summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Improve python code indentation warning / error messagesdocstring-indentationWilliam S Fulton2015-07-306-23/+55
|
* Function comment header formatting correctionsWilliam S Fulton2015-07-301-46/+66
|
* Improve Python docstring indentation handlingWilliam S Fulton2015-07-305-28/+308
| | | | | | | | | | | SWIG-3.0.5 and earlier sometimes truncated text provided in the docstring feature. SWIG-3.0.6 gave a 'Line indented less than expected' error instead of truncating the docstring text. Now the indentation for the 'docstring' feature is smarter and is adjusted so that no truncation occurs. Closes #475
* Avoid gcc uninitialized variable warnings in Python wrappers.xantares2015-07-251-1/+1
| | | | | | Just initialize the local array with zeroes. Closes #453.
* Merge pull request #467 from vadz/no-callback-in-autodocWilliam S Fulton2015-07-245-16/+19
|\ | | | | Remove callback function from autodoc unit test.
| * Remove callback function from autodoc unit test.Vadim Zeitlin2015-07-194-16/+10
| | | | | | | | | | | | | | | | | | | | It doesn't seem to belong there at all, there is a dedicated callback unit test for it and it was added to the initial version of autodoc.i back in 124253d69828b4323f939bf118254ee96aefcdc7 without any explanation, so just remove it. As this callback was used in a PHP test, perform this test for callback.i now and use "%(uppercase)s" construct inside %callback to test that this works.
| * Make callback unit test pass for PHP backend.Vadim Zeitlin2015-07-192-1/+10
| | | | | | | | | | Deprecated %callback(1) doesn't work with PHP, use "%s" to give the same name to the callback as to the C function explicitly instead.
* | Typemap attribute fixesWilliam S Fulton2015-07-242-50/+52
| | | | | | | | | | | | | | | | Fix for breakages in previous few commits: - Perl test-suite - the "varout" typemap "type" attribute is now expanded in typemap.c instead of Perl.cxx. - The swig_typemap_warn errors testcase showed that $1 was no longer being expanded correctly when used in output typemaps (lname not set).
* | Merge branch 'LindleyF-typemap-in-descriptor'William S Fulton2015-07-2210-6/+516
|\ \ | | | | | | | | | | | | | | | | | | * LindleyF-typemap-in-descriptor: Add documentation and CHANGES for special variables and typemap attributes. Support special variable expansion in special variable macros in typemap attributes. Enable variable and typemap substitution in typemap kwargs, and a test that verifies this works for directorin:descriptor.
| * | Add documentation and CHANGES for special variables and typemap attributes.William S Fulton2015-07-223-0/+109
| | | | | | | | | | | | | | | Also add info about special variable expansions in special variable macros.
| * | Support special variable expansion in special variable macros in typemap ↵William S Fulton2015-07-224-16/+234
| | | | | | | | | | | | | | | | | | | | | attributes. Add test cases for special variable expansions and special variable macros (aka embedded typemaps) expansion.
| * | Enable variable and typemap substitution in typemap kwargs, and a test that ↵Lindley French2015-07-224-0/+183
| | | | | | | | | | | | verifies this works for directorin:descriptor.
* | | Merge pull request #472 from wkalinin/ignore_warningVladimir Kalinin2015-07-231-11/+9
|\ \ \ | |/ / |/| | "ignore" warning (2)
| * | refactoring: 2 ways of ignoring symbol in add_symbols() merged for clarityVladimir Kalinin2015-07-221-11/+9
|/ /
* | Merge pull request #470 from wkalinin/ignore_warningVladimir Kalinin2015-07-211-0/+2
|\ \ | | | | | | missing feature:ignore
| * | explicitly %ignore'd symbol does not get feature:ignore if it is only added ↵Vladimir Kalinin2015-07-211-0/+2
|/ / | | | | | | to C symbol table
* | Octave 4.0 fails in TravisWilliam S Fulton2015-07-191-0/+3
|/ | | | Mark as failing in Travis due to gcc 4.6 internal compiler error
* Merge branch 'm7thon-issue-445-python-class-docstrings'William S Fulton2015-07-182-6/+24
|\ | | | | | | | | | | | | | | | | * m7thon-issue-445-python-class-docstrings: Add changes note for Python tp_doc slot and docstring Set class docstring in tp_doc slot for python -builtin Conflicts: CHANGES.current
| * Add changes note for Python tp_doc slot and docstringWilliam S Fulton2015-07-181-0/+6
| | | | | | | | Issue #461
| * Set class docstring in tp_doc slot for python -builtinMichael Thon2015-07-091-6/+18
| |
* | Clearer variable name in Java director generated codeAnbiru Shouta2015-07-171-4/+4
| | | | | | | | Closes #463
* | warning fixesWilliam S Fulton2015-07-171-1/+1
| |
* | String / char * usage in parser fixesWilliam S Fulton2015-07-171-40/+21
| |
* | const char * fixes in the parserWilliam S Fulton2015-07-171-13/+15
| |
* | Update CHANGES.current and Octave.html to indicate Octave 4.0.0 supportKarl Wette2015-07-172-1/+4
| |
* | Merge pull request #462 from mtmiller/fix-mkoctfileKarl Wette2015-07-171-1/+1
|\ \ | | | | | | Octave: use correct mkoctfile executable
| * | Octave: use correct mkoctfile executableMike Miller2015-07-111-1/+1
| | | | | | | | | | | | | | | Allow a versioned Octave executable to be configured via --with-octave and the correct corresponding mkoctfile executable to be used.
