A GCC information message starting with 'in constexpr expansion'
was incorrectly parsed as an error.
Change-Id: If6142f8486e97c088315060c826d59e61566692b
Signed-off-by: Stephan Oostveen <stephan.oostveen@nextlevel-electronics.com>
SWT.COLOR_GRAY is gray in dark or light mode, but COLOR_TEXT_DISABLED_BACKGROUND
changes with the theme so always leaves the text legible.
Change-Id: I3d0c2d1a19baa618ef08c5c40e35fbdda11691e2
The select can be called twice when the expansion first opens,
so clear out the old selection before processing the new one.
If this isn't done, the oldStyles gets set to the highlighted
version on the second call, which prevents deselect from
removing the highlight.
Change-Id: Ibc8ae45d4164d380b7946cd492c29f33c9fac981
Each source file could have few custom templates. If user selects a
specific template from the combo box for a specific file type, this
selection will be persistent for next file creation, but only for same
file extension.
Change-Id: I7dde198d6fefdfbbc124b8c52517dfe751009fca
Signed-off-by: Lidia Popescu <lidia.popescu@windriver.com>
Because the dll/jnilib is modified in place, the natives are qualified
with their build date, rather than their git date as the checked-in
libraries are not signed.
Change-Id: I3078f5040f7ef9590bb4ab5d031dcb29b3c3bdde
The changes adds sorting for source templates in alphabetic order.
Makes possible to bring to the top the new custom templates provided by
extension points, as the most frequently used ones and to avoid the need
for using the drop down list for the right option.
Change-Id: I931bd2fc08f3e37178a64e4b7908db73af1fbc2c
Signed-off-by: Lidia Popescu <lidia.popescu@windriver.com>
- hover with ctrl+mouse underlines word under cursor
- ctrl-click tries to open the word:
- if a relative path (not starting with /) a full path is
obtained by prepending the shell cwd
- if the fullpath maps to a workspace file, it is opened
- otherwise open the OpenResource dialog with the word as
filter text
- if there is line/column information (separated by colons)
then the opened editor jumps to that line
- http and https words are opened in a browser window
Change-Id: I3f46accbf1eac6743d7b0c3b34bf30ac5e7523bb
Signed-off-by: Fabrizio Iannetti <fabrizio.iannetti@gmail.com>
Also-by: Jonah Graham <jonah@kichwacoders.com>
Signed-off-by: Jonah Graham <jonah@kichwacoders.com>
The ThrowNew JNI method requires the message to be encoded in
"modified UTF-8". The FormatMessage WinAPI method can return a string
using any encoding, so it needs to be converted to UTF-8 in order
to have it visible in the exception message.
To further help, try to extract the message in English and fall back
to the Windows installation language as a last resort.
Prefix the error message with the error code from the GetLastError()
function.
Contributed by STMicroelectronics
Change-Id: Id76ffd83e2d3ad1f061780c7ee0892c9b378649b
Signed-off-by: Torbjörn SVENSSON <torbjorn.svensson@st.com>
Java 16 restricts some reflection. The serializing of scannerInfoCache
with gson does some reflection on Pattern which isn't allowed.
Fortunately the reflection is not needed, we just have to prevent
gson from calling setAccessible by providing our own serialization
for the IExtendedScannerInfo
Change-Id: I0db5cec9cfec1ac4caabd2e67f1d6a336e361023
rebuilt docker images have small binary differences
in the jnilibs due to a rebuilt macOS toolchain being
used.
Change-Id: I25dc951e1e8c7c94a799d63a4c0db644d1f1be1e
Added missed license headers
Updated baseline for the parent pom to CDT 10.2
Updated version for pom.xml to 10.3
Updated version for features and bundles to CDT 10.3
Updated copyright for about.properties to be 2021
Incremented version + 100 where needed
Change-Id: I79666fcc0402fee6607499d7dce1eaf87a5f446d
Signed-off-by: Alexander Fedorov <alexander.fedorov@arsysop.ru>
The "filterPattern" attribute is used to filter templates by language
ID, not by build configuration ID.
Change-Id: Idd89fffc2ec3aefc3c1421b4cbbc7e1a19dbb7f3
to exclude flaky and slow tests from gerrit runs and main build
tests can be tagged as such. See BTreeExpensiveTests for example
of a slow test and Bug_303953Test for an example of a flaky one.
The root README.md has a few notes on converting tests to JUnit5
and adding annotations to mark them slow or flaky.
Change-Id: I03a4004112e6a500d8ec2771d68f14f7dc5f67fb
This change adjusts behavior in ElfParser.hasInterpProgramHeader()
(added for bug 512822) to throw an IOException that occurred, as opposed
to logging the exception and continuing. This excludes exceptions thrown
by Elf.getSections() indicating that ELFhdr.ELFCLASSNONE is read; a new,
more specific IOException type is defined to allow handling this as a
special case.
A "generic" IOException indicates an I/O problem with the binary. E.g.
the binary can be deleted by the user or by Eclipse, while being visited
in CModelManager. If hasInterpProgramHeader() returns instead of
throwing, more problems can occur "later on".
Change-Id: I8852d9cbaa17dc97f668bb1666e9b046bbde1fca
Signed-off-by: Simeon Andreev <simeon.danailov.andreev@gmail.com>
When GCC skips template instantiation in its output it was parsed
as an error instead of an informational message.
