562f97b0d251d6b326e9aecb45d03de
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90415
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
See instructions in solenv/gbuild/Trace.mk . This generates a file than
can be viewed e.g. in the Chromium tracing view.
Change-Id: I5f90647c58ca729375525b6daed2d4918adc8188
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/88754
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: Id5babe1e16c8f8a8c72077fc95508a567138766a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87889
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I69cc1b352221ca053ccd0c5b78e926480a8c9ccd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87884
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Iab35a8b85b3ba1df791c774f40b037f9420a071a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86708
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: If0fe2feb6b6b4aa595f69c1a1448a809438ec077
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87508
Tested-by: Jenkins
Reviewed-by: Muhammet Kara <muhammet.kara@collabora.com>
|
|
Unify Windows header usage (wrap into prewin/postwin, remove redundant);
replace "using" with explicit namespace usage; drop some indirection.
Change-Id: Ia635e732de7165b161935c05d54d0490464248b8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87571
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
... which obviously are rejected by SHGetFileInfoW and SHParseDisplayName
that it calls internally.
Change-Id: I2f5f3c675ea6aa1c2d92eef30be4399a8d600255
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87565
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
* In the URIS_ONLY case (which is intended to open documents etc.), set
preprocessed_command to the filesystem pathname the first time that is
computed (and which will no longer fail for file URLs with fragment since
14b36a16b225bf7c988f118d499a7287c47cd83e "Remove a fragment from a file URL
early on").
* In the !URIS_ONLY case (which is intended to run other programs), we will
generally be called with aCommand already being a filesystem pathname. But
even if it should be a file URL, it appears to be OK to pass that as-is to
ShellExecuteExW: At least on Windows 8, passing
"file:///C:/Windows/System32/notepad.exe" does start it.
* The code for <https://bz.apache.org/ooo/show_bug.cgi?id=4789> "Hyperlinks
doesnt works" should no longer be relevant: In the URIS_ONLY case, any
fragment (called a "jump mark" in that code) has already been removed from the
incoming URL now. In the !URIS_ONLY case, we should generally not be called
with a URL with a fragment, but even if we are, it should be OK to pass that
as-is to ShellExecuteExW and let it handle that (see above).
* Similarly, the code for
<https://bugs.documentfoundation.org/show_bug.cgi?id=54204> "Hyperlinks
between documents not works if link contains anchor at the end" is no longer
relevant for the same reason.
Change-Id: Ia6ec80a30f6d0603bccc87b9d6dd93ca6a84c370
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86975
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
...as ShellExecuteExW would ignore it anyway
Change-Id: I969db094bb7d2ea230ac8c36eb23d71a90fbe466
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86868
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I65ac9a8767c298c50606a3b269a227dd1513a112
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86550
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I59667664fc097989e0a78d8876f2be8353911236
Reviewed-on: https://gerrit.libreoffice.org/85701
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
PROPERTYMAP members pszXPathParent and pszValueNodeName were unused ever since
85e0cc246d4350122b17fbe2bfd15635a190ad77 "dv17: #i70994#: Use property handler
instead of column info for Windows Vista" introduced the code. (And removing
them allowed to simplify the code further in the obvious way.)
Change-Id: Ie1247cbee31a172a15fd57c4b752d74711b1d3a1
Reviewed-on: https://gerrit.libreoffice.org/84860
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
* OldEntry in fpicker/source/aqua/resourceprovider.mm was apparently unused ever
since it got introduced with 00657aef09d854c74fb426a935a3e8b1fc390bb0
"migrate to boost::gettext"
* impl_throwError is used from multiple TU,
connectivity/source/drivers/macab/MacabStatement.cxx just missed the relevant
#include
Change-Id: Iba131da57aa20085bb1c634ba9a3a59566070abd
Reviewed-on: https://gerrit.libreoffice.org/84653
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
...plus loplugin:consttobool and loplugin:fakebool fallout
Change-Id: Ie3d8121815c080b13bea6d9deca1eb138ca56138
Reviewed-on: https://gerrit.libreoffice.org/84515
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
which merely announce that the next declaration is a class
Change-Id: Ifdb1398bcd99816b13e0b3769b46d0562bfbc1dc
Reviewed-on: https://gerrit.libreoffice.org/84229
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I81fea38cd737a8be74e6ece333ca37cc434a1c33
Reviewed-on: https://gerrit.libreoffice.org/83765
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
...plus follow-up loplugin:implicitboolconversion and loplugin:redundantcast
Change-Id: I9fc9c5cb46fbb50da87ff80af64cb0dfda3e5f90
Reviewed-on: https://gerrit.libreoffice.org/83207
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I9da73ceef20a0b2535694438a3a6b20876dccd0f
Reviewed-on: https://gerrit.libreoffice.org/83208
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Where option "-R" means, according to the open(1) man page: "Reveals the file(s)
in the Finder instead of opening them." This appears to always reveal in
Finder, even for *.app "application directories", so should not affect
CVE-2019-9847.)
