Added .mol functionality

[nwoodweb]

The file iodata/formats/mol.py implements V2000 *.mol compatibility, based heavily off of currently implemented *.sdf compatibility

• I had used load_one on a phenol.mol file I had made
• I had passed Roberto CI check

[codecov[bot]]

Codecov Report

Merging #290 (7f61aec) into master (73eee89) will increase coverage by 0.03%. The diff coverage is 96.75%.

@@            Coverage Diff             @@
##           master     #290      +/-   ##
==========================================
+ Coverage   95.00%   95.04%   +0.03%     
==========================================
  Files          74       76       +2     
  Lines        8790     8975     +185     
  Branches     1213     1227      +14     
==========================================
+ Hits         8351     8530     +179     
- Misses        191      194       +3     
- Partials      248      251       +3     

Impacted Files	Coverage Δ
iodata/formats/mol.py	91.42% <91.42%> (ø)
iodata/test/test_mol.py	100.00% <100.00%> (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 73eee89...7f61aec. Read the comment docs.

[FarnazH]

@nwoodweb thanks for your contributions. As shared in issue #269, having a test for every line of code is part of our contribution guidelines, so we cannot review your PR until suitable tests are added.

[nwoodweb]

@FarnazH Ok thankyou for clarifying. I will add a testing script in iodata/test as well as at least 2 example files and a V2000 unit test in iodata/test/data

[nwoodweb]

I accidentally pushed to master, which is part of the pull request, then pushed my unit tests to new-feature so I had simply merged new-feature with my forks master

I added unit test file iodata/tests/test_mol.py as well as various example mol files

[tovrstra]

Thanks for the contribution and my apologies for the delay. The code is almost identical to that for the SDF format. When looking for the differences, I could only find one thing related to the end marker for a single molecule:

• For SDF this is

  M  END
  $$$$
  

• For SDF this is

  M  END
  

Are there any other differences? If not, I'd suggest extending the SDF code to also handle the .mol case correctly. This way, we can avoid duplicated code.

[nwoodweb]

Hi @tovrstra

.sdf and .mol are for the most part identical besides the termination string at the end. In hindsight it might be better to use branch statements to differentiate M END and M END $$$$.

However, I do remember being confused about V2000 and V3000 formatting, as well as the bond block, so I will review those before implementing your recommendation.