Needed so help index is rebuilt against latest lucene so it can be
usable.
Change-Id: Ifc9c65d50e6c9ea0e2735ec7587c09e6f46699b0
Signed-off-by: Alexander Kurtakov <akurtako@redhat.com>
- Return ProblemBinding if conversion from init list failed
- In conversion: don't allow implicit conversion with explicit ctor
Change-Id: I3145b89df778a035ced9999aff4d8a4164eac17f
Signed-off-by: Hannes Vogt <hannes@havogt.de>
The automated license change didn't handle this case properly. Add the
missing #s.
Change-Id: I11ad2c22b3be7b4d3db66aa69386619007aa8b6e
Signed-off-by: Simon Marchi <simon.marchi@ericsson.com>
Arduino-Connections can be created and edited using Preferences->Remote
Development->Remote Connections.
Change-Id: Ie92330689bdad4987c12d2bd17828dbe5e862b90
Signed-off-by: Lutz Hamann <lhamann@planettime.de>
The ConsoleOutputStream method added in 6e1b9b4 must be overridden here,
otherwise text sent to it does not appear along with the other console
output.
Change-Id: I1a3803ffb8962140537b877f0df328a4037b4dfb
Signed-off-by: Christian Walther <walther@indel.ch>
The original fix (a733900) only fixed part of the problem: It worked for
the case where a complete CExternalSetting was removed and replaced by a
different one, but not in the case where individual entries from a
CExternalSetting were moved to a different one, but others remained
(and, in both cases, the two CExternalSettings applied to the same
ICLanguageSetting).
This commit
- adds a test for the additional condition, which would previously fail
- reverts the previous fix, which is made redundant by the new one
- fixes both cases by applying removals before additions with
ICSettingEntry granularity per ICLanguageSetting rather than for whole
CExternalSettings.
Change-Id: I1b1ee7443b83189c29e458eef12be9cad6b3965d
Signed-off-by: Christian Walther <walther@indel.ch>
- As of [temp.func.order] p5 only parameters for which there are explicit
arguments should be considered.
- Implement [temp.deduct.partial] p11 (resolution of DR1395).
Change-Id: I56262229a6fe4717c22aac814aa9ea42eb37a10d
Signed-off-by: Hannes Vogt <hannes@havogt.de>
HierarchicalProperties already allow for OS-specific overrides, so
1 - convert platformProperties to HierarchicalProperties
2 - change HierarchicalProperties to allow overwriting of non-empty
properties. Before it would only use the platform child if the value for
the entry itself was null. With this change, if a platform child exists
its value is used if
- it contains a property or
- has an empty property and no further children. In this case the value
was specifically overridden with null
This change also allows proper parsing of esp32 platform.txt
Change-Id: Ie13a1bee57c216f6ab37cfc0208ef7711dfc62e8
Change-Id: Ie13a1bee57c216f6ab37cfc0208ef7711dfc62e8
Signed-off-by: Lutz Hamann <lhamann@planettime.de>
In some situations, it makes sense to have more than one GDB instance.
Allow to reuse the GDB controling layer to avoid more code duplication
than neccessary to have a second GDB session.
Change-Id: I641919157c5b86c14ff768c8cd3fcddc9dcdbeb2
Signed-off-by: Torbjörn SVENSSON <torbjorn.svensson@st.com>
Alias templates can be instantiated with dependent arguments, which
can themselves contain a pack expansion, so we need to take care
that pack expansion expressions are instantiated correctly.
This was previously fixed for pack expansions of types and template
arguments in bug 486971.
The patch also fixes a bug in CPPTypedef.getType() where alias
declarations weren't handled correctly. (This bug would only occur
during debugging as normally the type would be computed via setType().
Change-Id: Ie70a923fc9dd0f177b7bfb429b8f1387966b416d
instantiateCommaSeparatedSubexpressions() is a clunky name, and in C++17
(with fold expressions) the function will no longer be limited to comma-
separated expressions.
Change-Id: Id242f58bb291e79cefe2b28db12dbde4bafba4ed
Take number of explicitly passed arguments into account.
Change-Id: If85f072d8b18f3fd1ca28f3c5688e864b0ffb4cd
Signed-off-by: Hannes Vogt <hannes@havogt.de>
We weren't checking the case where the size of the toolchains array
was zero. Return a null config in that case and fix up the
CommonBuildTab to deal with that.
Change-Id: Ibd8f659988f1463570bdb2654d6ea2b4a45866c2
The config is not valid without this. Using the same algorithm CMake
projects do to find the default toolchain.
Change-Id: I871da3019b7d440fbd6c1b2a4935d424f084a603
This target file is not for building releases, but rather to help
early identify whether we have problems coming down the pipe from
CDT dependencies.
Change-Id: I481cffb7076ba20d8293540852435975dbbd7ea8
Some commands, such as -file-list-exec-source-files, can generate
10s of thousands of lines of output.
Change-Id: I07fc3ef99bdd8e9247c75e2e9733e43b35657687
Signed-off-by: Baha El Kassaby <baha.elkassaby@gmail.com>
Also-by: Jonah Graham <jonah@kichwacoders.com>
Provides GDB version detection as a public API that can be used without
having a saved launch config, or having a launch config at all.
Change-Id: Ifaee337899c01fe7066b27f7d66484bfc728f5ed
Signed-off-by: Max Bureck <max.bureck@fokus.fraunhofer.de>
The DSF Examples contains some commented out source that is enabled
via a script so this commit cleans that up for new formatting rules.
Change-Id: I82c5e195cc3746415c659b2c0fc72b8118cdc56c
The previous alignment of all the warnings/ignores
led to too many warnings that weren't there before. This
commit relaxes them a bit.
The core/org.eclipse.cdt.core/.settings/org.eclipse.jdt.core.prefs
is still the "master" copy, with
releng/scripts/check_code_cleanliness.sh containing the
exceptions that apply to test plug-ins.
Change-Id: Ibd4e31ade0b42b31e7cbe5a94f06c6fc15183a56
git ls-files has two benefits:
- Only finds tracked files
This is necessary because we were having a few cleanups
being attempted on ignored files.
- Is much faster
Change-Id: I5a53497f8ef9c839676b7ef7e3be35171140f4b9