Change-Id: I775db7bf71c09adc6570931a977b82dfee8dc95a
Reviewed-on: https://gerrit.libreoffice.org/82749
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Non-ASCII file links didn't work when the link contains anchor at
the end.
Note: The "correct" way to convert from a LO-internal to -external
file URL would be the translateToExternal method of the
css.uri.ExternalUriReferenceTranslator UNO service. But that
translates the URL to be interpreted according to the current Windows
code page (i.e., osl_getThreadTextEncoding), so only supports
characters covered by that code page, and given that the result is
passed into a Windows wchar_t API, ToIUri nicely avoids such potential
conversion failure.
Change-Id: I1a5b07366b0fea3da2fbe11a9378aacd765b5e04
Reviewed-on: https://gerrit.libreoffice.org/82219
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I324496ff7c61d87a83b6b378810aa5c78cd7dba3
Reviewed-on: https://gerrit.libreoffice.org/81405
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Replace them with default initialization or calloc
Change-Id: I747f53c2ced2d0473fd5a5ede4f8520a0633dcc1
Reviewed-on: https://gerrit.libreoffice.org/80805
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
which defeat the *StringConcat optimisation.
Also make StringConcat conversions treat a nullptr as an empty string,
to match the O*String(char*) constructors.
Change-Id: If45f5b4b6a535c97bfeeacd9ec472a7603a52e5b
Reviewed-on: https://gerrit.libreoffice.org/80724
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
No idea why that started to show up now, but
C:/Program Files (x86)/Windows Kits/10/Include/10.0.17763.0/um/combaseapi.h
contains
> #define STDMETHOD(method) virtual COM_DECLSPEC_NOTHROW HRESULT STDMETHODCALLTYPE method
while
C:/Program Files (x86)/Windows Kits/10/Include/10.0.17763.0/um/winnt.h
contains
> #define STDMETHODIMP HRESULT STDMETHODCALLTYPE
which caused failures like
> [build CXX] embedserv/source/inprocserv/dllentry.cxx
> embedserv/source/inprocserv/dllentry.cxx(261,40): error: 'QueryInterface' is missing exception specification '__attribute__((nothrow))' [-Werror,-Wmicrosoft-exception-spec]
> STDMETHODIMP InprocEmbedProvider_Impl::QueryInterface( REFIID riid, void ** ppv )
> ^
> embedserv/source/inprocserv/dllentry.cxx(143,15): note: previous declaration is here
> STDMETHOD(QueryInterface)(REFIID riid, void ** ppvObj) override;
> ^
Change-Id: I0fe3554c2da4089bf0f883e1132d6f2ee95ae2c3
Reviewed-on: https://gerrit.libreoffice.org/79970
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I3d40104d6822ebb80ebec44b36bcdfa5c7e953ed
Reviewed-on: https://gerrit.libreoffice.org/79960
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Id1bcc878f40fba804f78f7a5b7cd463db8dc5988
Reviewed-on: https://gerrit.libreoffice.org/79957
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Ie55967043fd6584c8b9a0ad66fad118c70bda8f0
Reviewed-on: https://gerrit.libreoffice.org/79927
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Ib8b2bc1c5f7b27a646036ce23cae2b6a06edd038
Reviewed-on: https://gerrit.libreoffice.org/79922
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Id1a82cea4444255fdb693e126b7571a406094624
Reviewed-on: https://gerrit.libreoffice.org/79916
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I08da288a88c2bce1d4250ec77f17bd483e6bc09c
Reviewed-on: https://gerrit.libreoffice.org/79911
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I13f626c29ceeafbca91b06e5efd11b79b0f25a76
Reviewed-on: https://gerrit.libreoffice.org/79679
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Thanks to the Paul's (paul@claws-mail.org) contribution
Change-Id: Ie4144ab78691b33d8d341d4520c40c56010a0dd8
Reviewed-on: https://gerrit.libreoffice.org/79423
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
To complete this:
https://gerrit.libreoffice.org/#/c/78312/
This is a massive replace for
".." instead of "..." between words.
