Commit Graph

587 Commits

Author SHA1 Message Date
Neil MacIntosh
d023ad51a7 Merge pull request #90 from john-lynch/master
Fixing move constructor/semantics of Final_act
2015-09-28 23:15:24 -07:00
john-lynch
ae24c0fe06 Added test that Final_act object can be moved properly. While this is not a common case, it may happen if the user calls finally and the compiler fails to perform RVO. 2015-09-29 00:03:15 -05:00
john-lynch
1d11cd1ed1 Fixing typo in move constructor implementation that resulted in compile error during tests. 2015-09-29 00:00:21 -05:00
john-lynch
422e7164d5 Fixing move constructor of Final_act to take non-const r-value reference and move properly from other Final_act object so that correctness is not dependent on copy elison. 2015-09-28 23:40:25 -05:00
Neil MacIntosh
065f4880c3 Merge pull request #88 from trebconnell/fixmaybenull
Fix issue #49
2015-09-28 18:54:31 -07:00
Treb Connell
444bf9640a Fix issue #49 2015-09-28 18:49:53 -07:00
Neil MacIntosh
318cee0489 Merge pull request #87 from trebconnell/fixmaybenull
Fix issue #50
2015-09-28 18:40:57 -07:00
Neil MacIntosh
6a1179974d Merge pull request #86 from kevgs/master
it's better to pass a small (e.g. 16 bytes) object by value then by c…
2015-09-28 18:39:26 -07:00
Treb Connell
83333419de Add test that reproduces issue 2015-09-28 18:34:04 -07:00
Treb Connell
b29566628e Revert "Fix issue #49"
This reverts commit a46d6fcf0d.
2015-09-28 18:26:35 -07:00
Treb Connell
a46d6fcf0d Fix issue #49 2015-09-28 15:17:37 -07:00
Treb Connell
27ff004e84 Merge branch 'master' into fixmaybenull
Conflicts:
	include/gsl.h
	tests/maybenull_tests.cpp
2015-09-28 14:56:08 -07:00
Kosov Eugene
e46160c18e it's better to pass a small (e.g. 16 bytes) object by value then by const reference 2015-09-28 23:20:28 +03:00
Neil MacIntosh
08064f8521 Merge pull request #85 from kevgs/master
fix clang -Wunused-parameter warnings
2015-09-28 11:28:55 -07:00
Kosov Eugene
3402b92ef6 fix clang -Wunused-parameter warnings 2015-09-28 21:20:02 +03:00
Neil MacIntosh
3571792058 Merge pull request #78 from kernhanda/kernhanda/heterogenous-assignment-not_null
Add various copy assignment operators to not_null and maybe_null_*.
2015-09-28 07:05:43 -07:00
Gabriel Dos Reis
6554e83c79 Macro expand constexpr to nothing under MSVC
Replace `_CONSTEXPR` with plain `constexpr`.
2015-09-28 05:10:44 -07:00
Gabriel Dos Reis
4e596761eb Update list of known platforms where GSL was successfully tested 2015-09-28 04:43:59 -07:00
Kern Handa
7d8e609faf Merge remote-tracking branch 'upstream/master' into kernhanda/heterogenous-assignment-not_null 2015-09-28 07:37:06 +00:00
Kern Handa
783eaabf9d Add various copy assignment operators to not_null and maybe_null_*.
Also removed unused constant member variable that seemed to be there
to prevent maybe_null_* being used with anything other than a pointer,
which is being taken care of with a static_assert now.
2015-09-28 07:35:18 +00:00
Neil MacIntosh
62226021a8 Merge pull request #74 from neilmacintosh/master
Fixed leak in owner<> test. Ha!
2015-09-27 18:18:14 -07:00
Neil MacIntosh
bb169976da Fixed leak in owner<> test. Ha ha ha! 2015-09-27 18:06:51 -07:00
Neil MacIntosh
1a0df71f2f Merge pull request #73 from neilmacintosh/master
Correct fix for int/size_t mismatch.
2015-09-27 16:58:07 -07:00
Neil MacIntosh
99746e2d57 Correct fix for int/size_t mismatch. 2015-09-27 16:53:58 -07:00
Neil MacIntosh
fb91393bb2 Fixing size_t/int mismatch in loops. 2015-09-27 16:25:43 -07:00
Kern Handa
f1be21aa36 Merged Microsoft/GSL:master into branch and resolved conflicts. 2015-09-27 23:25:20 +00:00
Neil MacIntosh
bad8545eee Merge pull request #70 from kernhanda/kernhanda/not_maybe_null
not_null and maybe_null variants should only work on nullptr-assignable types.
2015-09-27 12:43:44 -07:00
Kern Handa
2b6d90436f not_null and maybe_null variants should only work on nullptr-assignable types.
This is in accordance with the GSL.View guidance on not_null and maybe_null
types in the CppCoreGuidelines document.
2015-09-27 19:35:23 +00:00
Gabriel Dos Reis
599a354b1c Merge pull request #59 from kernhanda/kernhanda/size_t
Rank and dimensions should be size_t.
2015-09-27 12:21:22 -07:00
Gabriel Dos Reis
66131c985d Merge pull request #69 from bashrc-real/patch-1
Guide::finally can make use of move semantics
2015-09-27 12:17:27 -07:00
saurabh singh
437791e504 GSL::finally can make use of move semantics
for eg consider this case
[code]
string value = "someVeryLongErrorMessageIAm";
finally([value] { PrintErrorMessage(value); }
[/code]
With the current changes before the call to PrintErrorMessage there will be 3 calls to copy constructor for string(1 when it's captured in closure, 2nd when finally is called and 3rd when it's passed to Final_act . With my patch there will be 1 call to the copy constructor and 2 to the move constructor for the scenario in example, so 2 potential deep copies will be saved for some objects.
Validated that code builds from root, and all tests pass after my change. Also validated that indeed copy constructor calls are saved for objects that support move semantics.
2015-09-27 16:11:12 +05:30
Seth Cantrell
96e5757181 Add .clang-format file
To have any hope of achieving consistent formatting there needs to be
a standard and a mechanism for enforcing it. This is a first step in
that direction.

