travis: Use g++-7 direcly from packages (and fix ubuntu 16.04 problems) #113
Conversation
marehr
force-pushed
the
travis_upgrade
branch
3 times, most recently
from
6081ecd to
16ef1ee
Compare
marehr
force-pushed
the
travis_upgrade
branch
2 times, most recently
from
3ba7ece to
decd172
Compare
|
Whats going on here ❓ |
|
I try to use the gcc-7 compiler from https://launchpad.net/~ubuntu-toolchain-r/+archive/ubuntu/test, because our current compiler does weird things. Apparently the native gcc-7 compiler has some bug that I try to workaround/fix :( |
marehr
force-pushed
the
travis_upgrade
branch
2 times, most recently
from
f50a890 to
e281a86
Compare
|
What is the status here? |
marehr
force-pushed
the
travis_upgrade
branch
2 times, most recently
from
bafac27 to
c501fa1
Compare
|
@rrahn mergeable |
| { | ||
| //!\privatesection | ||
|
|
||
| /*!\interface seqan3::sequence_concept_modified_by_const_iterator <> |
| { val.erase(val.begin(), val.end()) } -> typename type::iterator; | ||
|
|
||
| // workaround a travis bug where insert/erase can't take a const iterator, e.g. cbegin() | ||
| requires detail::sequence_concept_modified_by_const_iterator_bug || detail::sequence_concept_modified_by_const_iterator<type>; |
There was a problem hiding this comment.
Please add a check, that this is only tested if the queried type is in fact a std::basic_string.
| * failing on travis anymore. | ||
| */ | ||
| constexpr bool sequence_concept_modified_by_const_iterator_bug | ||
| = sequence_concept_modified_by_const_iterator<std::string> == false; |
There was a problem hiding this comment.
Can we not use basic_string? Wouldn't that be safer?
| //!\privatesection | ||
|
|
||
| /*!\interface seqan3::sequence_concept_modified_by_const_iterator <> | ||
| * \brief Can insert and erase used with const iterators |
There was a problem hiding this comment.
Maybe: Checks whether insert and erase can be used with const_iterator.
And maybe also add an explantation why this was separated from the original concept definition,
and that it should go back in, if the travis bug is not of any concern anymore.
marehr
force-pushed
the
travis_upgrade
branch
2 times, most recently
from
78479cb to
0d98402
Compare
marehr
force-pushed
the
travis_upgrade
branch
from
0d98402 to
5ba7526
Compare
marehr
changed the title
#109: switch to a release version of gcc-7 (currently we use an older snapshot)