* | | Merge pull request #460 from opoplawski/octave4Karl Wette2015-07-174-16/+59
|\ \ \ | |/ / |/| | Support for octave 4.0.0
| * | Add octave 4.0 to travisOrion Poplawski2015-07-101-0/+2
| | |
| * | Fix default_constructor_runme.m testOrion Poplawski2015-07-101-16/+2
| | |
| * | Add #include <climits> for INT_MAXOrion Poplawski2015-07-101-0/+1
| | |
| * | Update print() signature for octave 4.0Orion Poplawski2015-07-091-0/+12
| | |
| * | Support for octave 4.0.0Orion Poplawski2015-07-081-0/+42
| |/
* | Fix compile flags for Travis C++11 testingWilliam S Fulton2015-07-111-2/+2
| |
* | Travis testing c++11 and gcc5William S Fulton2015-07-102-12/+31
| | | | | | | | | | Modify testflags.py to control the -std= option passed to gcc Install new boost for c++11 with gcc5 testing
* | c++11 test case fixesWilliam S Fulton2015-07-102-2/+3
| |
* | Travis testing extended to test c++11 and c++14 with gcc-5William S Fulton2015-07-091-7/+20
| | | | | | | | | | | | - SWIG executable compiled with gcc-5 using both c++11 and c++14 standards - cpp11 tests enabled using c++11 standard - Add clang compile using c++11
* | Configuring C++11 compiler flags for testing fixWilliam S Fulton2015-07-091-0/+1
|/
* Don't generate constructor wrappers if a base class has a private constructorWilliam S Fulton2015-07-0713-72/+45
| | | | | | | | | | | | | | | g++-5 errors out with this now with errors such as: default_constructor_wrap.cxx:665:27: error: use of deleted function ‘FFF::FFF()’ result = (FFF *)new FFF(); ^ default_constructor_wrap.cxx:314:7: note: ‘FFF::FFF()’ is implicitly deleted because the default definition would be ill-formed: class FFF : public F { ^ default_constructor_wrap.cxx:301:4: error: ‘F::~F()’ is private ~F() { } ^ default_constructor_wrap.cxx:314:7: error: within this context
* Test-suite fixes for c++11 compilation by g++-5.1William S Fulton2015-07-063-8/+4
|
* Fix incorrect test case codeWilliam S Fulton2015-07-061-1/+1
|
* Bump version to 3.0.7William S Fulton2015-07-066-207/+211
|
* Unused method warning suppression for Javascript v8v3.0.6rel-3.0.6William S Fulton2015-07-051-18/+18
|
* Add 3.0.6 release notes and release dateWilliam S Fulton2015-07-055-5/+10
|
* Html doc fixesWilliam S Fulton2015-07-054-9/+11
|
* Testcase workaround for SolarisWilliam S Fulton2015-07-051-0/+4
|
* parent_class testcase name warning fixes for PHPWilliam S Fulton2015-07-051-2/+2
|
* Perl5 carrays testcase fixWilliam S Fulton2015-07-041-1/+1
| | | | Number of loops is different since seg fault fix in e543299
* Merge branch 'appveyor-check-test-suite'William S Fulton2015-07-0428-74/+89
|\ | | | | | | | | | | | | | | | | | | | | | | | | * appveyor-check-test-suite: Appveyor testing expanded Fix array overrun in li_carrays testcase Warning fixes in generated Python code for 64bit Visual C++ on Windows. Warning fixes in generated C# code for 64bit Visual C++ on Windows. Warning fixes for 64bit visual c++ on Windows Warning fixes in generated Java code for 64bit Visual C++ on Windows. Warning fixes for 64bit visual c++ on Windows C# gc tests failure fix Add a space between literal and string macro
| * Appveyor testing expandedWilliam S Fulton2015-07-041-2/+17
| | | | | | | | | | | | | | | | | | | | - New variable to control version of Visual Studio to use on appveyor - Enable VS2015 (14.0) for C# - Run full check-test-suite and not just partialcheck-test-suite since Appveyor performance improvements since using dedicated Hyper-V instead of Azure. - Allow 64 bit Python 2.7 to fail on Appveyor as a vector container slicing bug needs fixing.
View Lorry Controller Status