This format specification is intended to mimic the existing style
insofar as the current source has any consistent style. For example
the determination as to whether ref qualifiers should bind to the
type or variable name was made by trying both and seeing which
produced fewer changes. One exception is the use of tabs vs. spaces,
which was decided based on a comment by neilmacintosh [here][1].

To use clang-format with Visual Studio [download][2] and install the
clang-format plugin. In VS ctrl-r, ctrl-f formats the current line or
text selection.

[1]: https://github.com/Microsoft/GSL/issues/55
[2]: http:llvm.org/builds/
2015-09-26 18:00:54 -04:00
Kern Handa
c4f9b87d96 We should be using standard algorithms where possible.
Use of algorithms in the STL should be promoted where possible.
Also fixed up some whitespace issues.
2015-09-25 17:11:17 -07:00
Neil MacIntosh
b40d3466c6 Merge pull request #52 from trebconnell/addheaderguards
Fix issue #39: Add header guards
2015-09-25 15:17:05 -07:00
Treb Connell
1a791992a0 Add equality operators to maybe_nul_ret 2015-09-25 12:16:39 -07:00
Kern Handa
e157026807 Rank and dimensions should be size_t. 2015-09-25 00:42:38 -07:00
Treb Connell
51da13607c Fix issue #39: Add header guards 2015-09-24 18:08:34 -07:00
Treb Connell
35fb11853f Fix issues #48 #49 #50 2015-09-24 18:02:37 -07:00
Neil MacIntosh
8ae77b1fd5 Merge pull request #46 from trebconnell/fixmaybenull
Fix issue #45: comparing two maybe_null_dbg's can cause fail_fast
2015-09-24 15:22:43 -07:00
Treb Connell
996aa06e08 Fix issue #45: comparing two maybe_null_dbg's can cause fail_fast 2015-09-24 14:09:40 -07:00
Gabriel Dos Reis
0a88570c26 Merge pull request #29 from tamaskenez/master
add travis testing (gcc5, clang36)
2015-09-24 13:34:21 -07:00
Gabriel Dos Reis
2ae46395fa Merge pull request #30 from PSPDFKit-labs/spelling
Fixes spelling of Xcode
2015-09-24 07:53:51 -07:00
Peter Steinberger
2c8886bdfb Also renames MAC OS to OS X 2015-09-23 23:19:18 +02:00
Peter Steinberger
c3b85f9571 Fixes spelling of Xcode 2015-09-23 22:12:49 +02:00
Tamas Kenez
9b454b7a9b add travis testing (gcc5, clang36)
- also relaxed CMake version to 2.8.7, the version default in travis
2015-09-23 17:43:36 +02:00
Andrew Pardoe
186001611a Merge pull request #25 from RicoAntonioFelix/master
Corrected grammatical error...
2015-09-22 11:19:16 -07:00
Rico Antonio Felix
1228e17762 Corrected grammatical error... 2015-09-22 13:38:47 -04:00
Andrew Pardoe
e7c0b89ed4 Removed reference to Microsoft private fork 2015-09-21 19:58:25 -07:00
Neil MacIntosh
670c59a0b1 Merge pull request #24 from galik/master
Issue #23   missing header <stdexcept>
2015-09-21 18:15:41 -07:00
galik
19c5a89bca Merge remote-tracking branch 'upstream/master' 2015-09-21 23:08:35 +01:00