It passed "make check" on Linux.
Change-Id: I144d8061fca9f545c762941551e59dffdd3650e8
Reviewed-on: https://gerrit.libreoffice.org/78357
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Jenkins
|
|
Change-Id: Iee965c3f720827b20347f6228e891562c8295d22
Reviewed-on: https://gerrit.libreoffice.org/78327
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
... to executables on Windows
Change-Id: I31bd2447052915bb299820034f99128e0e97db54
Reviewed-on: https://gerrit.libreoffice.org/78290
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
'strdup' doc [1] mentions:
"The returned pointer must be passed to 'free' to avoid a memory leak."
This silences a "Mismatched free() / delete / delete []" error that
valgrind outputs.
[1] https://en.cppreference.com/w/c/experimental/dynamic/strdup
Change-Id: I28877be762256a7b995c09415a8ad9977b7998c4
Reviewed-on: https://gerrit.libreoffice.org/77974
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: I40e2a1be0ce7df627d3820192d24821ddd7c3e2f
Reviewed-on: https://gerrit.libreoffice.org/77527
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ic431ef6c3555f02fbc204a5b0af5f9bfe62f4a30
Reviewed-on: https://gerrit.libreoffice.org/77286
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: Icb9d9e1cd21e2506e36fe40a3b93b6a2521a868c
Reviewed-on: https://gerrit.libreoffice.org/77239
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
This only fixes part that the URL refuses to open the target file.
Honoring fragment isn't fixed here, since it's the system call to
ShellExecuteExW that in this case internally converts the file URL
into a system path, and strips the fragment from it.
Regression from commit d59ec4cd1660410fa1b18c50d2d83b1417a82ddc.
Change-Id: I6c9ed27e9a5bd7f2780dd3be96f816a6e825e043
Reviewed-on: https://gerrit.libreoffice.org/76778
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: I0045eac5f2395f56dcb1ed8a958c93c28de496a9
Reviewed-on: https://gerrit.libreoffice.org/76647
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
There is already a large comment in initAttachmentList about
problems with MAPI_DIALOG_MODELESS. Using MAPI_DIALOG_MODELESS with
Outlook 2016 and multiple attachments sometimes produces a
MAPI_E_FAILURE, but most times it crashes for me. And it's not a
problem with the removed temporary files, as uncommenting that
shows the same problem. And there seems to be many more problems:
https://social.msdn.microsoft.com/Forums/en-US/5d8fece6-9d93-490c-9331-625c17e3291d/mapisendmailhelper-and-mapidialogmodeless
But actually I don't see any blocking of LO, if I switch to
using MAPI_DIALOG, as senddoc.exe already runs in the background.
So this switches MAPI_DIALOG_MODELESS to MAPI_DIALOG.
This reverts commit 5874c76371562c3e2d8564b1fb04df1997091d27
("tdf#116074: Use modeless dialogs with supporting mailers").
Change-Id: Ie0f8f22196d1a174dfeada2bc4aabb1717ee16a7
Reviewed-on: https://gerrit.libreoffice.org/76155
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
It passed "make check"
Change-Id: I50c3639ec258c7fb6d73104f8e777f96995aa8e3
Reviewed-on: https://gerrit.libreoffice.org/76543
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|