Change-Id: If5dca2d1430a6b5e9cb9317649dea8291bfc0356
Signed-off-by: Stephan Oostveen <stephan.oostveen@nextlevel-electronics.com>
This change adjusts replaces the following actions (part of action set
org.eclipse.cdt.ui.buildConfigActionSet) with commands and menus:
* org.eclipse.cdt.ui.manageConfigsAction2
* org.eclipse.cdt.ui.buildConfigMenuAction
* org.eclipse.cdt.ui.wsselection
This is done to avoid menu extension errors on perspective
customization, due to problematic code in CustomizePerspectiveDialog. In
particular the customize perspective dialog will populate main menu
submenus only after going over actions; this causes the menu paths of
the actions above to be detected as invalid, despite actually being
valid and functional.
The original action classes are deprecated and marked for removal. The
respective classes are defined in non-internal packages; removal would
be considered API breakage and so must be done in a major version bump.
Change-Id: I31517697689772395b7e1868ef4cab07ad946085
Signed-off-by: Simeon Andreev <simeon.danailov.andreev@gmail.com>
This change renames all suites to be called ...TestSuite so that
the default includes don't pick them up. With the name ...Tests it
means that those tests referenced by the suite are run multiple times.
Tests which relied on being run as part of a TestSuite have been
refactored to run as individual tests by moving the logic from
the suite into the test (e.g. CommentHandlingTestSuite)
Tests that were not runnable have been renamed from ...Test to ...Tester
to match a convention already used in CDT.
Lots of tests did not use standard name (i.e. didn't end in no Test) so
this was fixed.
Many tests were really abstract tests, so the missing abstract was added,
e.g. ArrayTests
The default excludes in Maven behaviour differently than may be expected,
so rather than remove static inner test classes, update the excludes,
see pom.xml change
Change-Id: Ia91e12fe76c3ec2d914463a28400d21b9daf1910
This test started failing (and wasn't in testsuite previously) because
it had significant trailing whitespace in some tests.
The fix is to use ${whitespace_eol} which I have also added to some
of the javadocs to make it easier to find next time.
Change-Id: Ib364d8a400bcdeb21445dde8428e0fd4c77db990
This style is the JUnit3 way of doing JUnit4 @AfterClass. It was
commented out as part of Bug 564002 (commit 9e303185f9).
Change-Id: I56e2754f3cce447c47546c91cdc16b302e18b4a6
The lines are printed by the GNU ld with the flag --print-gc-secions
Change-Id: Ie09229427a29e9959f8ffe4920a1645ad3dcb086
Signed-off-by: Torbjörn SVENSSON <torbjorn.svensson@st.com>
This change adds a close action
(org.eclipse.ui.console.actions.CloseConsoleAction) to the CDT build
console.
CDT accessing a closed console via BuildConsoleManager will result in
re-adding the respective console page to the console view, if the
console page was closed.
Change-Id: Ifc4d4c6ed329f1c1c7e70f7903ed660ba85306c5
Signed-off-by: Simeon Andreev <simeon.danailov.andreev@gmail.com>
This change adds an action to the CDT build console, to disable
activation on build output. The action uses the existing preference
"build console on top when building" from C/C++ -> Build -> Console.
The action icon is copied from platform debug UI, see:
org.eclipse.debug.internal.ui.IInternalDebugUIConstants.IMG_ELCL_STANDARD_OUT
org.eclipse.debug.ui/icons/full/elcl16/writeout_co.png (and @2x)
The change also ensures the existing preference works, as it currently
has no effect on not showing the build console on CDT build.
Change-Id: I0d94583b85e1a13f18d43de8c10355ccdab7f259
Signed-off-by: Simeon Andreev <simeon.danailov.andreev@gmail.com>
Add enforecment script that verifies that only specified file types are
allowed to be mared as executable.
Change-Id: I4b40e3a46b03bfc78a9dcd52e8bf29cb079b7f3b
Signed-off-by: Torbjörn Svensson <azoff@svenskalinuxforeningen.se>
Advapi32Util.registryGetValues() is more efficient to fetch all the
values since the regiter only needs to be opened once.
XXX: Due to a cycle in the dependencies we need to split this
in two parts, this commit handles the core.native API change
and the next commit handles the use of the new API
Change-Id: Ifd9f1ccc44c652ef3b517278bd342a486155c5fb
Signed-off-by: Torbjörn Svensson <azoff@svenskalinuxforeningen.se>
ICPPUnaryTypeTransformation was meant to be used only when the
transformation is applied on a dependent type. But it was actually
always used when creating types for decl specifiers, regardless if a
dependent type was involved or not. The untransformed type was causing
issues because code dealing with ITypes doesn't apply the transformation
everywhere. It seems better to apply the transformation early when
possible and let the rest of the logic intact.
Change-Id: I1b6d77a857e901f71f00e935e75d24cea87c3118
Signed-off-by: Marc-Andre Laperle <malaperle@gmail.com>
Don't define __STDC__ for clang-cl, just like msvc. Added a new compiler
type to differentiate clang-cl from clang in order to be able to add the
define just for clang and not clang-cl. I initially thought that the
unresolved symbol was caused by a missing built-in symbol (_alloca) so I
started to implement MSVC as a separate language from GNU and not
pollute GCCBuiltinSymbolProvider with a new concept of compiler type,
etc. But it turns out _alloca is also in declared in the headers and was
inactive because of the wrongly defined __STDC__. So this version of the
fix of adding a compiler type to GPPScannerExtensionConfiguration is
much less intrusive and risky, although adding a new language could have
its merit in case of more substantial change.
Change-Id: Ieb27b7dbcc531b8fd3fe30777cd7f343fd3ba66f
Signed-off-by: Marc-Andre Laperle <malaperle@gmail.com>
