Issue 42 fixes

[AaronKutch]

While I figure out what to do about the BitWidth situation, I decided to go ahead an fix most of #42.

The only question I have now is if we should replace the single bit set/unset/flip functions with a single function set_bit that takes a boolean as an argument.

[AaronKutch]

Also, should I split up the first commit for you, because I originally wanted it to be two separate commits but forgot to commit in the middle, and it is tedious to separate the two?

[coveralls]

Coverage increased (+1.07%) to 81.913% when pulling 29c7ee448ea18f5ef9a4d1ced2024c1cc5bd941c on AaronKutch:issue-42 into 130949d37601a809d8d463430a94cbb17251a7a6 on Robbepop:master.

[codecov-io]

Codecov Report

Merging #56 into master will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##           master      #56   +/-   ##
=======================================
  Coverage   80.47%   80.47%           
=======================================
  Files          23       23           
  Lines        5229     5229           
=======================================
  Hits         4208     4208           
  Misses       1021     1021

---

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 130949d...fd83c51. Read the comment docs.

[Robbepop]

While I figure out what to do about the BitWidth situation, I decided to go ahead an fix most of #42.

The only question I have now is if we should replace the single bit set/unset/flip functions with a single function set_bit that takes a boolean as an argument.

The set_bit function makes sense to replace set and unset, however, maybe for flip we want to keep it since it does two things (lookup and modification at bit level).

[AaronKutch]

I have changed my mind and want to keep them for symmetry purposes. If I really want to bikeshed, then I would implement bit indexing for ApInts, but that leads right into bitslices and other stuff that will probably come after 0.3.0. I think this is ready.

[AaronKutch]

The CI needs cargo-fmt installed. Do you need to do rustup stuff on caches your end? I actually do not know much about CI in particular.

Also, I noticed that cargo test (no --release) is being used in travis. One cargo test is needed to test for stray overflows, but is a waste of time to run on more than one platform because of how slow the fuzz tester is on debug mode.

[Robbepop]

The CI needs cargo-fmt installed. Do you need to do rustup stuff on caches your end? I actually do not know much about CI in particular.

Also, I noticed that cargo test (no --release) is being used in travis. One cargo test is needed to test for stray overflows, but is a waste of time to run on more than one platform because of how slow the fuzz tester is on debug mode.

Fuzz tests should always be put behind a feature flag so that they only run when the feature flag is enabled since cargo test should by default only ever run fast tests for a nice debug-exec cycle.

[AaronKutch]

I wanted to finish my work on specialized-div-rem before working on apint again, but things are taking longer than expected. I decided to look at this PR again. I think this PR is still good, the only change I just made is to remove the commits about CI, which you are planning on replacing with github actions.

[AaronKutch]

I have added the one and is_one methods back to Int and UInt

[codecov-commenter]

Codecov Report

Merging #56 into master will increase coverage by 1.40%. The diff coverage is 80.92%.

@@            Coverage Diff             @@
##           master      #56      +/-   ##
==========================================
+ Coverage   80.47%   81.87%   +1.40%     
==========================================
  Files          23       23              
  Lines        5229     5132      -97     
==========================================
- Hits         4208     4202       -6     
+ Misses       1021      930      -91     

Impacted Files	Coverage Δ
src/apint/relational.rs	27.10% <0.00%> (+1.88%)	:arrow_up:
src/apint/serialization.rs	92.42% <ø> (+0.11%)	:arrow_up:
src/lib.rs	100.00% <ø> (ø)
src/int.rs	18.98% <46.87%> (+9.15%)	:arrow_up:
src/apint/arithmetic.rs	95.77% <80.00%> (+2.34%)	:arrow_up:
src/apint/bitwise.rs	89.23% <85.00%> (-1.50%)	:arrow_down:
src/apint/utils.rs	79.42% <94.73%> (+0.06%)	:arrow_up:
src/uint.rs	32.35% <95.12%> (+21.32%)	:arrow_up:
src/apint/casting.rs	72.84% <100.00%> (-3.56%)	:arrow_down:
src/apint/constructors.rs	99.75% <100.00%> (+2.24%)	:arrow_up:
... and 23 more

---

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 130949d...29c7ee4. Read the comment docs.

[Robbepop]

Thank you for your work @AaronKutch !