Fix bug in zero relative series to power zero.

[wbhart]

Fixes #931

[codecov]

Codecov Report

Merging #938 (4576ccf) into master (9f54b08) will decrease coverage by 0.12%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #938      +/-   ##
==========================================
- Coverage   82.53%   82.41%   -0.13%     
==========================================
  Files          86       86              
  Lines       20065    20058       -7     
==========================================
- Hits        16560    16530      -30     
- Misses       3505     3528      +23     

Impacted Files	Coverage Δ
src/RelSeries.jl	94.89% <100.00%> (+0.58%)	⬆️
src/generic/SparsePoly.jl	25.72% <0.00%> (-1.36%)	⬇️
src/generic/LaurentSeries.jl	92.88% <0.00%> (-0.62%)	⬇️
src/generic/RelSeries.jl	96.95% <0.00%> (-0.51%)	⬇️
src/Poly.jl	91.13% <0.00%> (-0.36%)	⬇️
src/generic/RationalFunctionField.jl	69.19% <0.00%> (-0.13%)	⬇️
src/generic/PuiseuxSeries.jl	84.24% <0.00%> (-0.05%)	⬇️
src/generic/FunctionField.jl	74.19% <0.00%> (-0.04%)	⬇️
src/generic/AbsMSeries.jl	94.01% <0.00%> (-0.03%)	⬇️

---

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 9f54b08...4576ccf. Read the comment docs.

[thofma]

Failure on nightly looks scary, but not our fault I think.

[thofma]

Ah, should we the example from the issue as a test case?

[wbhart]

Yes, I can add the test case.

The failure on nightly is obviously not related. It has to do with our use of set_field in the map code. Presumably they don't actually want 1.7 to be backwards compatible with 1.0 after all.

[thofma]

Thanks.

PS.: I opened an issue on the julia repostitory.

[wbhart]

Thanks, you beat me to it!