Fix compiler and linker warnings

[betatim]

Some trivial fix ups to reduce the number of compiler warnings generated by khmer.

@luizirber do you know what the -mmacosx-version-min=10.7 really does? I don't really know more than what you can deduce from the name, which makes me wonder if adding it breaks something (we aren't testing for)?

• Is it mergeable?
• make test Did it pass the tests?

[codecov-io]

Codecov Report

Merging #1605 into master will not change coverage.
The diff coverage is 0%.

@@          Coverage Diff           @@
##           master   #1605   +/-   ##
======================================
  Coverage    0.06%   0.06%           
======================================
  Files          78      78           
  Lines        9792    9792           
  Branches     2457    2457           
======================================
  Hits            6       6           
  Misses       9786    9786

Impacted Files	Coverage Δ
src/oxli/assembler.cc	0% <0%> (ø)	⬆️
src/oxli/traversal.cc	0% <0%> (ø)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update edabebf...80b32b6. Read the comment docs.

[betatim]

Ready for review! @luizirber, @camillescott, @standage , or @ctb

[betatim]

Somewhat related, my terminal is filling up with large number of copies of this warning. It has "always" been there but usually only once, is the duplication related to the multiple template instantiations we do for the read parser?

In file included from third-party/seqan/core/include/seqan/basic.h:62:0,
                 from third-party/seqan/core/include/seqan/seq_io.h:44,
                 from lib/read_parsers.hh:41,
                 from lib/hashtable.hh:57,
                 from lib/kmer_filters.cc:40:
third-party/seqan/core/include/seqan/basic/test_system.h:92:10: warning: ‘template<class> class std::auto_ptr’ is deprecated [-Wdeprecated-declarations]
     std::auto_ptr<Test> instance;
          ^~~~~~~~
In file included from /usr/include/c++/6.3.1/bits/locale_conv.h:41:0,
                 from /usr/include/c++/6.3.1/locale:43,
                 from /usr/include/c++/6.3.1/iomanip:43,
                 from third-party/seqan/core/include/seqan/basic/debug_test_system.h:46,
                 from third-party/seqan/core/include/seqan/basic/basic_debug.h:52,
                 from third-party/seqan/core/include/seqan/basic.h:52,
                 from third-party/seqan/core/include/seqan/seq_io.h:44,
                 from lib/read_parsers.hh:41,
                 from lib/hashtable.hh:57,
                 from lib/kmer_filters.cc:40:
/usr/include/c++/6.3.1/bits/unique_ptr.h:49:28: note: declared here
   template<typename> class auto_ptr;
                            ^~~~~~~~

[standage]

is the duplication related to the multiple template instantiations we do for the read parser?

SeqAn is heavily templated as well, so that may also be a factor. There are some warnings I see duplicated several times during compilation, but I haven't kept track of which they were and haven't seen a substantial increase since introducing templating to the read parser.

[standage]

👍

[betatim]

Ok. For sure the original sin comes from seqan being a bit old, hence the warnings. Downside to all the "visual noise" is that I train myself to ignore it all which means sometimes you miss a new warning :-/

[standage]

Downside to all the "visual noise" is that I train myself to ignore it all which means sometimes you miss a new warning :-/

Yep, the struggle is real

[betatim]

@standage this can be merged right?

[standage]

"Yes", given the caveats in our current #khmer@slack discussion.

[betatim]

Had forgotten about that, let's wait for a decision re: merging into feature/cython_all_the_things

[betatim]

I am impressed merging master just worked. 😀

[betatim]

From scrolling through the travis output there are less warnings, I think. Merge first, squash more warnings later?

[standage]

I did have to resolve a couple conflicts, but honestly there were few enough changes that I could easily have redone the changes manually.

On Thu, Jul 27, 2017 at 7:56 AM Tim Head ***@***.***> wrote: I am impressed merging master just worked. 😀 — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1605 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAimJxylFzhBWr_YZconPjF2alvWHS7Jks5sSKUzgaJpZM4L0EAg> .

--

…

-- Daniel S. Standage, Ph.D. Postdoctoral Scholar Lab for Data Intensive Biology University of California, Davis

[standage]

Most of the remaining warnings are related to the third-party code. But there are a couple from khmer/oxli. This isn't an urgent PR. Let's see if we can squash the rest of the warnings (from our code) before merging.

On Thu, Jul 27, 2017 at 7:58 AM Tim Head ***@***.***> wrote: From scrolling through the travis output there are less warnings, I think. Merge first, squash more warnings later? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1605 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAimJ16jbuRQPvgWDuFuBsx_KrNcxTZNks5sSKWWgaJpZM4L0EAg> .

--

…

-- Daniel S. Standage, Ph.D. Postdoctoral Scholar Lab for Data Intensive Biology University of California, Davis