fixed check_space_for_hashtable to use args.n_tables

[ctb]

Fixes bug where table size was calculated based on args.ksize instead of args.n_tables. Important bug to fix for some cloud computing systems w/limited disk space.

[ctb]

• Is it mergable
• Did it pass the tests?
• If it introduces new functionality in scripts/ is it tested?
  Check for code coverage.
• Is it well formatted? Look at pep8/pylint, cppcheck, and
  make doc output. Use autopep8 and astyle -A10 if needed.
• Is it documented in the Changelog?

[ctb]

Fixes #380.

[ctb]

Ready for review, if tests pass.

[ged-jenkins]

Test FAILed.
Refer to this link for build results: http://ci.ged.msu.edu/job/khmer-multi-pullrequest/477/

[ctb]

test this please.

[ged-jenkins]

Test PASSed.
Refer to this link for build results: http://ci.ged.msu.edu/job/khmer-multi-pullrequest/479/

[ctb]

ready for review.

[mr-c]

Should we define these arguments to be type float to avoid the need for this (and future potential bugs)?

[ctb]

On Sun, Apr 13, 2014 at 11:58:32AM -0700, Michael R. Crusoe wrote:

@@ -53,7 +53,7 @@ def main():
check_file_status(_)

 check_space(args.input_filenames)

• check_space_for_hashtable(args.ksize * args.min_tablesize)
• check_space_for_hashtable(float(args.n_tables * args.min_tablesize) / 8.)

Should we define these arguments to be type float to avoid the need for this (and future potential bugs)?

I don't think it's necessary, I guess. I could go either way. Any points
of evidence that this is needed? I can't think of anywhere else this is
done.

--t

C. Titus Brown, ctb@msu.edu

[ctb]

I've added an issue re improving the test coverage.